Music vs Sine Waves

MEASUREMENTS & AUDIOPHILES: One of the goals behind this blog is to explore some of the more popular audiophile beliefs. Which ones are true, partly true, or completely false? When it comes to measuring audio gear there are many different beliefs but I often run into variations of these three (photo: Leon Wilson):

• Measurements use test signals, not music, so they’re of limited use
• Measurements fail to account for real world usage and loads
• Measurements matter little as you can only trust your ears.

MUSIC vs TEST SIGNALS: Intuitively music is far more complex than test signals. So it’s not surprising many believe such measurements cannot accurately convey the performance of audio gear. But there’s lots of well documented research demonstrating the right measurements using test signals can help predict the sound quality of a lot of audio gear. Some things to consider (photo: Inha Leex Hale):

• Sine Waves - Sine waves are not some abstract signal created in a lab. They’re the primary building block of all sounds we hear. Analogies would be a single color of light or a pure chemical element from the periodic table. All the colors we perceive are combinations of individual wavelengths of light. And everything we experience in the physical world is made up of elements from the periodic table. And, in much the same way, music is just a collection of sine waves. A perfect sine wave is a single pure tone and has no distortion of its own. It's the most pure component of sound.
• Steinways & Yamahas - The note "A" above "Middle C" on a piano strongly resembles a 440 hz sine wave. It's a relatively pure tone at a single frequency. The wood structure, nature of the strings, hammers, etc. all slightly alter that 440 hz sine wave. A Stieinway grand might have a slightly faster attack, a longer decay and a different set of distortion products than a Yamaha grand. These subtle properties are well enough understood it’s possible to simulate the sound of different pianos using software (photo: Mrs Logic).
• Amplifier Distortion – Just as the sound of a particular grand piano can be simulated by understanding its distortion, the same can be done with amplifiers. In the mid 80’s Bob Carver challenged high-end audio magazines and ended up duplicating the sound of a seriously expensive Conrad Johnson tube amp using one of his inexpensive mainstream solid state amps. He did this by simply measuring the tube amp using test signals. The golden-eared editors had a very difficult time telling which amplifier was which despite the massive price difference. There are many other examples that demonstrate the power of proper measurements. If Bob Carver can “describe” and essentially equal the sound of a high-end audiophile amp using measurements that says a lot.
• YACA (Yet Another Car Analogy) – Much as music is more complex than test signals, public roads are more complex than a closed circuit on a racetrack. But when someone wants to figure out if Car A outperforms Car B they take them to a track where they can be evaluated under identical controlled conditions. There are too many uncontrolled variables in real world driving such as other traffic. Few dispute a racetrack is the best overall way to evaluate acceleration, braking and handling limits of cars. A test bench for audio gear offers the same for audio gear. A fair and valid performance comparison is only possible under rigidly controlled conditions—not casual listening or driving.
• Correlation – Lots of studies compare measurements made with sine waves to perceived distortion when listening to music. And, for decades, the research has supported a strong correlation between the measurements and what we hear. It’s not black and white when you’re dealing with human perceptions, but virtually all of the research has pointed towards various measurement thresholds that help define what people can perceive under various conditions.
• Sufficiently Transparent – When there are audible differences between audio gear it’s sometimes difficult to definitively pick a clear winner as that’s subjective and personal preferences will bias the result. This is especially true with things like headphones, speakers, and phono cartridges. But with electronics like amplifiers, DACs, pre-amps, etc. measurements can help a lot. It’s been shown once all the right measurement thresholds are met, the equipment in question becomes essentially “transparent” in the signal chain—i.e. it doesn’t alter the audio signal in an audible way. For example, Meyer & Moran demonstrated you can insert an A/D and D/A operating at 16 bits and 44 Khz into a high resolution SACD signal path and even skilled listeners could not tell when the extra hardware was present. The A/D and D/A were sufficiently transparent and did not alter the sound enough for anyone to detect.
• Myth Busted – Sine waves are not some artificial signal with no basis in reality. They’re  a well proven method to reveal distortion in audio gear and they’re a building block of all sound we hear including music. And audio engineers have more tricks up their sleeves than just sine waves.

REAL WORLD USAGE & LOADS: Some claim measurements fail to account for real world conditions. But that’s all in how the tests are done. Some tests account for a much wider variety of conditions than typical listening tests:

• Test Loads – Proper tests are done using proper loads. And it’s not uncommon to run at least some tests with reactive loads or even real loads. It’s not difficult to model a loudspeaker or headphone driver on a test bench. And it’s easy to compare the results with various kinds of loads including real ones. Hence there’s a pretty good understanding and body of evidence as to how simulated loads affect the performance of audio sources.
• Complex Test Signals – Not all tests are just a simple sine wave. Some involve sweeps or other techniques covering the entire audio spectrum, using multiple tones, impulse signals, chirps, pink noise, square waves, etc. And measuring noise doesn’t require any test signal.
• Worst Case Testing – It’s not that difficult to come up with worst case operating conditions that represent real world usage. I did just that in developing the criteria for the O2 headphone amp. Once you establish the worst case criteria, tests can be run to verify the performance under those conditions. If the gear measures well enough to be transparent even under worst case test conditions, it’s a safe bet it will also be transparent under realistic conditions in the real world.
• Audio Differencing – It’s possible to test many kinds of audio electronics using a method known as analog or digital audio differencing. These tests can be done using real music and real loads (i.e. headphones). In essence, the input and output are matched in level and subtracted from each other. This method was originally put forward by Baxandall and Hafler as a method for evaluating power amplifiers under real world conditions. Differencing captures nearly all forms of distortion and can be quantified objectively by measuring its level and spectral properties. It can also be quantified subjectively by listening to the nature of the difference signal and how unpleasant it is.
• Myth Busted – If anything it’s more common to fully challenge audio gear on a test bench than in listening tests. Just like testing a car on a track, it’s generally best to find the ultimate limits on a test bench rather than in real world usage.

TRUSTING YOUR EARS: Audiophiles put a lot of trust in their ears and subjective impressions certainly matter. But there's a problem. Human senses, including hearing, are greatly influenced by other factors (photo: Travis Isaacs).

• Primal Brain - A primal and involuntary part of our brain constantly filters our senses to avoid sensory overload. For example, your brain automatically filters out other conversations at a noisy party so you can better hear the person you’re talking to. Check out how the brain involuntarily filters what you hear with this brief BBC video demonstrating the fascinating McGurk Effect
  • BBC McGurk Effect on YouTube
• Seeing & Hearing – In the video above, if you close your eyes your hearing is accurate but with your eyes open, your brain deceives you. It turns out when listening to audio gear you have go one step further and remove the knowledge of what gear you’re listening to. Otherwise much the same thing happens—the brain tries to help out and serves up an altered version of what you’re listening to. The objective geeks call this “sighted listening bias” and it’s been well proven in many studies.
• Bed Sheets & Hearing – The simple act of throwing a bed sheet over an equipment rack can make allegedly obvious differences in sound quality disappear. The listener’s abilities, room, music, and the hardware remain unchanged, yet just removing the knowledge of which equipment is playing removes the previously audible differences. This has been proven again and again, even in the homes of audiophiles, by listening tests such as this one: Matrix Audio Test
• Conditions Are Different – Just because a headphone amp sounds good with 300 ohm Sennheisers doesn’t mean it will sound equally as good with 25 ohm Denon headphones. And gear that might sound great with classical music may fall on its face with the next guy’s hip hop. So listening tests are often only valid for a particular set of audio preferences, conditions, type of music, volume, etc. And all those things differ widely from one person to the next. I’m not saying these tests are useless, but they’re highly subjective and very difficult to compare between people and conditions.
• Ears Are Different – If Michael Fremer at Stereophile listens to a piece of gear and declares it best-in-class what does that really mean? It’s much like a wine critic saying the same thing about a particular wine. But the next wine critic will often choose a different wine as he had different tastes. The same is true in audio and it’s a fundamental problem with subjective listening. Everyone’s tastes, preferences, priorities, hearing acuity, listening skills, etc. are different. One man’s “detailed” is another man’s “excessively bright”. So it’s difficult to trust someone else’s ears. And, there are not many stores where you can walk in and audition high-end headphones to hear them with your own ears before purchasing. Measurements greatly supplement these subjective reviews and provide a much better means of comparison.
• No Contest – With other controversial topics, say global warming, there’s typically conflicting research. But, in this case, there’s lots of research supporting the problems with sighted listening and essentially nothing credible opposing any of it. If there’s some problem with blind listening tests, why hasn’t the well funded high-end audio industry managed a single study supporting the supposed advantages of their products and/or sighted listening?
• Myth Busted – The points above explain why it’s difficult to trust your own ears and trusting someone else’s ears is much like trusting a wine critic. Hence the classic “you can only trust your ears” belief is highly suspect. Objective measurements, however, generally can be trusted as they’re immune to all the issues above.

OTHER RESOURCES: If you’re still not convinced the three popular beliefs are more myth than reality, or if you want more information, the following links are worth a look:

• Audio Myths Workshop Video – This is a fascinating video covering everything from human psychology to assorted listening trials. One of the presenters is Ethan Winer who has also made available audio files to allow your own comparisons.
• Science and Subjectivism in Audio – Douglas Self shares his views on this debate in an older but still poignant article. He’s the engineer behind some high-end gear such as the current flagship Cambridge Audio products. He’s published some of the most highly regarded books on audio design in the world.
• Subjective vs Objective Debate – This is my own article covering more of the philosophical differences between the “trust your ears” and “measurements are best” camps. The comments are also enlightening.
• Testing Methods – I share some of my thoughts on how I test audio gear and why. Tests should be done in standard ways to allow fair comparisons between gear and they should be verifiable by others.

BOTTOM LINE: If you want to know the ultimate performance limits of a car you take it to a test track or race course. And if you want to know the ultimate performance of audio gear, you put it on a test bench and use an audio analyzer to make appropriate measurements. Subjective impressions are still important in both cases. For example, the numbers tell you nothing about how easy the controls are to use. But when it comes to determining if the BMW or Mercedes is the higher performance car, measurements offer the best answer. The same is true when comparing audio gear.

Op Amp Measurements

INTRO: In last week’s Op Amps: Myths & Facts article I covered the background for op amps, why they’re important, and some common audiophile op amp myths. This is the follow up with more technical information, a summary of many popular op amps, and some of the measurements I made.

THIS STUFF MATTERS: A lot of people play around with “op amp rolling” but they mostly lack the ability to measure the results. A lot of op amps get “upgraded” to much more expensive parts usually with zero real world benefit. Worse, these usually faster parts can easily be unstable in a circuit designed for a much slower part. Some of the differences heard could be high frequency instability. But when someone “upgrades” to an expensive op amp with high expectations many assume the different sound must be better sound. This article is all about measuring the results and putting some facts around the differences between op amps in a typical audio application.

REALITY CHECK: This isn’t a comprehensive comparison. It’s mostly a side story to my research and development of the O2 Headphone Amp. So please don’t expect a technical paper worthy of the Audio Engineering Society. A complete op amp survey is at least several full weeks worth of work. For an example of a more ambitious effort, look up Samuel Groner’s “Operational Amplifier Distortion” published in 2009. It’s a 436 page monumental effort in the form of a huge 35 MB PDF (which is why I’m not linking to it here). I’ve also referenced the work of Douglas Self who has run many op amps through an Audio Precision SYS-2700 series audio analyzer in various configurations.

LIMITED SCOPE: The O2 headphone amp uses DIY-friendly DIP8 dual op amps. It also requires op amps that can operate comfortably at +/- 12 volts (24 volts total). This rules out single devices, surface mount packages, and lower voltage op amps. While I did adjust the compensation and feedback impedance with some op amps, ultimately, the measurements are all in the same circuit configuration unless otherwise noted.

MOUSER’s SELECTION: One of the design goals of the O2 was to purchase as many parts as possible from a single vendor and, for some good reasons, that turned out to be Mouser Electronics. Mouser offers Ti, Burr Brown, JRC/NJR, On Semi and others, but they don’t offer National, Analog Devices or Linear Technology. So most of my evaluation was limited to what Mouser stocks. But I did test a couple parts from National and Analog Devices.

BLIND TESTING CHALLENGE: For those of you who believe there’s more to op amp sound than just the measurements, please read the Op Amp Myths article if you haven’t already. And if you’re still a skeptic, let’s schedule your YouTube video appearance with an ABX comparator, the op amp of your choice, and a $0.60 NE5532 using two O2 boards and the op amps in the gain stage. If you can tell them apart, I’ll give $500 to the charity of your choice. If you can’t tell them apart, you donate $500 to the charity of my choice. The challenging op amp’s measurements must at least meet the minimum criteria outlined in the O2 Design Principals. The test would be administered by an independent third party (I won’t even be present). Win or lose the video goes on YouTube and the story is published on this blog. And if you think the O2 is the weak link, please see: O2 Open Challenge

BOTTOM LINE: For those wanting to skip the Tech Section, the conclusions can be summed up as follows:

• At gains less than 4X nothing overall could beat the $0.39 NJM2068 in the O2’s gain stage. This is especially true if you’re concerned about power consumption for battery operation.
• At gains of 5X and higher nothing could beat the $0.65 NE5532 except by a few dB in noise performance.
• The older TL072, despite being used in thousands of audio devices, is way out of its league against newer op amps. It’s no longer even in the game in a low noise application like the O2.
• For the output stage, the NJM4556 was untouchable among compatible DIP8 dual op amps.

 

---

TECH SECTION

---

CONSTRAINTS: The testing was mostly limited to op amps suitable for the gain stage of the O2. See the O2 Design Process article for more.

APPLICATIONS: Most of this article is written from the perspective of using op amps for headphone and line level audio applications. I’m excluding high gain applications like microphone preamps, phono preamps, instrumentation equipment, etc. Used in high gain critical applications the rules can be different.

OP AMP SPECS: When you start “screening” op amps for a project you have to look at packages (DIP, SOIC, etc.) and decide if you want a single, dual, or quad part. Beyond that it mostly comes down to the specs and here are the highlights:

• Intended Application – The semiconductor companies actually do know what they’re doing. They know what parameters are most important for audio applications and most design op amps with audio specifically in mind. It’s best to use those parts designed for audio, not those optimized for say video use. This seems obvious enough but several audio designers break this rule like the OPA690 in the AMB Mini3 which isn’t even specified at audio frequencies and has rather horrible specs for audio. Anyone who thinks they know more about op amp applications than the guys designing the op amps is very likely mistaken. It’s like putting tractor tires on your sports car thinking they’ll improve the traction. Stick to the tires made for sports cars.
• Operating Voltage – The O2 op amps need at least a 24 volt (+/- 12) power supply. This rules out a lot of the newest parts that are designed for low voltage portable use. Also note some op amps are optimized for single supply use and, while they’ll usually work with dual supplies, they may not be an optimal choice.
• Unity Gain Stable – There are some “un-compensated” op amps available that are not unity gain stable. They specify the minimum gain they can be used at. These op amps are generally best left to those who understand the complex math involved with calculating things like phase margin and who have the test equipment to verify the result (RMAA and low bandwidth scopes don’t qualify). Otherwise, stick with Unity Gain Stable compensated op amps.
• Package – The O2 requires a dual op amp in a DIP8 package. Op amps also come in single and quad versions, various surface mount packages, round metal TO99 “cans”, and some high power packages for higher thermal dissipation.
• Single or Dual – The crosstalk between sections of dual op amps designed for audio use is typically better than –105 dB. That’s vastly better than the rest of a headphone amp can manage so it’s not an issue. There’s no need to use a single op amp for each channel. The 3 contact plug on almost every pair of headphones creates more crosstalk all by itself. Some claim thermal coupling between channels is a problem in dual parts, but the power consumption of a gain stage is usually dominated by the quiescent current which is relatively constant. So there won’t be significant thermal variations on the die due to the music signal. A dual op amp also has the advantage of providing better matching between the channels. So for cost savings, and to save board space, the choice is obvious—use a dual op amp.
• FET or Bipolar – You can broadly divide audio op amps into these two categories based on the type of transistors used in their input stage—FET or bipolar. Some claim various types of FET (JFET, Bi-FET, Di-FET, etc.) inputs “sound better” but they’re usually more noisy in real world use. The main reason to use FET inputs is for their superior DC performance (especially bias current) which is mostly useless in an audio application. The one exception being, if you’re going to hang a volume control right off the input, a FET op amp will generally put less DC through the volume control wiper. This helps reduce the “rustling” noise you get when you adjust the volume. It also can allow larger resistor values without excessive DC offset. But the O2 solves these issues in other ways so FETs are generally more of a liability if accuracy is the goal. The ultra high-end FET devices, like the Di-Fet OPA627, are an exception to the rule but the FET OPA2134 is not. It performs worse in some ways than the much cheaper bipolar 5532.
• Quiescent Current – In a battery powered device the quiescent (idle) current is a critical spec. Most of this current is for biasing the internal transistors. Some op amps need as much as 9 ma per amp section or more. While low power op amps typically have poor drive capability, higher noise, lower GBW, and poor slew rate. There are significant trade-offs between power and performance.
• Gain BandWidth Product (GBW/GBWP/GBP) – The GBW helps indicate how much feedback will be available at higher frequencies and higher gains to correct distortion. But higher GBW usually means more power drain and possibly less stability. For low distortion at 20 Khz it should be at least 5 Mhz for G=1. Some manufactures specify the Unity Gain frequency instead which is the same as GBW with G=1. To compare GBW between different op amps it needs to be specified at the same gain—i.e. G=1, G=10, etc. Most manufactures provide a graph as well.
• Open Loop Voltage Gain (myth busted) – The open loop gain combines with the GBW above to determine how much feedback (NFB) is available across the power band. Generally, the higher the better and it should be at least 90 dB worst case. The best audio op amps are around 140 dB typical and it can make a big difference in the performance—especially at higher gains. Nearly all the audiophile myths about too much feedback have been proven false by lots of research—some of it relatively recent. Bruno Putzeys (a truly brilliant guy), in particular, has put most of the remaining persistent myths to their overdue death. May they rest in peace (but I’m sure they won’t). More on this below.
• Slew Rate – Here’s another huge myth: The faster the slew rate the better. But the fastest “slew” you can get from a 16/44 digital recording has a period of about 22 uS. So the O2, worst case, needs to slew 20 volts in 22 uS or a slew rate of 0.9 V/uS. And even considering those rare 24/96 recordings, the requirement is still 1.8 V/uS. SACD or 24/192, in theory, could require about 3.6 V/uS but, in practice, never will as that would be a 0 dBFS signal close to 100 Khz. Such a signal would never make it through the recording chain, and if it did, would fry tweeters and cause bats to fly into the side of your house. The peak levels in music generally fall as the frequency rises. This is especially true at ultrasonic frequencies. If you think I’m wrong, please reference a recording than can challenge even a 1.5 V/uS headphone amp. The industry rule of thumb, even on the recording side of the signal chain, is 0.2 V/uS per volt RMS of maximum output. I’ve doubled that just to be ultra conservative so 7*0.4 = 2.8 V/uS for the O2’s requirement. As I have explained elsewhere,slew rates well in excess of what’s needed come at a price. Somewhere in the range of 2.5 – 5 V/uS is ideal for a high output headphone amp like the O2. 
• Feedback & Speed Myths – First of all, some audiophiles are under the mistaken impression that negative feedback (NFB) is mostly a bad thing. I just read, in an audiophile magazine, that NFB can only work correctly if an audio amp has zero delay. And they claimed such amplifiers don’t exist so the next best thing is either zero feedback or an extremely fast amplifier. The amp being discussed was touted as good for “250 Mhz” (yeah right!) resulting in “essentially zero delay”. This is a huge pile of false technobabble. The whole idea of “by the time a feedback loop corrects an error, it’s too late” is just layperson nonsense. Look up what Bruno Putzeys and Doug Self have written on the topic if you want the real truth. To use a layperson’s analogy, nearly any DC motor driving a turntable is controlled with a feedback loop. The turntable has rather massive delays with respect to current inputs to the motor vs speed due to the mass of the platter. The feedback pundits argue by the time the feedback detects an error, the speed is already wrong and can’t be corrected due to the delays. But, somehow against all those supposed audiophile gurus, the speed is rock steady and the feedback works perfectly. The same is true in an amplifier. The only speed issue that matters is the amp should never be slew rate limited. As long as that condition is satisfied, there’s no such thing as “Transient Intermodulation Distortion” or other such feedback related problems. Those are all dinosaurs from the era before fast power transistors were available for high power amps. It’s a non-issue at headphone voltages and below even using $0.39 parts like the NJM2068. Really.
• CMRR – Common Mode Rejection Ratio is the ability of an op amp to reject input signals that are in-phase with respect to ground. This matters most when using an op amp in differential mode but it can help lower distortion in many single-ended circuits. For example, with the popular voltage follower op amp “buffer” shown to the right, the distortion will rise with input impedance in proportion to the op amp’s CMRR. In other words, if you stick a volume control in front of a buffer, and turn down the volume, an op amp with better CMRR may well outperform one with lesser CMRR. The close quarters inside a small portable amp results in more electromagnetic coupling of high current signals throughout the circuitry. Here again, higher CMRR helps the critical gain stage reject what otherwise might become distortion. For this application at least 80 dB worst case and 100+ dB typical is desirable. It’s also worth noting CMRR is often asymmetrical which is only sometimes disclosed on datasheets
• PSRR – Power Supply Rejection Ratio is the ability of an op amp to reject common mode signals on the power supply rails such as ripple and distorted versions of the music signal. The grain of truth behind the common audiophile desire for massive overkill power supplies is they can make a difference but only when the audio circuit has insufficient PSRR. And a lot of audiophile designs have poor PSRR as they use discrete circuits or op amps not designed for audio. The OPA690 in the Mini3, for example, has poor PSRR for audio use. But if you simply use op amps with excellent PSRR, any variations on the power supply end up well below the noise floor of the entire design even with an inexpensive power supply. It’s a much more elegant and rational solution than a power supply with 50 times the required current capability. Like CMRR, PSRR can by asymmetrical. 
• Turn On/Off Characteristics – Unless there’s an output relay to protect the headphones it’s important the op amp be reasonably well behaved on power up and power down. Not all are. And it’s often not covered in the datasheets so you just have to test it with a real scope at a slow sweep speed (most AC coupled soundcard scopes and RMAA won’t work). Don’t try to subjectively judge the “thump” in a pair of headphones as it’s hard to hear large amounts of DC. And a DMM is nearly useless due to their very slow update rate (they’ll miss the true peak of the transient by a mile).
• Drive Capability – To keep Johnson noise low, the op amp should have low distortion into a 1K load. A lot of op amps fail here—especially low power versions. This can be hard to figure out from many datasheets. Some of the better audio op amps are specified into 600 ohms. If you only see loads of 5K or higher, the part probably isn’t suitable for high-end audio. Ideally look for a THD graph at 2K or less with some gain.
• Low THD – This is also often poorly specified. And if it’s not specified at all, or given as x dB for specific harmonics, you’re probably looking at the wrong op amp for an audio application. Ideally you want to know the THD at various gains, versus frequency, into various loads. In practice, this often requires testing with an audio analyzer. If the entire amp is to meet a 0.01% THD goal, the gain stage needs to be well under that.
• Low Noise – Op amp noise is modeled two ways--current noise and voltage noise. The O2 is a low impedance application and the current noise will likely dominate but both have to be considered along with the Johnson noise. The math isn’t trivial. You can’t just compare the raw datasheet numbers and know how the amp will really perform in a specific application unless both the voltage noise and current noise are lower than another op amp. And even then Johnson noise could still dominate so both would perform roughly the same. And, in reality, the voltage and current specs tend to be mutually exclusive—i.e. amps with the lowest voltage noise usually have higher current noise and vice versa.
• Phase Reversal – Some op amps, when their inputs are overloaded, violently slam into the opposite supply rail. And sometimes that nasty fact isn’t even on the datasheet or is buried in a foot note somewhere. Phase reversal is unacceptable in this application.
• Rail-to-Rail – This isn’t required for the O2, but would be nice—especially when running from batteries. The problem is most rail-to-rail audio op amps are newer designs and often only come in surface mount packages and/or can’t handle a 24 volt supply. And the few that exist and meet the other requirements tend to be expensive. Even non-rail-to-rail op amps vary widely in how close they can get to the rails with various loads and operating voltages. The datasheets usually have a graph or two showing voltage swing versus load resistance and/or current output. It’s rarely worth using an op amp that’s otherwise poorly suited to high quality audio just because it’s rail-to-rail.
• DC Offset Voltage & Bias Current – This usually isn’t an issue in an AC coupled application like the O2 but can very be a serious issue in a DC coupled application not using a servo or a high-pass filter in the feedback loop. The higher the gain the more it can matter.
• Short Circuit Protection – Any op amp driving an external output (line outs, headphone jacks, etc.) should have short circuit protection unless you can add enough series resistance without compromising the performance (which is often impossible). Beware even with short circuit protection it might be short term if the device does not also have thermal protection. The datasheets are usually fairly clear on this. There’s also the chance of exceeding the Safe Operating Area (SOA) of the output stage at higher supply voltages. Some designs may require some series output resistance to meet all these requirements. There can be trade offs involved. Enclosing short circuit protection in the feedback loop, as done in the AMB Mini3, can create stability issues and increase distortion.
• Equivalent Schematics – Some manufactures publish so called “equivalent schematics” for their op amps. The implication is it’s a fairly accurate representation of how they actually made the part. In reality, however, it’s often not as they don’t want to help their competitors out too much. The schematics have become rather generic so as not to give too much away. So don’t try to take them literally. There’s often a lot more going on.
• Unique Advantages – For those who prefer discrete designs over op amps, I refer you to Q9 in the small snippet to the above right. Op amp designers are free to optimize every single transistor in the device including creating things you can’t buy off the shelf—like  transistors with 3 collectors. This, combined with the inherent matching of transistors made on a single die, and near perfect thermal tracking, provides a bag of tricks no discrete designer can match. Sure there are advantages both ways, but the op amp guys have an awful lot at their disposal—particularly compared to some guy working in his basement.

COMPENSATION: Unity gain stable op amps, as being evaluated here, use dominant pole compensation for stability. This is to make sure that the amp does not oscillate when feedback is applied. In a low gain application, the amp might still show ringing on square waves due to the phase margin becoming small. This is usually compensated for by adding the correct value of capacitance in parallel with the feedback resistors from output to the negative input. These capacitors cause a zero and a higher frequency pole in the feedback loop response effectively cancelling one of the amp's own non-dominant high frequency poles (by placing a zero near it) and replacing it with a new one at a higher frequency. This results in an increased phase margin, improved stability, and improved transient response. It’s important to note this compensation isn’t “one size fits all”. It’s specific to many variables including the gain, the op amp’s own loop response, the impedances and reactive components in the circuit, etc. Like many things in audio it can be a delicate compromise say in an output stage with unknown external loads, or in a gain stage with two different gain options.

THE OP AMPS: Below I’ve sorted some of the more popular op amps into several categories. The ones marked with an asterisk are op amps I’ve tested in the last few months and at least some data is shown for most of them in the next section.

THE ELITE FEW: Designer op amps are amazingly like hyper-expensive supercars. They’re vastly more expensive than their more mainstream counterparts, relatively high strung, and very high maintenance. While they may show their true colors in critical applications, like say inside a $20,000 audio analyzer, it’s hard to imagine where they offer any real-world benefit inside a headphone amp. It’s analogous to taking your half a million dollar Ferrari down to the Quick-E-Mart for a corndog. The only reason not to take a perfectly competent, and more comfortable, BMW instead is so you can be seen in your Ferrari. It’s not like the Ferrari will get you through urban traffic any quicker and you’re more likely to mow down some kid on a bicycle because of the poor visibility out of that low slung bodywork. It turns out the following op amps find their way into headphone products for exactly the same reason—to be seen. It has very little to do with better real world performance or a genuine difference in sound quality. Here are some of the elite:

• OPA627 – The Burr Brown/TI OPA627 might be the Bugatti Veyron of op amps as both offer some of the best performance numbers around. But, just like it’s pretty much impossible to explore the Veyron’s full performance anywhere but on a race track, the same is true with the OPA627. Ferdinand Piëch was playing corporate one-upmanship with the Veyron, and Burr Brown did the same with the OPA627. Basically it’s the op amp of choice for those who keep their towels on in the gym locker room. In exchange for your $24, TI gives you a stereo pair of op amps so quick they can easily go sideways. Unlike the Veyron, there’s no built-in stability control—that’s up to the designer to get right. With a slew rate of 55 V/uS, the OPA627 demands very careful attention to PC board layout, power supply bypassing, parasitic inductances and capacitances, and reactive loads on the output. Basically if you’re only going to drive on public roads in traffic--i.e. headphone audio applications--it’s expensive overkill and more trouble than it’s worth.
• AD797 – The Analog Devices AD797 is probably the OPA627’s closest competition. There’s no question it’s a good part, but it’s also over $10 for a single op amp per package. If anything, due to it’s more rational speed (slew rate = 20 V/uS), and better CMRR/PSRR at power supply ripple frequencies, it’s probably the better part for audio applications. But this game is all about bragging rights. So, sadly, it often gets pushed aside in favor of the more trendy and excessive OPA627.
• OPA604 – The Burr Brown/TI OPA604 is also designed for audio use and, unlike the 627 above, even comes in a dual version. But I’ve never bothered with it after Doug Self extensively tested it and concluded: “it is not very clear under what circumstances this op-amp would be a good choice”. He found the much less expensive 5532 outperformed it in most every way.
• *AD8610 – Analog Devices lists “high performance audio” among the applications for the AD8610/AD8620 and, at around $10 each for the single part and $17 for the dual part, they aren’t cheap either. And, like the choices above, they’re overkill for headphone amp applications. I tested this part in the QRV09 and, while it’s very respectable, I suspect the NE5532 would match it to the limits of most audio analyzers.

HIGH PERFORMANCE: If the above are the ultra expensive supercar elite, these are the more attainable high performance cars. They’re still expensive and demand extra maintenance but they’re more sanely priced and closer to offering performance you can actually use now and then:

• *LM4562/LME49860/LME49720 – National had a dedicated team of high-end audio engineers and they turned out some really great parts. The most popular are these op amps which have found their way into a lot of high-end audiophile blessed gear. The LM4562 and its siblings, in any headphone or DAC application I can imagine, should match any of the elite parts above. That’s objectively on an audio analyzer using the conventional suite of audio tests and subjectively in blind listening tests. The LM4562 was the first op amp to unseat the NE5532 as Doug Self’s overall benchmark. The LM4562 is electrically identical to the LME49720 but cheaper ($3) and comes in a DIY friendly DIP8 package. Samuel Groner did raise some common mode concerns with the LM4562 depending on how it’s configured. Doug Self, who also tested it in some similar configurations, seems less concerned.
• *OPA2134/OPA134 – The Burr Brown dual FET OPA2134, and the OPA134 single version, are the BMW 325 sedan of audiophile op amps. You see lots of them around, they’re not exactly cheap, and they offer better than decent performance. I’ve included this part in my comparison mainly because it’s so popular. They’re $3 – $4 each for the dual part. Personally, for the price, I’d much rather have the LM4562 which outperforms it overall. I had to laugh when I read Doug Self’s critique of this part. Commenting on Burr Brown’s claims of “superior sound quality” Self said, “regrettably, but not surprisingly, no evidence is given to back up this assertion.”
• *OPA2227/OPA227 – This OPA2227 is a more expensive Burr Brown bipolar op amp that offers lower noise specs and some other improvements over the 2134 above. This part is considered an upgrade over the popular OP-27 and LT1007 so I’ve included it in my tests. It has lower noise than the LM4562 but little else to recommend it in typical headphone/DAC applications.
• OP275 – The Analog Devices OP275 part is another part Doug Self rated a solid FAIL. His summary: “it is probably best avoided.”
• Linear Technology – Linear Technology makes lots of op amps with typically each having an impressive spec or two but they don’t seem to have the sort of home run all-rounders to match something like the LM4562. Samuel Groner tested several LT parts with mixed results. He was rather unimpressed with the LT1007 for example as it had obvious crossover distortion and other flaws.

GIANT KILLER: The following op amp deserves its own category:

• *NE5532 – The NE5532, as mentioned in the last article, was originally developed by Signetics mainly for audio use. At the time it, for audio applications, it was like comparing a modern day Ferrari to an old Volkswagen Beetle. It really wasn’t until the LM4562 (see above) came along there was much reason to use anything else for most consumer audio applications—especially if you exclude microphone and phono preamps. There are op amps with better DC specs, faster settling times, higher slew rates, lower offsets, etc. But none of those improvements help 99% of audio gear perform any better. And those “partly better” parts are often worse in more critical areas. The 5532 is the corner stone of nearly all high-end professional audio gear which means most of your music has already been well steeped in 5532 goodness. If you dislike the “sound” of 5532 that means you must also dislike most of the music available.

BARGAINS: When price matters, you can get amazingly good performance if you know where to shop. The best deals I’ve found are from NJR/JRC. For more on NJR see the Circuit Description section of the O2 Details Article:

• *NJM2068 – You’ll find the NJM2068 in a fair amount of pro audio gear—some of it with big names on the outside. I haven’t seen it as much in consumer electronics but it does show up now and then. It has significantly better noise performance than even the NE5532, LM4562 and far better than the OPA2134. It’s $0.39 and requires about a third less power than most of its competitors. It can’t drive as low of impedance loads, and can’t quite match the 5532’s high frequency distortion at higher gains, but it comes so close in most applications it’s essentially a tie.
• *NJM4562 – How about a $3 LM4562 (see above) for $1? That’s pretty much what you get with the NJM4562. My measurements show the two to be very similar with the NJM version even needing less quiescent current. It has slightly more noise than the 2068 above but can better drive low impedance loads, with lower high frequency distortion, at higher gains. For gains < 4X into “easy” (non-reactive) loads of 2K or higher, the NJM2068 is the better part. If you’d rather have the Audi A3 over the very similar but much cheaper Volkswagen GTI, get the National part, otherwise save some money with the NJM4562.

HIGH CURRENT: For driving low impedance and/or significantly reactive loads, there are not many choices in dual DIP8 packages that work comfortably at +/- 12 Volts (24 volts total). The short list:

• *NJM4556 – If you need lots of drive capability the NJM4556 can beat every dual DIP8 op amp I know of. This part is discussed more in the O2 Design Process article.
• *RC4580/NJM4580 – The 4580 isn’t a bad part but, in my tests, it’s inferior to the NJM4556 in pretty much every way. Most significantly it gives up up at around 50 mA while the 4556 can manage double that. And they’re only about $0.20 cheaper.

DON’T BOTHER: While the following parts had their glory days, they’re long gone. They offer no advantage over the above parts and are often much worse:

• *TL071/TL072 – These were mainstream audio parts and used in lots of pro gear before the 5532 came along. For all those speed freaks, it’s worth noting they’re rated only slightly slower than the elite chips at 15 V/uS. They were also among the first reasonably priced FET op amps that were useful for high quality audio. But they suffer from very limited output current, they can’t get very close to the rails, have higher distortion, and they’re relatively noisy. Their lower power cousins the TL061/TL062, and the lower spec TL081/TL082 are even worse.
• LM833 – This was one a National’s early attempts at a high performance audio op amp like the 5532 but they failed to beat the 5532.
• 4560 – The 4560 has been eclipsed by the NJM2068, 5532, 4556, etc.
• 4558 – Same as above
• Most Any CMOS Op Amp – There are lots of CMOS op amps (such as the TI TLC and TLV series) with various marketing claims being made. They’re great for lots of non-audio applications. But I’ve not seen any that can even come close to the $0.39 NJM2068 for serious audio performance into similar loads, etc. Unless you have a specific reason (such as a strict low power/low voltage requirement) to use one there are generally better choices
  

---

MEASUREMENTS

---

WHICH PARTS: To avoid spending 100+ hours on this, I’m only publishing the cream of the crop. Overall I tested nearly two dozen op amps. But it quickly became apparent that many were either obviously inferior, or in this application, approached the limits of the dScope which means they were way past being “good enough”. So I then narrowed the choices down further based on price, popularity, and power consumption. For the output stage it was no contest, the NJM4556 killed everything else. For the gain stage the final list worked out to:

• NJM2068 – Cheap and Cheerful
• NE5532 – The reasonably priced objective benchmark
• OPA2134 – A mainstream audiophile favorite
• OPA2227 – Pushing the high-end of the budget with excellent specs
• OPA2277 – For the low power O2 (long battery life version)
• LM4562 – An audiophile favorite (NJM4562 stocked at Mouser)
• TL072 – A former audio benchmark to put the above in perspective

THE CIRCUIT: The O2’s gain stage at the right was used with various resistor values from R16 and C19 to ground to achieve 2.5X, 3X and 7X gain (as noted in the tests). All the Op Amps were measured with the same 1.5K feedback resistor and 220 pF dominant pole capacitor. Pin 7 was connected to a 10K pot yielding about a 1.3K net load. For the full schematic, and more details, see the Circuit Description section of the O2 Details Article.

ALL THE SAME: Here’s a list of measurements that were essentially unchanged by any of the op amps and why:

• Frequency Response – The 220 pF feedback cap determines the upper –3 dB point (~250 Khz) with all the op amps and the low frequency limit of just the gain stage was down to DC. There were no measureable differences. The only thing that would cause a problem would be high frequency instability might cause a rising HF response. But I checked for that where applicable.
• Slew Rate – All were over the 2.8 V/uS requirement for the O2. Put another way, all were over the requirement for any dynamic headphone amp.
• Square Wave/Stability – Stability issues showed up in the THD vs Frequency sweeps as added high frequency distortion or worse. So I did check the square wave response for any op amps showing unexpected behavior. None of those op amps are shown below (i.e. everything measured below was stable as far as I know).
• Phase – This is set by the 220 pF feedback compensation capacitor to less than 1 degree error at 10 Khz with all the op amps.
• Crosstalk – The internal crosstalk in any of the parts I tested was lower (better than –100 dB) than the rest of the circuit (such as the volume control and headphone jack) so it wasn’t a significant factor.
• THD 20 hz & 20 Khz – These are mostly covered by the THD vs Frequency sweeps.

BEING LAZY: I probably should have tested these, but didn’t:

• Clipping Performance – I probably should have verified the clipping behavior on a scope, but I didn’t. The good news is the THD vs Output sweep tends to show red flags for any weird clipping behavior and all op amps were driven to clipping in that test.
• SMPTE & CCIF IMD – IMD testing might have revealed some minor differences. But the few I checked were so similar I didn’t bother going further. I suspect only the TL072 would have stood out.
• DC Offset – It doesn’t matter in the O2 because the gain stage is AC coupled to the output stage so it wasn’t worth measuring. The datasheets should yield accurate numbers for those interested.
• Distortion Magnifier – It would be interesting to use a distortion magnifier to increase the resolution of the dScope distortion measurements by an order of magnitude to reveal more differences. I have one, but it’s a lot more work to use than the dScope’s native hardware. I’m really not that concerned with distortion in the “triple zero” range given that it’s at least ten times below the most conservative estimates as to what’s audible (0.01%). The dScope reads as low as 0.0005% on its own. I’m not trying to count DNA strings in an electron microscope. I just want to listen to music!

dSCOPE RESIDUAL THD+N: Several things are important to know about the dScope. First these tests are using its Continuous Time hardware analyzer which computes the THD+N in real time. This dramatically speeds up sweeps with 100 points for each pass but it also limits the accuracy and ultimate noise floor compared to using FFT post-processing. The white line in the two graphs below is the residual noise floor of the dScope in loopback and the THD+N is as low as 0.0005% which is plenty low enough for these measurements. But, op amps have become so good they can easily approach or even better that. Keep in mind that 0.0005% is the combined distortion of the signal generator, output amplifier, all the balanced isolated input circuitry (which can handle anything from a few microvolts up to 200 volts RMS!), and the 24/192 A/D in the dScope. So there’s a lot more going on in the dScope than just a single op amp so 0.0005%, viewed in that light, is rather impressive.

DROOPING HF DISTORTION: It’s somewhat debated, but the generally accepted practice limits THD+N measurements to the audio band. This has especially been true in the “digital age” where there can be significant out of band noise from digital circuitry, switching power supplies, etc. If you include the out of band stuff, the entire noise floor rises and masks the true audio performance below 20 Khz. So I’ve cut all but one of the sweeps off at 22 Khz which is the default standard in the dScope. The distortion drops as the harmonics fall above this frequency. For example the third harmonic of 8 Khz is 24 Khz and would be rejected by the analyzer.

JAGGED LINES & LEGENDS: The jagged bits in the traces are mostly the input circuitry auto-ranging to keep the signal within 2 dB of the dScope ADC’s 0 dBFS point. This maximizes the dynamic range of the analyzer but causes some “steps” in the sweep when the relays switch different attenuators into the circuit. Audio Precision analyzers do the same thing (unless you average it out with lots of smoothing).  Finally the “A” in the legend names is “Channel A” of the analyzer and it’s added automatically. It’s not my usual poor grammar.

THD+N vs OUTPUT: The most interesting thing about this result is how boring it is. Note: This is just the gain stage. The dScope was connected directly to the output pin of the op amp being tested (the NJM4556 output stage is not included). The default NJM2068 is shown at gains of both 2.5X (yellow) and 7X (aqua) while the others are shown at only 7X. The OPA2134 (blue) and the NE5532 (pink) get slightly closer to the rails before they clip but otherwise deliver very similar performance—all 3 parts are so close to the analyzer’s residual it’s mostly a tie. I didn’t even include the OPA2227 or LM4562 as the lines were right on top of the NE5532. The TL072 (orange), however is a different story. The TL072 clearly isn’t liking the ~1.3K load. Note the graph has been re-scaled from my usual 1% THD to only 0.1% at the top. The DC rails are +/- 11.85 V:

 

LOW POWER OPA2277 THD+N vs OUTPUT: Here’s the low power op amp I’m recommending for the O2 low power option (25+ hour battery life) at both gain settings compared to the standard NJM2068. While the clip points differ by about 10% the distortion performance is very similar. The OPA2277 needs only 0.8 mA per amp or 1.6 mA total vs about 4.5 mA total for the NJM2068. The big downside is it’s more than ten times the cost ($4.40). It’s the best performing low power part I found, many others like the On Semi MC33078, TL062, NJM022, etc. are seriously awful:

 

THD+N vs FREQUENCY: Here are the favorites plotted vs frequency also taking the output directly from the gain stage op amp under test. Again, I didn’t bother with the expensive OPA2227 or LM4562 as they just further blurred everything together with the residual of the analyzer. At 2.5X gain the NJM2068 matches everything else. At 7X gain (aqua), however, it’s not quite as happy with the ~1.3K load and starts to show the limitations of a $0.39 part. However, to put it’s 7X gain 0.0018% max THD+N in perspective, consider that’s 28 times lower than the 0.05% high frequency requirement. Even if you use the much more stringent –80 dB threshold of 0.01% it’s still more than 5 times lower. So it’s extremely unlikely to be an audible problem. And, perhaps more important, the NJM2068 uses significantly less power and has significantly less noise. The TL072, on the other hand, is obviously having a hard time as it did in the earlier sweep:

 

LOW POWER OPA2277 THD+N vs FREQUENCY: Here’s the recommended low power op amp versus the default NJM2068. At 2.5X gain the OPA2277 is still under 0.002% while at 7X gain it hits 0.0045% which is still well under the 0.05% high frequency goal and even comfortably under 0.01%. This is a worthwhile trade off for needing only about 1/3 the power:

 

OTHER TESTS: Here the dScope bandwidth extends to 44 Khz testing the older V1.0 O2 board at 3X gain.The measurement is at the headphone output into 600 ohms at full volume. The important thing to note is even  the relatively high-end LM4562 and OPA2227 don’t make any difference:

 

dSCOPE RESIDUAL NOISE: Here’s the noise floor of the dScope in dBV (the unit dBV is always referenced to 1 Volt RMS) in loop back with the output of the generator muted and set for a 25 ohm source impedance. To reference these numbers to the full output of the O2 (7 V RMS) you would add 17 dB so the A-Weighted number would be 137 dBr (ref 7V). Also note the noise floor is well below –150 dB except close to 20 hz (which is actually an artifact). The total sum of the noise out to 22 Khz is how the –116.8 dB number is calculated:

 

NJM2068DD NOISE: The default op amp is impressively quiet even at 7X gain as measured here—especially for $0.39! Being a bipolar input device designed for audio and low noise I would expect decent performance but this is borderline stunning matching even the nearly $5 OPA2227 below:

 

NE5532 NOISE: The professional audio favorite, the NE5532, is about 3 dB worse at 7X gain than the less expensive and less current hungry NJM2068 above but still very quiet. I have heard some versions of the 5532 are a few dB better than others. This was a TI NE5532AP:

 

LOW POWER OPA2277 NOISE: This is the recommended op amp for the low power version of the O2. As is often the case with lower power op amps, despite it’s $4+ price, it has roughly 6 dB more noise than the $0.39 NJM2068 but it’s not too bad all things considered and matches the popular OPA2134. Most of the other low power parts I tested were much worse:

 

OPA134/OPA2134 NOISE: Op amps with FET inputs are typically noisier in many audio applications. And despite the Burr Brown OPA134/2134 being marketed as “low noise” it’s has nearly 3 dB more noise than the much cheaper NE5532 and nearly 6 dB more noise than the cheaper still NJM part. The –130 dB spike at 60 hz might be related the high impedance FET inputs or it could just be changes in the AC field around my test bench but it’s not significantly affecting the numbers. You can plainly see the entire noise floor is about 3 dB higher than the graph above:

 

OPA227/2227 NOISE: This is the most expensive op amp at nearly $5 and, for 12 times the cost of the NJM2068, it only manages to match its noise performance! This is a bipolar input device specifically designed for low noise applications in professional audio equipment. In the O2 this would be a complete waste of nearly $5:

 

LM4562 NOISE: While the National LM4562 isn’t available from Mouser, the NJM4562 is and they perform very similarly. Here’s the genuine National part and it’s noise performance turns out to to be nearly identical to the NE5532:

 

TL072 NOISE: The TL072’s FET inputs and ancient design really show here. It’s nearly 12 dB worse than the NJM part which is fairly huge—especially considering the TL072 is more expensive (about $0.60). The 60 hz spike is back which has me wondering if FET inputs and EMI susceptibility go together (see the FET OPA2134 above):

 

MEASUREMENT SUMMARY: While there are some differences, if you exclude the ancient TL072, everything else delivers very similar distortion performance, especially at more modest gains. There are perhaps more significant differences in noise performance and that’s where the NJM2068 excels. The NJM2068 might have slightly more high frequency distortion at 7X gain, but it also has 3 dB lower noise and uses 1/3 less power than the NE5532. The overwhelming conclusion is you can spend less than $1 to get performance well beyond the point of diminishing returns. Spending more doesn’t buy you anything useful in this typical application. Hopefully this article at least helps demonstrate there are not big differences between appropriate op amps in a well designed circuit. If you had a headphone amp with an NE5532, and tried to “upgrade it” to an OPA2134, you would only end up with about 3 dB more noise for your trouble.

O2 Details

TWO BUCK CHUCK:  The Objective2 (O₂) headphone amp design has already proven surprisingly popular. Just as a $2.49 wine can beat much more expensive wines in blind tasting, so can an inexpensive headphone amp. In this case, with as little as $25 worth of parts, you can have a headphone amp that rivals the headphone output of the Benchmark DAC1 Pre in blind listening tests.

START HERE: There a lot of documentation on the O2. If you landed on this article first, a better place to start is probably here:

• O2 Summary

$150 FULLY ASSEMBLED & COMPLETE: An experienced USA DIY builder is offering to sell complete O2 amps for around $150 with everything included. Made in China it could easily be $99 or less. See O2 Resources below.

LESS WORRY: Unlike 99% of headphone amps, the O2’s performance has been fully documented on professional equipment under real world conditions. The results were published in the first O2 article. Some of the competition would rather make vague and sometimes even misleading claims leaving you to wonder. And a lot of amps only work well with some headphones but are a poor match with others. The O2 comes close to a one-size-fits-all portable amp. For the history see the first article covering the premise, and the second article discussing the design process.

THE PRICE DETAILS: All the parts to build the O2’s fully functional circuit board should be around $30. Just add two batteries, and perhaps the 2011 version of a Cmoy Altoids tin, and you have a fully functional very high quality headphone amp for under $40. If you want a extruded aluminum case, pre-made front panel, rechargeable batteries, and AC wall adapter, just add another $45 or so. Throw in shipping costs, taxes, title, and insurance, and you can still might headphone amp objective nirvana for under $100. The price details are all in the Bill Of Materials. If someone wants to roll the dice and make a few hundred O2 amps things get even better. They can buy in volume, run the circuit boards on a wave solder line, and probably sell the O2 at a nice profit for $99 or less.

THE COMPETITORS: To put the O2 in perspective, you might want to compare the O2 against the $180 (assembled) Mini3, $350 - $400 HeadAmp Pico, $450 Headroom Portable Micro, or $650 SR-71 Blackbird. I’ve already shown the O2 far outperforms the Mini3 in every single category. And I’m 99% sure it will also, overall, outperform the others. I believe it will even embarrass many far more expensive desktop amps in accuracy—i.e. distortion, output impedance, noise and real world power requirements. Sorry if I’m being less than modest, but I’m trying to get the point across $30 worth of parts really can rival or beat much more expensive products.

THE CHALLENGE: I’ve offered both an objective and a subjective (listening) challenge for the O2. If you have an amp you think can beat the O2 on either playing field, check out: An Open Challenge.

THE RESULT: Even if the O2 loses a challenge, everyone else wins. The whole idea is to raise the bar. Until now, if you only wanted to spend a few hundred dollars, the bar has been set surprisingly low mostly in a quagmire of high distortion, limited output, high output impedance and other difficult to excuse problems. Or it’s been seriously expensive to buy genuinely good objective performance from a company like Violectric. The reality is you can have your cake and eat it too—for around $40 – $150.

THE LICENSE: Anyone is free to use the O2 design, as presented here, if they comply with the Creative Commons License. I don’t want any revenue from the O2 but I do humbly request everyone please respect the license which includes proper attribution. It’s good Karma. Also please note while the design is open (i.e. the files on Google Docs) the content on this blog is copyrighted and cannot be used without permission.The O2 is Open Source Hardware under a Creative Commons license. In addition to the language in the CC-BY-ND License, the O2 design is offered “as-is” with no warranty of any kind, either expressed or implied, including its suitability for anything you might want to use it for. You build and use it at your own risk. In plain English: If you can’t make the O2 work correctly, or as you had hoped, or damage your headphones, or your house burns down and your wife leaves you, you can’t blame me. Any similar DIY project has similar risks. For more, please see: Open Source Hardware

This work is licensed under a Creative Commons Attribution-NoDerivs 3.0 Unported License.

THE RESOURCES: Pictured to the right (click for larger) is the final PCB version. Here the resources for the O2 project:

• O2 Summary – A concise summary of the O2 project with links to all the appropriate sections for more details. Consider it a landing page or table of contents. 
  
• The First Article (premise and measurements) – Why I tackled this project, the basic design goals, blind testing, and the detailed measurements
  
• The Second Article (design process) – The “how” from a design perspective, detailed performance goals, and the design methodology.
  
• This Article – How to build your own O2 including options, buying parts, usage, and a detailed circuit description
  
• O2 on diyAudio – This is the best place for the DIY crowd to discuss building the O2, creative enclosures, add-ons, etc.
  
• O2 on ABI – AnythingButiPod is a great friendly site and it’s the best place for non-DIYers to discuss the O2. The more hardcore DIY guys are probably better off at diyAudio.
  
• Documentation Package – Revised November 30th 2011. The documentation is available in PDF format on Google Docs.
  
• Bill Of Materials Spreadsheet – Revised November 30th 2011. This is the parts list (included in the PDF above) in interactive spreadsheet form on Google Docs. You can download the spreadsheet and even cut and paste from right into Mouser’s website to order parts.
  
• PC Board Gerber & Drill Files – Revised September 14th. These are the files for getting PC boards manufactured. The September revision has several improvements including accommodating a wider variety of volume controls and battery terminals. Check the included readme.txt file for the PC Board specs and file descriptions.
  
• Front Panel Express CAD File No Lettering  – Here’s the basic front panel for the B2-080 enclosure without any lettering which prices out under $17. It’s in the native format for the free Front Panel Designer software. There’s also some talk on diyAudio of a front panel group buy. See the link below.
  
• Front Panel Express CAD File With Lettering – Here’s the front panel for the B2-080 with simple lettering. It’s more more expensive depending on the type of lettering (bare exposed aluminum or in a contrasting color). Also see the group buy possibility below.
  
• JDS Labs Pre-Assembled O2 Boards – JDS is selling in their online store pre-assembled O2 boards and possibly other components like bare PC boards and front panels.
  
• Jokener Worldwide Parts Group Buy – Jokener is operating his second group buy for O2 boards and parts out of Europe.
  
• Epiphany Acoustics Complete O2 Pre Order – Epiphany Acoustics in the UK is gauging interest in completed amplifiers. If they get enough interest they may put the EPH-O2 into production and have complete amplifiers in stock.
  
• Aerohoff O2 – Assembled boards and possible laser cut acrylic front panels (if he has any left or is doing another run)
  
• Front Panel Group Buy – Flynhawaiian may still be doing group front panel sales through FPE.
  
• PC Board UK Group Buy – A gentleman in the UK on diyAudio generously organized a group buy that’s now closed but the thread is still a valuable resource. There may be some surplus boards still available.
  
• O2 MrSlim Build Service – MrSlim, an experienced custom DIY builder, has offered to use the above PC boards to build O2s. The cost, excluding a few items, is around $115.
  
• GB, EU, Parts Group Buy – Jokener on diyAudio generously volunteered to put together kits of parts for European O2 builders to save on shipping costs, etc. This is also closed.
  
• Australia/New Zealand Group Buy – Blue Fusion offered a parts group buy as well and it’s closed.
  
• O2 Wiki – Started by a volunteer. Hopefully it can become an easily referenced repository for helpful future O2 info. This article and several of the other has “anchors” that allow linking to specific sections. If anyone wants to turn these links into a Table Of Contents for the Wiki please contact me using the link on the right of this blog.

 

---

Important Details (and what’s new)

---

POWER CYCLING & NOVEMBER DESIGN CHANGE: I revised R25 from 2.7M ohms down to 1.5M ohms and R9 from 40K down to 33K. These changes are related to some users reporting their O2 would rapidly turn on and off when the batteries were low. This issues seems to be very battery specific (my O2 amps shut down OK). The new values should help it stay off at least long enough to turn it off. Either way it’s a signal your O2 needs to be shut off and charged. The BOM and PDF files in the Resources section were revised.

MOUSER INVENTORY: As of November 30th everything on the current BOM, except the volume control, is in stock and all prices are correct. The first alternate volume pot is in stock and differs only in taper. See: Obtaining Components.

CHANNEL MIS-MATCH & GAIN SWITCH CLEARANCE: The corner of the gain switch closest to R21 may touch a via (small exposed pad on the PC board) and cause a gain change in one channel making the channels sound unbalanced. See the Circuit Board Construction section.

METAL SHAVINGS FROM CASE SCREWS – A few users have reported the screws to attach the panels to the Box Enclosures case produce aluminum shavings that can end up creating unwanted connections on the PC board (short circuits) which may damage the amp, possibly even headphones and/or at least cause odd behavior. So, as a precaution, install all 8 self-tapping screws first into an empty case, remove them, and blow out any shavings that might be created. Once the holes are “tapped” there shouldn’t be a problem.

GET YOUR OWN O2: The biggest question most have is “how do I get one?” Here are some possibilities:

• Perfboard DIY – Don’t try this! While a basic Cmoy might work cobbled together on perfboard or protoboard the O2 is different. The higher performance and greater complexity mean you’ll almost certainly have problems without using a proper PC board.
  
• Get Your Own PCB Made – You’re free to use the gerber/drill files (see: O2 Resources) and get your own PC board made. There are many PC board services that cater to hobbyists and small quantities.
  
• Buy A Bare PCB – See O2 Resources for PCB sources.
  
• Group Buy – See O2 Resources for existing group buys or start your own.
  
• Commercial Pre-Assembled Boards/Kits – See O2 Resources for existing pre-assembled boards, etc. This eliminates the need for soldering or special skills. 
  
• Professionally Hand Built – See: O2 Resources
  
• Commercial Complete Amps – If the O2 becomes popular enough, it’s possible one or more sources may decide to build a production run of O2s. These could be sold at one of the online dealers specializing in headphone gear and/or on eBay. See O2 Resources.

PLEASE DON’T MESS UP A GOOD THING: Many of the smaller commercial audiophile manufactures apparently lack the ability to properly measure the performance of their products. And DIY designers are typically even less well equipped. For these reasons, and the profit driven desire of some to cut corners, the O2 is offered only under a “no derivatives” license. Put simply, you can’t modify it or use it as part of a different design and then offer it to others without prior approval. This is to help prevent inferior versions of the O2 that might disappoint whoever buys/builds them. Even the PC board design is an essential component of the O2’s performance and changing it could have dramatic consequences. If you want to modify the design, and offer it to others, please contact me first. I’ll be investigating O2 designs offered commercially and alerting everyone if I discover unauthorized changes or performance problems. The idea is the best implementations of the O2 should win.

WORK IN PROGRESS: Like the previous two articles, this one is a work in progress. I have added several sections and made many revisions to the first two articles and will be doing the same with this one.

 

---

Cautions!

---

HEADPHONE DAMAGE: While the O2 uses current limiting to decrease the risk, it still has enough output power to damage some headphones—especially small in-ear types. This is one of the few negatives to the One Size Fits All approach. It’s important to operate the O2 in the Low Gain mode with delicate headphones, keep the volume down, and generally be careful. The O2 can be easily modified without a soldering iron to reduce the gain for use with sensitive in-ear headphones. This reduces the risk considerably. See the Sensitivity & Gain section below.

HANDLING THE PC BOARD: If you operate the O2 with the bare PC board exposed (not in an enclosure) be aware as much as 60 volts DC (+/- 30V) can be present between certain points on the top and bottom of the board. While that’s not generally dangerous it’s not a good idea to handle the board when the AC adapter is connected and/or the batteries are installed. Don’t touch any connections on the board while the amp has power, or place the board on any metal surfaces or have anything conductive nearby (like wire clippings, etc.). You could damage the amp and there’s a risk of electric shock. This is true of many headphone amps (and much more so when the include power line mains voltages).

BATTERIES: Don’t short out the Ni-MH batteries. Alays make sure the amp is turned off when removing or installing the batteries. Observe the correct polarity when connecting them. Don’t under estimate their ability to get things hot, damage parts, etc.

SMART HEADPHONE USE: Always turn the volume fully down when you plug in or unplug headphones to help avoid damaging your headphones and/or the O2. Be cautious with the volume control as the O2 has enough power to possibly damage some headphones.

HEARING DAMAGE: The O2 is capable of more output than most headphone sources so it’s possible to listen at levels known to cause permanent hearing damage with most any headphone. Even extended listening at an average level of 85 dB SPL can cause hearing damage. If you don’t know what’s safe here’s a good article: Loud Music Sucks

THOSE NEW TO DIY: While the O2 is certainly easier and safer to build than many projects, especially those involving surface mount technology or 120/230 volt line power, the O2 isn’t intended to be a step-by-step project for novice DIYers. If you encounter problems the O2 doesn’t come with free technical support (beyond others possibly responding to questions on diyAudio, etc.). If you’re hesitant about your DIY skills, or any other aspect of this project, you should wait until several others have built one and there’s more of a support network. Or just wait for pre-assembled boards to become available. Please understand if you choose to build the O2 it is entirely at your own risk.

MAXIMUM INPUT: Using the default Low Gain setting of 2.5X it’s extremely unlikely the O2’s input will be overloaded by any reasonable fixed output source. But it could happen at higher gain settings. If you plan to use more than 2.5X gain, especially with an unusually high output source, see the Maximum Input section.

POWER MANAGEMENT CIRCUIT: Do not eliminate or bypass the power management circuit even for desktop only operation without batteries. It is still necessary to prevent potentially dangerous transients on power up and power down.

ERRORS & TESTING: There are some tips in the Testing section to help make sure your O2 is working correctly before plugging your headphones in. Like any most high quality amp, errors in the assembly of the O2 could cause large amounts of DC to appear on the output which may damage headphones.

SINE WAVE TESTING: While I’ve abused the O2 in a variety of ways playing real music into assorted worst case loads, and it has survived nicely, it will eventually overheat the output op amps with sustained high power sine wave (or square wave) testing into low impedance loads. This is true of many high power headphone (and loudspeaker) amps that are designed for music not test signals. It’s important to limit full power sine wave testing to a few seconds below about 150 ohms and let the output ICs cool down between tests. You should also run sine wave tests with at 15 VAC or greater wall transformer. This is discussed more in the Circuit Description section.

 

---

Using The O2

---

INTRO: This section is for anyone considering the O2 and covers the operation and a few things that might not be obvious.

THE BASICS: The O2 is fairly simple to use. From left to right on the front panel:

• Power Jack – For AC line power or charging, connect the AC wall transformer to this jack. The standard plug is 5.5mm outside and 2.1mm inside. The transformer output can be anywhere from 14 VAC to 20 VAC and at least 200 mA. Don’t use more than 20 VAC or less than 14 VAC (except for the 12 VAC transformers listed in the parts list). Do not try to use a DC adapter. A DC adapter may appear to work if batteries are installed, but the amp will be at least partly running off the batteries. 
  
• On/Off Switch – The switch by the power and headphone jack is the on/off switch. Press it in to turn the amp on. The LED should come on.
  
• Output Jack – The left 3.5 mm jack (J2) by the power switch is the headphone output. If your headphones have only a 1/4” plug, use a “pigtail” style adapter. Don’t try to use a rigid one piece 1/4” to 1/8” adapter. These adapters create a very long “plug” that puts excessive stress on any 3.5 mm jack and careless handling can damage or break the jack.
  
• Power LED – This tells you the amp is on. If the batteries are low the amp may not operate but the LED will remain on as a reminder to turn the amp off (and hopefully charge it soon).
  
• Volume Control – Always start out with the volume all the way down to avoid unwanted surprises or possible headphone damage.
  
• Gain Switch – The out position is Low Gain (2.5X or 8 dB) and the in position is High Gain (6.5X or 16 dB).
  
• Input Jack – Connect your source to the 3.5 mm (1/8”) jack (J1) on the far right of the board. It can be connected to the line outputs of most anything or driven from a headphone output (although see the Maximum Input section below if using the High Gain setting). RCA to 3.5 mm cables are widely available for connecting to a home source as are LOD cables for portable players. It’s best to always have the amp turned off or the volume all the way down when making connections. This is true with any high powered amp.

BATTERIES: The batteries are optional and not required for AC operation. If you want to use batteries, consider the following:

• TYPE OF BATTERIES - The O2 is intended to use Ni-MH “9 volt” (usually 8.4 volt rated) batteries. The Tenergy 8.4V batteries in the parts list are about 9.5 volts when fully charged and have 7 cells inside. There are also 6 cell (7.2 volt) and 8 cell (9.6 volt) batteries. While the 6 and 8 cell versions will sort of work, they’re not recommended. Alternately, you can also use disposable 9 volt alkaline batteries as you would in a Cmoy. Do not connect the AC power with Alkaline batteries installed (they’re not rechargeable).
  
• INSTALLATION – Observe the correct polarity! It’s best to slide the board out and clip the batteries in place rather than trying to slide them in from the back of the case. You also need a small piece of foam or something similar to hold the batteries down against the board so they don’t come loose. Thin double sided foam tape to stick them to the board is also a good idea if you plan to leave them in a while. Loose batteries bouncing around inside could be bad! If your batteries are a bit shorter you may also need some sort of “shim” between the base of the batteries and the back panel to make sure they stay snapped in place.
  
• RUN TIME - The default design with 250 mAH Ni-MH batteries has 7 – 9 hours of run time on a charge. The Low Power version uses more expensive low power op amps and has somewhat more (but still very likely inaudible) distortion but the run time is around 20 - 30 hours. See: Low Power. Higher capacity batteries in the 250 – 300 mAh range will extend the battery life with either version.
  
• LOW BATTERY SHUTDOWN - The power control circuit in the O2 monitors the batteries and automatically turns the amp off if they drop much below 7 volts each. The amp will simply stop playing, or possibly cycle on and off, but the LED remains on to remind the user to turn the O2 off. Either way you should turn it off and charge the batteries as soon as possible.
  
• BATTERY CHARGING - The batteries charge whenever AC power is connected with the O2 turned on or off. The charge current automatically tapers off as they reach full charge and you can leave the AC power connected indefinitely if you want. But if you’re going to use the O2 mainly as a desktop amp for months at a time you should remove the batteries. Charging time depends on how low the batteries are and which batteries are used but will generally be around 8 – 24 hours.
  
• BATTERY SHELF LIFE - Most Ni-MH batteries have a significant self discharge rate. This means just sitting on the shelf with the O2 turned off the batteries are slowly dying. You can get “low self discharge” batteries that are significantly less prone to this problem. Just like an iPod or other portable player you should charge the O2 at least once a month even if you’re not using it. If the batteries are left dead for an extended time they may be damaged.

HEADPHONE SENSITIVITY & GAIN: Always start with the O2 set for Low Gain when using a new source and/or new headphones. Only try the High Gain setting if it won’t play loudly enough using Low Gain. Always turn the volume all the way down before selecting High Gain. For more information, please see Gain Settings.

MAXIMUM INPUT: This is something many don’t think about but many headphone amps have a maximum input level before the input stage “clips” creating excessive distortion. The FiiO E5 and E7, for example, are limited to about 1.2 V RMS before their inputs overload. The desktop FiiO E9 overloads at 2.1 V. This is true at any setting of their volume controls. The O2 was designed with the volume control “in the middle” for maximum performance (as explained in the O2 Design article) but this means some care may be required to not overload the input stage at gains above the default 2.5X. If you’re going to change the O2’s gain or use the high gain mode, be aware of the following:

• Gain Explained – See: All About Gain.
  
• Low Gain = No Worries – The O2 is unlikely to have any problems at 2.5X gain or less on AC power with home sources or on battery power with portable sources. It can handle up to about 2.8 V RMS on AC power at 2.5X gain which is higher than just about any source I know of that doesn’t have a volume control. On battery, as would be used with lower output portable sources, the O2 can handle up to 1.8 V RMS input. That’s higher than any portable source I know of including 5 V USB powered DACs. If that works for you, and it probably will, you can ignore the rest of this section.
  
• Source Volume Control – The sources that may possibly overload the O2 also often have volume controls which eliminates the overload problem. You can just leave the volume all the way up on the O2 and use the volume control on your source (which might be more convenient anyway). Used in this way, the input stage will never overload. The O2 will reach maximum output first.
  
• Distortion – If you hear distortion, switch to the Low Gain mode or try turning down the volume control on your source (if it has one).
  
• Portable Sources – You can normally use Low or High Gain as most appropriate. Most portable players output around 0.5 V via their line outputs (LOD) if they have one and 0.5 – 1 volt from their headphone jacks where you can use their volume control to set the level. So even at the 6.5X High Gain setting you won’t overload the input.
  
• Home Sources – For most home sources you want to use the default Low Gain setting of 2.5X (8 dB). The Redbook digital audio standard for home line outputs is 2.0 V RMS. A few, like the HRT Music Streamer II, go a bit higher to 2.25 V RMS which is still well under the 2.8 V limit.
  
• The Output Clips First At Max Volume – When the O2 is set to max volume the output stage clips first at a level that far exceeds what your ears and/or headphones can likely handle. So, in practice, the input limit isn’t much of a limitation. In the real world most will never need anywhere close to the O2’s full output capability. So they can comfortably use lower gains and clipping of any kind won’t be an issue. The O2 has at least 6 dB of total headroom over most portable amps like the Mini3, etc.
  
• AC vs Battery – Because the supply voltages are lower during battery operation the O2 has lower limits when operating on battery power. This normally isn’t an issue as you’re likely to be using a low output battery powered source if the O2 is running on batteries. But don’t try to use a home source while operating the O2 on battery power unless you know it won’t be a problem. You won’t harm the O2, but you might get distorted sound. 
  
• The Math – For those who want the exact numbers, or want to change the default gain settings, on AC power the Maximum Gain = 7 / Vin_(max) and on (low) batteries it’s 4.5 / Vin_(max). So if you have a 2 volt CD player, the max gain is 7/2 = 3.5X on AC power. If you have a 0.5 volt LOD portable it’s 4.5/.5 = 9X running on the batteries.
  
• More Information – This was an intentional design trade off for maximum performance. See Gain Stage Overload in the Circuit Description section below. For more on calculating gains, see: All About Gain.

NOISE: The O2 by itself is dead silent. If you hear noise it’s coming from something else. Try your most sensitive headphones with nothing plugged into the input jack at any volume setting if you have any doubts. The O2 may pick up noise with an un-terminated input cable. There can also be noise if it’s connected to a source that’s turned off (or portable device sleeping). This is normal and will happen with many amps. If you hear hum with the O2 running on AC power make sure the case is grounded as documented in the next section.

 

---

Circuit Board Construction

---

SAFETY: Always unplug the AC adapter and remove the batteries when working on the board, moving it around, handling it, changing connections, etc. Ni-MH batteries can deliver considerable current and there can be as much as 60 volts DC present across the large filter capacitors. The capacitors will store a charge if the amp is turned off even after the AC power has been disconnected. Don’t have metal objects laying around (like stray lead clippings, tools, etc.) that might come into contact with the board while it’s powered up. If you plan to use the O2 without an enclosure, add some stick on feet to the bottom and consider using blue painter’s masking tape to cover the bottom of the board to help prevent accidental contact, shorts, etc. See the Cautions section.

BUILDING THE CIRCUIT BOARD: The design documents with a schematic, parts list and board drawing are linked under The Resources in the first section of this article.

REQUIRED TOOLS – Here are the tool basics if you’re new to DIY. Everything below can be ordered from Mouser with the parts and Circuit Specialists to save on shipping. You might also find even better deals on eBay or elsewhere:

• Soldering Iron - At a minimum you need a 25 watt or smaller soldering iron or ideally one with active temperature control. I’ve used this one and it’s the least expensive station I know of with true temperature control (which adds more heat to this tip as you solder for better connections). It’s basically a $30 Chinese clone of a $80 Hakko professional station. A cheaper option might be something like this but I haven’t used one and it’s not a closed loop temperature control like the $30 model. The tip should be fairly fine but not surface mount tiny. Plated tips work vastly better than un-plated ones. Non temperature controlled soldering irons tend to be hard on tips. They get oxidized from too much heat when they sit idle and then won’t tin properly. If the entire tip is not bright shiny and evenly “wet” when you wipe it with the iron on you need a new tip (or a better iron).
  
• Solder – High quality 60/40 or 63/37 solder is essential. Ideally it should be rosin activated (RA). I don’t recommended no clean or lead free solder but if you have some you like, that’s fine too. An ideal gauge (thickness) is around 0.032 inches (22 AWG). There is a recommend solder on the BOM. Kester 44 is also excellent. Just make sure you don’t use anything with corrosive flux as that will slowly eat up your solder connections long after you complete your amp. Most people I know who have trouble making good solder connections are using Radio Shack or hardware store solder that is better suited for stained glass windows than fine electronics.
  
• Solder Wick – Some solder wick (such as the Techspray 1810-5F on the BOM) is almost essential to fix mistakes.
  
• Cutters & Needle Nose - You also need some electronics sized semi-flush cutters (i.e. Xcelite 175 ($8 at Mouser), or for bargain shoppers these) to trim the leads, and a pair of small needle-nose pliers (such as these).
  
• DMM - Even a really cheap DMM, is also very useful (see the Testing section). They can be found on eBay for even less.

IMPORTANT!!! I know there’s a lot to read in this article, but unless you’re a very experienced DIYer who can build things blindfolded it’s very important to read through the following before you start building the circuit board. It’s very difficult to remove parts once they’re soldered in and with nearly 80 parts there are many chances to make mistakes or overlook something important. The O2 also has some unique requirements that, if ignored, will cause it not to work correctly or worse.

CONSTRUCTION TIPS: Experienced DIYers probably won’t have too much trouble but there are some things to watch out for with the O2. Novices should pay close attention to everything below:

• S2 PROBLEM – The September version of the PCB artwork (9/14/11 release) has the via between R21 and S2 located too close to the corner “foot” on S2. While most have not had any problem, it’s highly recommended to slightly bend the leg in that corner of the switch inward, or if you have some heavy cutters, trim that leg shorter (don’t use small “precision” cutters as you’ll likely damage the blades). At the least, make sure there is no contact when you’re soldering S2 in place.
  
• ESD (static discharge) – The semiconductors will come in ESD protected packaging with warning stickers. Most of the components in the O2 are reasonably robust but use care if you’re working in dry conditions, on carpet, etc. But the MOSFETs require special care regardless of your working conditions.
  
• MOSFET HANDLING - Q1 and Q2, the MOSFETs, are be extremely sensitive to ESD damage! There have been reports of ESD damage from those building the O2. First, please see Wikipedia ESD for some general information. Handle the MOSFETs only by the body and never touch the pins. Also beware of the pins touching anything conductive including the O2 circuit board. If you’re hanging onto the metal tab, and you touch the gate pin to something conductive, any static charge in your body will be discharged through the MOSFET likely damaging it. If either is damaged the O2 either won’t turn on properly, or could appear to work fine, but won’t properly shut down if there’s a power problem Either could damage your headphones!  Install at least all the resistors in the power management circuit before installing the MOSFETs and ideally wait to install the MOSFETs until you’re ready to start the first board tests and U2 is installed. The resistors help protect the MOSFETs while you’re handling the board. If you have one, wear a wrist ground strap, use a grounded soldering iron, and always try to avoid “completing a circuit” (loop) that includes the MOSFET gate pin (the pin closest to the batteries) in any way. It only takes around 30 volts to damage the gate and your body and other objects can easily have a static charge that’s much higher.
  
• METAL SHAVINGS FROM CASE SCREWS – A few users have reported the screws to attach the panels to the Box Enclosures case produce aluminum shavings that can end up creating unwanted connections on the PC board (short circuits) which may damage the amp, possibly even headphones and/or at least cause odd behavior. So, as a precaution, install all 8 self-tapping screws first into an empty case, remove them, and blow out any shavings that might be created. Once the holes are “tapped” there shouldn’t be a problem.
  
• Component Heights – U5, U6 and possibly Q1 and/or Q2 are the tallest things on the board using the specified components. They must be soldered so they’re nearly fully down in the holes (less than 0.7 in/18 mm high) with the wider part of the leads in the holes. Leave a bit of room (about 0.05 in or 1 mm) under the parts so the leads can absorb stress. If they’re taller than 0.7 inches the board won’t fit in the B2-080 case. If you have any doubt solder just one pin, clip some of the excess lead off and test fit the board before you solder the other pins as they’re difficult to reposition once fully soldered in..
  
• Trimming J2 – The outermost pin of the input jack, J2, needs to be clipped very close to the board so it won’t touch the “screw rail” in the B2-080 case. Don’t put a lot of solder on the pin as it needs to end up nearly flush to the board. Failure to clip this lead correctly may connect one input channel to the case leading to noise problems or one channel not working. When you slide the board into the case check for any contact with the board in a worst case position in the slot.
  
• Battery Terminals – The larger female terminals are installed in the “+” locations to the left closest to the edge of the board (these interface with the smaller terminals on the battery). Make sure you get this correct as they’re very difficult to de-solder and you may damage the amplifier with the batteries reversed. The terminals are a bit tricky to solder without burning yourself if you try to hold them in with a finger while soldering. They’re also tricky to align or the batteries may not fit properly. The holes are sized with some extra room to allow use of both Eagle and Keystone terminals. A dead disposable battery is highly recommended to clip into place before soldering the terminals but don’t try using a “live” battery—especially a Ni-Mh rechargeable. If you’re going to use thin double sided foam tape under the batteries to secure them to the board (recommended) put a piece of the tape on the bottom of your dead battery before soldering the terminals to adjust for the thickness of the tape. You may also want to add solder from the top side of the boards if it doesn’t flow all the way through from the bottom to make the terminals mechanically stronger. Wire leaded battery clips can also be used if you have other ways to secure the batteries and trim the leads to be as short as possible.
  
• Volume Control Options – There are several part numbers listed for volume controls. If you use one of the 15mm shaft versions, you have to cut off the small "nub” on the front of the body BEFORE you install the pot or it may interfere with the front panel. It’s extremely difficult to cut off after the pot is soldered in place. This is really easy to with a pair of diagonal cutters as the cast metal is fairly brittle. It just cleanly snaps off. Don’t grind or file it as the metal dust could end up inside the pot. The 15mm pots go in the set of 6 holes closest to the edge of the board. The 20mm pots go in the other set marked “20 mm”. When you have it right the end the shaft should be about 0.625” or 15.8mm from the edge of the board.
  
• Front Panel Component Alignment - To allow for manufacturing tolerances there’s some “slop” in how the components fit in their locations (especially the volume control and power jack). The 3.5mm jacks have small plastic alignment pegs. It’s essential these fit into the holes (not sit on top) with the jacks fully down on the board. It’s important all the front panel parts be correctly aligned and not rotated off to either side or sitting up off the board as they may not fit through the panel openings.
  
• R10 & S1 Clearance – R10 is very close to S1 (the power switch). Try to make sure R10 is centered or even towards R11. R10 should not touch the switch.
  
• Optional Gain Resistor Pin Sockets – There’s an optional part in the BOM for allowing the gain resistors (R17, R21, R19, R23) to be replaced without soldering. These “SIP” pin sockets are bit tricky to align correctly and can be installed in pairs.
  
• B2-080 Enclosure Ground – To reduce hum and noise it’s important to ground the enclosure to Pin 1 of the input Jack J2. That’s the center pin closest to the front panel. You can either use a short piece of lead wire from one of the small resistors (you want the thinnest wire possible) and make a loop where the screw hole is for the lower right corner of the front panel. Or, if you prefer, use fine gauge wire (like AWG30 wire wrap wire) from J2’s ground pin to the lower screw hole on the back panel behind the batteries. Run the wire under the board keeping it to the far right of the enclosure (under the batteries). Don’t forget about the wire if you try to slide the board out later. If you have some contact preservative or similar to help keep the aluminum from oxidizing you can apply a tiny bit to the connection but the front panel should keep it in solid contact. Click the picture for a larger version.
  
• B3-080 or Other Enclosure Ground – If you’re using panel mount RCA input jacks on a desktop amp use the nut and washer of one of the jacks as the ground point. Make sure to clean the inside of the panel where the nut fastens (the anodizing or coating on the panels is an insulator). Also scrape off some anodizing around the screw areas on the enclosure and panel to make sure the two make good electrical contact when the screws are tightened. If you’re not using RCA jacks, use the same method as above for the B2-080 enclosure. The enclosure should only be grounded at one point and only to the input ground.
  
• Off-Board Panel Mounted Components – If you plan to eventually remote mount anything like the volume control, LED, etc. plan ahead and don’t solder it into the board as parts can be difficult to de-solder. If you want to use remote input jacks you have to either leave the 3.5mm jack (J2) off the board or cut the two thin ground traces on the inner pins (see diagram). Otherwise the on-board jack shorts out the input when nothing is plugged in for lower noise. The remote input jack can be wired to P1 and a 1/4” remote headphone jack can be connected to P2—ideally with 4 wires. For P1 ground is the center pad, for P2 the two pads closest to the edge of the board are ground.
  
• IC Sockets – They’re optional but highly recommended. They allow powering up the amp the first time without the ICs installed for testing and easy op amp replacement should something go wrong. The measurements were all taken with the tin sockets specified in the datasheet. There are also upgraded gold sockets listed. The op amps in the O2 are very stable with sockets so there’s no reason not to use them. If for some reason you don’t use IC sockets, leave U1 – U4 off the board initially—see Initial Testing below. Don’t try to “wash” the board with flux remover once the IC sockets are installed. 
  
• IC Prep – Bend the “sides” of the DIP8 ICs gently against your work surface to straighten the pins before trying to put them in the board or sockets. IC pins come from the factory “splayed” out and you may damage the pins if you don’t straighten them first.
  
• Component Polarities – Installing a component backwards can destroy it and possibly damage other components when you first turn on the O2. Note the electrolytic capacitors “face” different directions. The large strip is negative and should match up with the mark on the circle on the board. The large caps have the negative stripe on the right and the small caps (C8, C9) on the left. The diodes also face differently. Make sure you get the op amps and comparator in correctly. Sockets will cover up the outline on the board. There’s a dot near pin 1 on the ICs and the board. The electrolytics are the only polarized caps. All the others can be installed in either direction as can the resistors.
  
• Holding Components While Soldering – It can be useful to either use your solder dispenser/spool or otherwise have a piece of solder be self supporting to free up one of your hands. That way you can hold the board and a component fully in place with one hand, hold the soldering iron with the other, and bring the pin/pad to the solder instead of the solder to the pad. A small vice or other way to securely hold the board vertical is also very useful as a “third hand”. Many components will get too hot too touch during soldering so it can be useful to have a small cloth or similar to keep from burning yourself. The resistors and diodes can be held in place by bending their leads out at an angle before soldering them. If any are not flat to the board, it’s easy to reheat one side and push them down later.
  
• Stuffing & Soldering Order – Generally it’s much easier if you install the lowest parts on the board first (the resistors and diodes) followed by the next taller (IC sockets), etc. This is especially true for a fairly crowded board like the O2. If you solder the tall parts in first your fingers won’t fit between the parts.
  
• Flux Removal (optional) – You can leave the flux residue around the solder connections. If you use the recommended solder it doesn’t hurt anything. If you plan to “wash” the board with flux remover, leave off the input/output jacks, switches, the pot, and the IC/pin sockets. Otherwise you’ll wash flux residue and other gunk into some or all of the above and likely cause problems.
  
• C8 & C9 Soldering – The inner pins of C8 and C9 form the star ground so there’s a lot of copper hanging off the pads. The copper acts as a heatsink and it will take more heat and longer to properly solder them. You might also want to flow extra solder between them along the exposed track.
  
• Feed-through Vias – There are a few holes in the board that don’t correspond to any components. They’re used for routing signals from the bottom layer to the top layer. Nearly all of these carry low currents except for the two under C9. Ideally you should add a bit of solder to those two when you’re soldering the board. 
  
• Checklist – It’s best to print out the board drawing showing the location of all the components as well as the Bill of Materials and mark each part on both as you install them. It’s much harder to de-solder something then just double-check the location and value of each part as you go. Your amp also stands a much better chance of working correctly. I build prototypes often and even I made a couple of mistakes with some of the O2 builds. Without professional de-soldering equipment some parts are very hard to remove without destroying them and/or harming the board.
  
• Mistakes – If you do make a mistake, and the solderwick or, if you have one, solder pump isn’t working to let you remove the part it’s often best to carefully destroy the part by cutting it off the board one pin at a time so you can de-solder the pins individually. If you keep trying to take it out in one piece you may damage the PC board itself which is much worse that just having to buy a new IC, etc.
  
• Output Impedance & Resistors - The default output impedance is near zero at 0.5 ohms. For those rare headphones that sound better with a higher output impedance, R10, R11, R15 and R18 can be increased but must always be equal. The output impedance is half their value. So, for example, for a 10 ohm output impedance all four should be 20 ohm resistors. Also, for the Low Power version you may want to increase these resistors to around 6.8 ohms each to reduce DC offset related power drain.
  
• Contact With Enclosure – With no batteries installed, and the AC power disconnected, make sure none of the connections on the bottom of the board, and U5, U6, Q1 and Q2 on the top of the board no contact with the inside of the case no matter how you move the board around in the lowest slot. The bottom of the enclosure has the 2 deep grooves on the outside. The edge pin on J2 has to be trimmed very close to the board to prevent contact. The front panel will only work with the board properly oriented.
  
• Test First! – See the Testing Section and follow the steps there before you complete the board and attempt to listen to it.

GAIN SETTINGS: If you’re considering modifying the O2’s gain, keep the following in mind:

• Gain Explained – See: All About Gain.
  
• Maximum Input Level - Be aware the maximum input level changes with the gain settings (it's not a problem for at 2.5X or less for nearly any source). See Maximum Input.

THE GAIN RESISTORS: Before you solder in the four gain resistors by the gain switch, you might want to consider different gains than the approximately 2.5X and 6.5X default values. You want just enough gain so typical music plays loudly enough with your headphones and source and not much more. Extra gain means using less of the volume control’s range, more noise, more distortion, and makes accidental headphone damage more likely. Here’s what you need to know about calculating gain:

• Lower Is Safer - Lower gain settings make the amp less likely to damage headphones by limiting the maximum output voltage to only approximately what’s needed.
  
• Lower is Cleaner – As shown in the first article, there’s a slight increase in distortion, especially at high frequencies at higher gain settings. Lower gains also result in lower noise.
  
• Resistor Values – The O2’s gain (for one channel) is:
       Low Gain Ratio = 1 +  R16/R17
       High Gain Ratio = 1 + R16/R19
       Voltage Gain in dB = 20 * Log(Gain Ratio)
  
• Example – The standard amp has R16=1500 and R17= 1K so 1 + 1500/1000 = 2.5X. And 20*Log(2.5) = ~ 8 dB.
  
• Feedback Impedance – For many reasons it’s generally best to leave R16 and R22 at 1500 ohms unless U1 is replaced with a weak op amp that can’t handle a 1.5K load.
  
• See The BOM - Gain resistors are pre-calculated in the BOM parts list for gains of 2X – 12X. For 1X gain just leave out the gain resistors (or clip one end if they’re already installed).

NJM4556 OP AMPS: The NJM4556 op amps, by far, work the best in this design. No other dual DIP8 op amp can deliver even close to as much current at as low of distortion. Few op amps are made for driving headphones but the NJM4556 is. It also works well paralleled and not all op amps do. The TLC2062 is an acceptable substitute for headphones 32 ohms or higher and also provides about 3 times longer battery life. See: Low Power Option. The RC4580 is a very distant third if have no other choice.

NJM2068/NE5532: If you’re mainly going to run the O2 from battery I strongly recommend the NJM2068 as it reduces the overall idle power consumption about 13% significantly helping battery life. At 2.5X or 3X gain there’s no difference in distortion performance between the NJM2068 and any other op amp I tried and it’s generally quieter. But, at the 6.5X gain setting, you can get a slight reduction in high frequency distortion from about 0.0018% to 0.0010% switching to the NE5532. Anything below 0.01% is very likely inaudible and both are well under that threshold. But the NE5532 is 3 dB noisier and shortens battery life. Anything more expensive than the NJM5532 is not only a waste of money but some very fast op amps may have stability problems, require different compensation, etc. See the Op Amp Measurements article.

AC ONLY DESKTOP VERSION: There’s a box on the schematic around the parts you can leave out for the AC only version. Don’t remove the power management circuitry as it’s still required to protect headphones from power on/off transients.

BATTERY ONLY VERSION: There’s a box on the schematic around the parts you can leave out for the battery only version. This is mainly to save money but if you use the amp a lot disposable batteries will eventually be more expensive so I suggest the full version unless you use an external charger,

NO GAIN SWITCH: You can jumper around the gain switch and leave out two of the gain resistors if you want only a fixed gain amp. But testing shows the gain switch does not hurt the performance and it adds a lot of versatility for future sources and headphones. There are jumper marks on the PC board along the sides of S2 showing where to install jumpers if you leave the switch out.

 

---

Enclosure Options

---

BOX ENCLOSURES B2-080 PORTABLE/DESKTOP AMP: The $11 extruded aluminum B2-080 enclosure comes in various colors with matching front and rear panels. It’s pictured in this article (for more pics see: O2 Summary) and is suitable for portable and desktop use. The PC board simply slips into the lowest set of internal grooves (the “V” shaped notches on the outside of the case are the bottom—it’s not symmetrical).

BOX ENCLOSURES B3-080 DESKTOP AMP: The B3-080  is a $12 slightly taller cousin to the B2-080 above. It  has room to add things like 1/4” headphone jacks, RCA inputs, etc. but makes the amp less portable. Again, the O2 board slips right into the lowest slots leaving room above it for added jacks. It’s also possible to use both the front and back panels. The O2 board can be installed “backwards” with everything at the back. But consider waiting for the desktop version of the O2 hopefully coming this fall. See: O2 Summary

PRE-MADE FRONT PANEL: Most will probably want to get a pre-machined front panel from Front Panel Express (FPE), their European parent Schaeffer AG, or elsewhere. For as little as $17 you can get a front panel already made without any lettering. Each character adds about ten cents. FPE engraves the characters into the aluminum. On a natural silver panel they’re barely visible but, for about $7 more, they can “infill” the engraving with a contrasting color. Or, if you use a different colored panel (which is only about $1 more), the silver engraved letters will show up without the infill process. The free software can be downloaded from FPE and you can start with the files I have provided. Just make sure you don’t move any of the openings around. Lettering can be added in different fonts, sizes and locations. For the bigger B3-080 you can also add extra openings for added jacks, etc. but you’re on your own. The software is very easy to use and there’s help available on the FPE website.

PANEL COLORS: As discussed above FPE has several colors available (which more or less match the choices from Box Enclosures). Please note all the machined areas, including the outer edge of the panel, will be natural aluminum colored (silver) regardless of what panel color you choose. To me this looks a bit odd but others like the effect. My personal taste is a black box, natural aluminum panel to hide the machining, and black infilled lettering but that’s also relatively expensive.

DIY FRONT PANEL: Please note: The front panel cannot be more than 2mm thick. If it is the connectors will not fully seat in the jacks and the volume knob may rub on the front panel. Someone skilled with metal or plastic work and access to a drill press could probably drill the required holes in the panel that comes with the enclosure. The front panel doesn’t have to be metal although ideally the rear panel should be and it’s still important to ground the metal shell to the input ground (see: Circuit Board Construction). One trick is to print the front panel to scale using the free Front Panel Express software, verify the scale is correct on the printed output (you can adjust it in the software if need be), and then tape a paper cutout of the printed panel to the metal panel as a drill guide.  If you end up ruining the panel that comes with the enclosure you can always still buy the pre-machined one from FPE. Just make sure to use a drill press (not a hand drill), take the appropriate safety precautions, and properly clamp the panel. Otherwise it can turn into a high speed body slicing projectile or crude finger hacking saw blade on the end of your drill.

DIY CUSTOM ENCLOSURE: The “X” and “Y” coordinates of the required openings to match up with the PC board are on the supplied drawing. The numbers are given in “mils” which is the USA standard for PC board mechanical dimensions. Just divide by 1000 to get inches. X is the distance from the center of the board (the LED). H is the height of the hole center above the top of the PC board. The board, if made to the specs, should be 62 mils (1.6mm) thick. So you have to add 62 mils to the H values if you want to reference everything to the bottom of the board. D is the diameter of the required hole. The diameters are either already standard drill sizes or can be “upsized” to the next larger size available. Upsizing the holes reduces the need to be 100% accurate but the end result won’t look as nice. It might be useful to still use FPE’s free software to view the supplied CAD files. You can set the reference anywhere you like for your particular design and then just read the coordinates of all the openings in the software.

 

---

Low Power Option

---

TRADE OFFS: If 7 – 9 hours isn’t enough battery life, you can get higher capacity batteries to improve that by a few hours. And if that’s not enough, you can greatly lower the idle power of the O2. Nothing comes for free, however, and the lower power version has some significant trade offs (photo: R. Dalton)

• Longer Run Time – The low power O2’s idle current drops below 8 mA compared to 22 mA with the normal version. Playing music softly into sensitive headphones this gives about three times the battery life. Because driving the headphones requires a bit of power too, a realistic range is 20 – 30 hours (or less if you’re rockin’ some hungry cans long and hard).
  
• More Distortion – Low power op amps don’t perform as well. While the ones chosen here will not slew rate limit, despite their much higher price, they can’t match the performance of the higher power devices. See the graphs in the Circuit Description section.
  
• Low Impedance Problems – The low power version performs very well into 150+ ohms and respectably well into 50+ ohms, but below that it starts to behave more like the Mini3. I would consider 32 ohms the absolute minimum. See the Circuit Description section.
  
• More Noise – I doubt this will be a real world issue for most, but the measured noise is somewhat higher—another side effect of low power op amps. With my ultra sensitive UE SuperFi Pros I can just barely hear some hiss at full volume and I do mean barely. Any other headphones I own are dead silent
  
• More Expensive – The low power op amps cost about $12 while the standard parts are less than $2. 
  
• Low Power Resistor Changes – Due to DC offset differences, power consumption can be lowered further by increasing the value of R10, R11, R15 and R18. I would suggest around 6.8 ohms for the longest battery life. See DC Offset under Circuit Description for more. You might also consider doubling the values of all six gain resistors and cutting the compensation capacitors in half to reduce the loading on the first stage. This will improve distortion performance and reduce power consumption. See Gain Stage under Circuit Description for more.
  
• The Decision – If you’re using 16 ohm headphones, forget this option. At 32 ohms it might be worth considering if you mainly will run the amp from the batteries and can’t or don’t want to charge it very often. At 50 ohms and higher definitely think about spending the extra $10 if triple the battery life is more important than vanishingly low distortion. If the op amps are socketed you can also easily switch back and forth.

 

---

Obtaining The Components

---

NON DIYers: I expect there will eventually be at least a pre-assembled version of the O2 circuit board available for purchase. Stay tuned for more information.

NORTH AMERICAN DIYers: A PDF with all the parts and a BOM spreadsheet are available in the Resources section. Here are the only sources you should need for everything:

• Mouser Electronics - Everything on the circuit board is available from Mouser Electronics. And the critical NJM4556 is not widely available elsewhere at a decent price except from Online Components so you would probably be placing an order with Mouser anyway.
  
• Allied Electronics – Has the best price and reasonable shipping costs on the B2-080 or B3-080 enclosures. But you can also get them directly from Box Enclosures for only slightly more or the Farnell subsidiaries like Newark or Element 14.
  
• Front Panel Express – If you don’t want to try to drill your own front panel (see the Enclosures section) FPE is the source for the panel. You can start with the supplied files (coming soon) and customize the panel color, lettering, etc. before ordering it using their simple free software. There might be some “refer a friend” discount codes available for FPE. It’s also possible commercial vendors could eventually start selling front panels at a better price. See the O2 Resources section for possible group buys.
  
• eBay – Mouser wants nearly $10 each for 9V Ni-MH batteries. You can get two higher capacity Tenergy batteries from All-Battery on eBay for $11 including shipping—a savings of $9. They have both 250 mAH conventional Ni-MH batteries and also a 200 mAH low-self discharge version that has a longer “shelf life” but less run time. Forget the other “premium” Tenergy batteries. The Mouser batteries are 180 mAH. If you want a desktop amp you don’t need any batteries.

GENERAL COMPONENT NOTES: Here are some things to keep in mind if you can’t find a part or are considering upgrades, substitutions, etc:

• Commercial O2 Designs – Please check with me before using other parts in a commercial version of the O2. I’m trying to encourage very similar performance to what I’ve measured.
  
• AC Wall Transformer – There has been some confusion over the wall transformer and it’s a critical component. Most wall transformers put out DC but the O2’s power supply requires at least 13.5 volts AC. The ideal transformer would be rated at 14 - 16 VAC and 400 mA or higher. In North America the Triad WAU12-200 from Mouser is rated at 12 volts but is really about 13.5 VAC with no load, and on normal 120 volt line voltage works fine for anything but full power sine wave testing or driving rare low impedance power hungry cans. If your line voltage is below 117 VAC  or 235 VAC, and/or you plan to drive difficult low impedance headpones (i.e. HiFiMan planars), I would suggest a 14+ VAC transformer at 400+ mA. The best Mouser transformers are the WAU16-400, 412-218054 or WAU16-1000 CUI. But those are more expensive than the WAU12-200. The WAU20-200 also works for higher impedance headphones. At least some European 230 volt input 12 VAC output transformers only measure around 11.5 to 12 volts on normal line voltage and won’t work (especially if designed for halogen lights). You can also change the power jack to a 5.5mm x 2.5mm version if needed to match the plug of different wall transformers.
  
• R9 & R25 – These resistors are now 33K and 1.5M respectively. If you have the old values that’s OK. See Important Details above.
  
• C6 & C7 – These have been the same 0.22 uF film caps used elsewhere in the O2 which have been working fine for hundreds of O2 builds so far. The 7812 regulator does not require any capacitor on the output and most datasheets for the 7912 also at least imply the capacitor is optional. But for the National LM7912, the data sheet says a 1 uF capacitor is required. I suspect this is just National being more conservative than their competitors, but just to be safe, you may want to use 1.0 uF ceramic MLC caps in this location such as the part used for C1 or, even better, the 5mm lead space version specified in the latest BOM spreadsheet.
  
• Volume Control Taper – The Alps volume controls are available with two different tapers. I have tested both the “3B” taper (used for the measurements in the first article) and the “15A” taper and I prefer the 15A as it’s more “spread out” more below 70% volume. Which you prefer depends on if you’ll be listening more at high volume or low volume settings. At low volume settings (below 70%) the 15A is better. Click the graph for a larger version (graph courtesy Alps). For most the 15A taper (the default RK09712200MC part in the BOM) is probably the better choice.
  
• Volume Control Options – The board was originally designed only for a 20mm shaft length pot. If you cut the tiny 0.8mm “nub” off the front of the 15mm shaft length pot (it’s a brittle casting and it snaps right off) then the pot can be mounted right up to the edge of the PCB and the shaft length is long enough to work with the specified knobs. The current board has locations for both 15mm and 20mm shaft length pots. Here are the pots that work (drawing and photo courtesy Alps):
  
  - Alps RK097122008T (20mm "D" shaft 3B taper)
  - Alps RK09712200HA (20mm "D" shaft 15A taper)
  - Bourns PTD902-2020F-A103 (20mm long "D" shaft)
  - Bourns PTD902-2020K-A103 (20mm long "star" T16 shaft)
  - Alps RK09712200MY (15mm long "D" shaft 3B taper)
  - Alps RK09712200MC (15mm long shaft 15A taper)
  - Bourns PTD902-2015F-A103 (15mm long "D" shaft)
  - Bourns PTD902-2015K-A103 (15mm long "star" T16 shaft)
  
  NOTE: The “star” T16 shaft pots from Bourns will require a different knob: Eagle 450-BA760
  
• SAFE MOSFET HANDLING – Leave the MOSFETs in their static safe packaging until right before you install them. Try to not touch any of the pins.
  
• R1 & R2 – For reasonably fast but still carefree battery charging use the default 220 ohms. If the amp will be mostly used on AC power 240 or 270 ohms are better values. These increase charging time but lower the trickle charge current helping the batteries last longer if the amp is left plugged in all the time. If you will use the O2 mostly from battery and will not leave it connected to AC power for extended periods use 150 ohms for the fastest charging (especially recommended for 270 mAH or 300 mAH batteries).
  
• 3.5mm Jack Colors – These come in different colors. You might want to use the PC standard colors of Green (or Black) for output and Blue for input. See the BOM.
  
• Op Amps – Those who either plan to mainly use AC power or don’t care much about battery life can upgrade the NJM2068 to an NJM5532 for a tiny reduction in the already  inaudible high frequency distortion at the expense of about 3 dB more noise. But, otherwise, don’t try other op amps. Any DIP8 op amp will perform worse than the NJM4556 in the output stage, and many may require circuit changes to replace the NJM2068/5532. If Mouser is out of the NJM4556 in both the D and DD versions try Online Components, or worst case, eBay. The low power TLE2062 can also be used (see the Low Power section). Worst case, the RC4580 is a very distant third choice.
  
• BOM Notes Column - Check the “Notes” column in the BOM if you have any problems or questions. There are a lot of alternatives and tips there.
  
• Physical Size – Because the O2 is a small portable amp with a fairly crowded circuit board there’s not a lot of extra room. So make sure any substitute components not on the BOM will fit, have the right lead spacing, are not too tall (>18mm), etc.
  
• MOSFETS (Q1 & Q2) – TO-220 parts must be in the insulated “F” package and all parts must have a Vgsmax of 25 volts or greater which is somewhat rare. See the BOM and Circuit Description for more.
  
• Battery Terminals – One or both of the Eagle battery terminals might be out of stock. I haven’t yet tested them, but the final PCB revision has been modified to accommodate Keystone 593 and 594 terminals which Mouser has plenty of and are also available from other distributors.
  
• LED – Note the LED isn’t just an indicator, it’s a functional part of the power monitor circuit. So if you change the LED it changes the forward voltage and hence the shutdown voltage. Also note it’s a high efficiency type to operate on very little current to prolong battery life. See LED under Circuit Description.
  
• Power Supply Regulators – The On Semi regulators have lower noise and are the default parts in the BOM. The Fairchild and National parts may be cheaper and will also work.
  
• Low Power Version – This involves different op amps and possible some different resistors and capacitors as well.  See: Low Power. You can always get both the regular and low power op amps and swap them in IC sockets. You may also want to change the gain resistors and compensation capacitors and the output isolation resistors.
  
• Capacitors – Don’t try to change the values or types of the electrolytic caps. The ones in the parts list were chosen with real world testing. Bigger won’t be better and may create other problems. C8 and C9 are low ESR types and should not be greater than 220 uF. C2-C5 are less critical so long as they fit the board and are 470uf and >= 35V. The 0.22 uF decoupling caps could probably be 0.1 uF but that’s not been tested.
  
• Resistors – I have specified low noise resistors in the most critical parts of the circuit. But due to the two stage design, and low impedances, the resistors are far less critical in the O2 than in many other designs. So just about any 1% metal film resistor is likely to perform as well.
  
• Extra Gain Resistors – Getting all of them is only an extra $2 so it’s worth considering. To order them later you’ll pay many times that in shipping cost.
  
• Volume Knob – DigiKey sells some higher bling all metal knobs in your choice of natural, black, satin or gloss (see the BOM). Mouser, sadly, doesn’t (that I know of). The shipping from DigiKey will cost more than the knob, so if others have better suggestions let me know? The Mouser knob is the one in my pictures and is “rubberized” for a decent feel. But it’s not high bling. Perhaps a “group buy” for knobs and B2-080 cases? Some of the Bourns volume pots (i.e. with 2015K or 2020K in the part number) require a T18 “star” style knob such as the Eagle 450-BA760. Beware some substitute knobs without a set screw may “bottom out” against the front panel.
  
• Colors – The case, front panels, volume knob, input/output jacks, power and gain switch buttons are available in other colors (see the datasheet or catalog pages). The knobs come with different colored pointers. So mix and match to make your O2 unique if you want.
  
• Rubber Caps – Box Enclosures and Allied sell the EC2-BK and EC3-BK soft end caps for the B2 an B3 enclosures. They soften the corners of the amp, hide the machined edge of the FPE panel, hide the screws, and allow the amp to be tossed around a bit more. If you add lettering to the front panel make sure the cap won’t cover it up.

FOR THOSE OUTSIDE NORTH AMERICA: You can get everything without too much hassle. The first stop for those outside the USA should be Mouser Worldwide where you can hopefully find a nearby localized Mouser site which should have most everything Mouser USA does. Another option is TTI Europe which only sells passive components but shares a lot of inventory with Mouser. And Farnell, with excellent worldwide distribution, should have most everything else including the Box Designs enclosures. I added a section to the BOM with Farnell part numbers and notes about other options. See also the O2 Resources section for possible kits or group buys in or near your country. Here are some general notes:

• AC Wall Transformer – For countries with 220 – 240 volt power you need a transformer with a rated output of at least 14 VAC and at least 400 mA. The no load AC voltage must be at least 13.5 and less than 22 volts AC. There are several part numbers listed in the BOM. See the comment under General Component Notes above.
  
• LED – If you use the pre-designed front panel and can’t get the specified LED, you need to bend the specified Farnell conventional wire leaded T1 LED to fit. Note the LED isn’t just an indicator, it’s a functional part of the power monitor circuit. So if you change the forward voltage that changes the shutdown voltage. Also note it’s a high efficiency type to operate on very little current to prolong battery life. See LED under Circuit Description.
  
• NJM4556/2608 – Mouser has the NJM parts. Or they’re on eBay for a modest price with shipping included. Another option if your headphones are 50 ohms or higher is the TLE2062 discussed in the Low Power section and stocked at Farnell. And the NE5532 can be used in place of the NJM2608 but needs about 2 more mA of power and has a few dB more noise. Dalbani might also be a source in Europe for the NJM4556.
  
• Battery Terminals – The Keystone parts in the BOM should be available. Worst case, you can use normal 9 volt battery clips with wire leads and secure the batteries.
  
• 3.5mm JACKS – Mouser/TTI are the best options. Farnell has one part that should fit the board OK, but they have metal threaded bushings and will require changes to the front panel and are not isolated which might create ground loops.
  
• Volume Control – MouseUK/TTI are the best options. Another is RS Components. There’s a Bourns part from Farnell that should fit the alternate board footprint. Some Bourns parts have a “star” shaft that requires a different knob (i.e. Eagle 450-BA7600).
  
• Power Switch – MouseUK/TTI  are the best options. There’s also an Alps switch listed from Farnell that should fit the board but will require changing the front panel opening and needs a different button.
  
• Coupling Capacitors – This one surprised me but I guess capacitors are not a strong suit of Farnell. If you want the correct 2.2 uF parts I tested, get them from Mouser/TTI. Otherwise the only parts I could find at Farnell are 1.5 uF or 1.0 uF which will increase the low frequency roll off from 1.8 hz up to 3 hz or 4 hz respectively which is still plenty low.
  
• Resistors – Any resistors with a body length less than 3.9mm should work fine. Testing has shown only very slight difference at high gains with the “low noise” versions. 
  
• Enclosure – Rapid Online has a great price on the enclosure but it doesn’t come with blank panels or hardware. Farnell stocks the Box Enclosures versions.
  
• Front Panel – Use Schaeffer AG instead of FPE.
  
• Other Distributors – Besides Mouser, Farnell and TTI there’s Rutronik, RS Components, Rapid Online, Future Electronics, Avnet, and Arrow Electronics.

MOUSER UPLOAD: Mouser has the ability to upload a Bill Of Materials (parts list) and that’s by far the easiest way to order the 35+ different parts on the O2’s circuit board and will save a bunch of time and many potential mistakes. Here’s the process:

• Register With Mouser - Like most websites you need to register your email with Mouser if you haven’t purchased from them before.
  
• Log On and Click “Services & Tools”
  
  
• Choose Import Bill of Materials from the menu above and you’ll get a screen like the one below. You should be able to cut and paste from the pink shaded area of the supplied BOM spreadsheet (from Google Docs) into the big empty white box and click Import BOM:
  
  
• Resolve The Conflicts – Mouser sometimes has multiple versions of the same part that come up under the same part number. There will be a few conflicts, especially with the NJM parts as seen below:
  
  
• Click “Select from a list of matching produce” (see above) and you’ll get a screen like the one below, follow the steps and click the big green Buy Selected button:
  
  
  
• After the conflicts are resolved check the entire shopping cart carefully for backorders like the one shown below. If you find any, note the part, find the alternates in the BOM and see if one of those is hopefully in stock and order the required quantity. Delete the backorders from the shopping cart after you find suitable in-stock alternatives:
  
  
  
• Add any extra parts from the rest of the parts list (the blue shaded area) you might want such as the AC wall adapter, extra gain resistors, jacks for the desktop version, etc. (anything that’s not part of the shaded area of the BOM). Double check the final shopping cart against the parts list.
  
• Get your credit card out :)

 

---

Initial DIY Testing

---

IMPORTANT! Any DIY amp can damage headphones and possibly even your source if it’s assembled wrong. So initial testing is important and it’s strongly recommended to follow each of the steps in this section.

DOUBLE CHECK EVERYTHING! I like to print out the drawing of the board and use a highlighter to mark each component as I verify the correct part is installed in that location with the correct polarity/orientation. When every part has been highlighted, flip the board over and check all the solder connections for any “bridges” of solder accidentally connecting pads together that aren’t supposed to be connected. Also check for anything that’s either not soldered at all or poorly soldered. These are the most common problems with DIY projects and can have very unhappy results.

CASE CLEARANCE: Before installing the batteries or connecting any power make sure the board slides into the bottom slots of the case without any contact at J2, U5, U6, Q1, Q2 or elsewhere. See the Circuit Board Construction section.

STEP BY STEP: After you’re pretty darn sure you have everything visually correct, it’s best to take small steps as follows:

• Don’t Connect Anything Yet – Don’t plug anything into the input or output jacks until you’ve completed all the tests below and verified the amp seems to be working normally.
  
• Leave U1 - U4 Off The Board – If you used IC sockets this is easy otherwise you hopefully read the tip earlier and didn’t solder them in yet. As a tip, always use a small flat blade screwdriver or similar to gently pry chips out of their sockets. Don’t try to use your hands, pliers, etc.
  
• Use a DMM - Any sort of Digital Multi Meter (DMM), even a sketchy $5 one (see the tools section), is extremely useful. It’s the best way to know the O2 is safe to plug your source and headphones into. If you don’t have one you can get one on eBay or from CircuitSpecialists and it’s cheap insurance. You can find tutorials on the web on how to use a DMM. It’s not difficult and all you need is one that can measure AC and DC voltage which is pretty much all of them (although measuring resistance—i.e. ohms--is also useful). If you don’t have a DMM you can still at least partially complete some of the steps but it’s crucial to use a pair of junk headphones you don’t care about for the initial testing (see Sacrificial Headphones below).
  
• Measure Resistances – If you have a DMM, with only U2 in its socket (U1, U3 and U4 sockets should be empty), measure across each of the resistors shown in the diagram to the right (click for larger version). The values marked with a “*” may depend on your DMM. If you get a very different reading, try reversing the leads. If it’s still different, check that resistor carefully. If in doubt, heat one end of the resistor from the top of the board and use a small screw driver to carefully pry that end up from the board so it’s only connected at one end. You can then measure it accurately “out of circuit”. If it measures OK, persuade the lead back into board using needle-nose pliars while you heat the pad and lead.
  
• CAUTION: Use the resistance (ohms) range of the DMM only with all power removed from the board and after waiting at least 15 seconds with the power turned ON (in) after the power is removed to let the capacitors discharge. Some inexpensive DMMs can be damaged by applying a large voltage when they’re trying to measure resistance. Also, beware up to 64 volts DC can be across the unregulated parts of the power supply (i.e. C2/C3/etc.). That’s high enough to be potentially hazardous. Hang onto the insulated parts of the probes and edges of the board. Try not to “ground” yourself.
  
• Measure Your AC Wall Transformer – If you have a DMM use it to measure the AC voltage at the coaxial power plug. It should be at least 13.5 VAC and under 22.5 VAC. If it’s above that range you need a different wall transformer. If it’s between 13 and 13.5 VAC you might be OK unless you plan to use power hungry low impedance headphones. If it’s less than 13 VAC or over 22.5 VAC you need a different wall transformer.
  
• Remove U1 – U4 (if socketed) – If you installed U2 for the resistance checks above, carefully pry it out with a small screw driver from each end. U1, U3, and U4 should also not be installed.
  
• Use The AC Wall Transformer Initially – It might not be intuitive, but fully charged Ni-MH or alkaline batteries can deliver more current into a dead short than the on board power supply can. The on board power supply, and wall transformer, will likely survive some abuse and the batteries won’t. If your O2 is a battery-only version you’re stuck using batteries. Consider getting some cheap carbon-zinc 9 volt disposable batteries for the initial testing. They’re inherently current limited. You could also use a dual bench supply (or two single supplies) with the current limit set for 0.1 A or less connected to the battery clips.
  
• Briefly Check The Supply Voltages – Before you power the O2 on for the first time hook up the DMM across the outermost of the four battery terminals (the + terminal of BT1 and the – terminal of BT2). With the power switch S1 off (the out position), briefly connect the AC adapter. The DMM should read around 23 - 24 volts (or 18 volts with batteries). If it doesn’t, something is wrong so remove the power plug (or batteries) immediately. If it does read 24 volts, briefly turn the amp on with the power switch and repeat the test. It should still read 24 volts and the LED should come on and nothing should get even warm let alone hot.
  
• Install Only U2 Next – With the power off, install U2 with pin 1 towards the batteries and repeat the above test with the power switch on. Be careful not to short anything out with the probe tips! If the battery terminals measure OK, then measure the voltage from the negative battery terminal of BT1 (the terminal closest to the gain switch which is ground) to pin 4 of the empty U4 socket (the pin closest to Q2). It should be very close to –11.8 volts. Then measure from the same battery terminal to U4 pin 8 in the opposite corner closest to C2. It should be very close to 11.8 volts. If the voltages at U4 are correct, the power controller is likely enabling both rails. Powering up the amp with only one rail working can create a large amount of potentially headphone damaging DC on the output so it’s important to make sure both are working. Again, nothing should get even warm.
  
• Install The Op Amps – Unplug the power and wait for the capacitors to discharge with the amp turned on. Install the three op amp ICs U1, U3 and U4. Turn the amp on again and verify the correct voltages pins 4 and 8 of U4 as above. Now the regulators, U5 and U6, will eventually warm up a bit but everything else should be fairly cool to the touch.
  
• Check The Raw DC Voltages – With amp turned on, no batteries installed, and no headphones, the DC voltages should measure approximately as shown in the diagram to the right (click for larger or right click and open in a new window) with the negative DMM probe connected to the negative terminal of BT1 (ground) as above, It’s important to verify the voltage shown on D3 and D4 are within 0.1 volt of each other but opposite polarities. For example if the banded end of D3 is 17.5 volts then the un-banded end of D4 should be 17.4 to 17.6 volts. If there a greater difference, something is likely wrong. See the Troubleshooting Section.
  
• Check The Output For DC - If everything looks good so far, with the amp on, measure from pin 1 (the square pad) of P2 to the other terminals of P2 as shown in the diagram above. The voltage at the two lower pins should be very close to zero (ideally under 0.008 V or 8 mV). Use the lowest DC voltage range on your DMM if it has manual ranges. As a double check, measure from the center offset pin of J3 to the outer pins at the back corners of J3. The voltage should be under 8 mV at both corners. If it’s much higher something is very likely wrong. It’s OK if both channels are not exactly the same. One channel might be say 3 mV and one 5 mV. That’s normal and due to normal production differences in the op amps.
  
• DC Offset Note – There’s been some confusion about DC offset when the amp is either turned off or in shutdown mode (due to low batteries, etc.). With no headphones or load connected you may measure up to 0.6 volts of DC on the output that will slowly decrease over time. This is completely normal and something most direct coupled amps will do. The offset is from the remaining charge in the power supply capacitors (C8 and C9) after the op amps have completely shut down (transistors need more than 0.6 volts to operate). If you connect headphones, the charge will quickly bleed off and the offset will drop to near zero. With headphones (or a test load) connected the offset should be under 20 mV (0.02 volts) with the amp on, off, or in shutdown mode.
  
• IMPORTANT! Low Voltage Shutdown - To verify the MOSFETs were not damaged by static electricity (ESD), and the entire power management circuit is working correctly, with nothing connected to the amp and running only from battery, pull one battery out and check for DC at the output jack as above. It should be less than 0.7 volts (700 mV). Verify both supply rails shut off by measuring from ground (BT1 neg) to pin 4 and pin 8 of U4. The voltage at each pin should eventually fall below 1 volt. Re-connect the battery and repeat the test by removing the other battery. If either test produces more than 0.7 volts at the output, or one or both supply rails is above 1 volt (or more negative than –1 volt), there’s probably something wrong in the power management circuit. See the Troubleshooting section.
  
• Check The Current Consumption (optional for more advanced DIYers) – If your DMM has a DC current range of at least 200 mA, and you have 9 volt batteries (or suitable bench power supplies), you can verify the amp is drawing the correct amount of power. With the amp powered off, and no AC adapter connected, clip the outer 9 volt battery onto just one clip and connect the DMM’s current jack to the unconnected battery terminal. The common wire from the DMM should go to the unconnected battery terminal on the board. Turn the amp on and the current should settle down to about 20 – 24 mA (or 6 – 8 mA for the Low Power version). If your DMM reads in amps, that’s 0.022 amps. Turn it back off, and repeat the test for the other battery (with the first battery reconnected). The current should be very close to the same.

SACRIFICIAL HEADPHONES: If all the above tests pass, or you don’t have a DMM, it’s time to connect a source and headphones. If you have some junk headphones, like say the ones supplied with a portable player or from an airline flight, use them rather than your $1500 HD800s. That way if something goes wrong and they burst into flames you’re not out much. Hook up a source, set the O2 to Low Gain (switch out), volume all the way down, headphones unplugged, and turn the O2 on. Listening to the headphones plug them in with the O2 powered on but the volume all the way down. There shouldn’t be much noise when you plug them in. If there’s a big pop that’s a sign there’s a DC problem. Also try turning the amp on and off with the headphones plugged in. There should be a modest “click” at power on and a soft “thump” on power off. If there are loud noises instead, that’s a sign of a likely problem. Assuming no huge pops, etc. try playing some music, and if it sounds OK (as OK as can be expected with junk headphones), it’s likely the amp is at least not dangerous. Congratulations!

MORE TESTING: Just like baking cookies, if you follow the recipe exactly, they should turn out great. But if you accidentally leave out the sugar they might look like cookies but they won’t taste very good. If you have an oscilloscope, RMAA set up, or other means of further testing the O2 it’s a good idea to do so before plugging in those HD800s. Even if it sounds OK there could still be some construction errors that might not be obvious. If you don’t have any other test gear, it’s even more essential to make sure you have all the right parts in the right locations, everything is soldered correctly, etc. See the Cautions section regarding sustained sine wave testing at full power if you plan to run full power bench tests.

OTHER PROBLEMS? They’re not likely, but check the Troubleshooting section first, and if that fails, post your question on the diyAudio O2 thread.

ENJOY! Once the O2 clears all the above hurdles it’s time to kick back with some favorite tunes, your favorite cans, and enjoy! Most find a certain magic in listening to something they built themselves. I hope the O2 delivers that magic!

 

---

Troubleshooting Problems

---

O2 PROBLEMS? So you built your O2 and (hopefully) went through the DIY Testing and either discovered a problem there or while listening to it. So far, out of hundreds of O2’s built, only a few have had reported problems. And most of those were fairly easily resolved. Most of the potential problems here haven’t been reported. I’m just trying to be thorough and imagine what possibly could go wrong.

TOP 5 ISSUES: So far, out of the hundreds of O2s built that I know of, only a few have failed the tests outlined in DIY Testing above. And of those few, most of them turned out to be assembly errors of one form or another. Everything so far, including several reports of supposed voltage regulator problems, have been one or more of these “Top 5” with the most common first:

1. Soldering problems including Solder “bridges” (shorts) between adjacent solder pads, “open” connections (not fully soldered), damaged PC board pads/traces (usually from too much heat, trying to remove a component, or excessive physical stress to component leads),
   
2. Components in backwards, including the battery terminals
   
3. The wrong component in one or more location or the wrong wall transformer
   
4. Static (ESD) damage to Q1 and/or Q2
   
5. User modifications that either went wrong or created other problems

CHANNEL BALANCE PROBLEM: If the channels are not equal in level make sure the gain switch, S2, isn’t contacting the via (exposed pad) by R21. See Circuit Board Construction. If that’s OK, make sure R3/R7, R14/R20, R12/R13, R17/R21, R19/R23 and R16/R22 are all matching pairs of values. If those all match, it’s very likely a soldering problem somewhere, or very remotely U1 or S2.

LOUD TURN ON TRANSIENT: A few have reported this problem and it’s so far been Q1 and/or Q2 are damaged. Check the U2 pin 1 and pin 7 voltages outlined in steps 13 and 17 later in this section. And repeat the battery removal test in the DIY Testing section. It’s likely Q1 and/or Q2 are either stuck on or no longer working in sync with each because one is damaged. It’s also possible C16, C21 and/or C1 are missing, not soldered correctly, or the wrong value. Before replacing Q1 or Q2 see Circuit Board Construction regarding static precautions and be extra careful with the new parts.

CLICKING OR RAPID TURN ON/OFF: This indicates the amp is shutting down due to low batteries or some other power supply problem. If you are working from earlier documentation, R9 has been changed to 33K and R25 to 1.5M to help the amp stay off when it shuts down. But these are not critical changes. Either way the amp is no longer usable until the batteries are charged. You can consider the clicking a reminder to turn off the amp.

DISTORTION: If you hear distortion in both channels, make sure the amp isn’t configured for too much gain for your source. See Gain Stage Overload. If you’re certain that’s not the problem, but both channels are distorted, that leaves just U1 and the power supply. Monitor the voltages (see the Step by Step sections below) at D5 and D1 while the amp is distorting. If they both remain very close to 12 volts the power supply isn’t to blame. If only one channel is distorted, try swapping U3 and U4 and see if the problem switches channels. Check pin 1 and pin 7 of U1. Both should be under 0.1 volts. If either is higher, check the pins of P1 with your source connected and turned on. There should not be more than +/- 0.02 volts at any of the P1 pins. If so, there’s a problem with your source. If P1 is OK, but pin 1 and/or 7 of U1 has significant DC, U1 is somehow developing a DC offset and either U1 is bad or there’s a soldering problem with it or all the parts around it.

THERMAL ISSUES (things getting hot): U5 and U6 are the only thing on the board that should more than mildly warm. The higher the voltage of your AC wall transformer, the hotter U5 and U6 will run. U3 and U4 will also get fairly warm if the amp is working hard into challenging headphones. R1 and R2 may get fairly warm if the batteries are especially low and being charged. But that’s it. If anything else is getting warm or hot, there’s a problem. And the above three “pairs” of components should be roughly the same temp—i.e. U5 and U6 should be similar, same for U3/U4 and R1/R2. If one is much hotter than its twin, something is likely wrong.

TROUBLESHOOTING BASICS: First of all, it’s best to take things one step at a time. If there’s something wrong with the amp, and you just randomly start trying things, you might make things worse, damage components, damage headphones, harm the batteries, etc. So you start with the amp turned off, the op amps out of their sockets, etc. The power supply is the “foundation” of the amplifier. If it’s not working right, nothing else will. So any power supply problems have to be resolved first.

REQUIRED: To perform the steps below you need to have the following. If you’re missing any of these it’s much more difficult to try and troubleshoot problems. All of these items should be readily obtainable at modest cost:

• 12 VAC – 20 VAC wall transformer (see Obtaining Components)
  
• Two reasonably fresh disposable 9 volt batteries (ideally the cheap kind—not alkaline)
  
• A basic Digital Multi Meter (DMM) capable of reading DC volts reasonably accurately (see DIY Testing)
  
• Some cheap/free or undesirable headphones you don’t care about
  
• An alligator clip, or alligator clip lead (technically optional but very useful to free up a hand)

TWO KIND OF MEASUREMENTS: There are two kinds of measurements in this troubleshooting section:

• Single Point Voltages (ground referenced) – These measurements are made with the positive (red) DMM probe while the negative (black) DMM probe is connected to ground. On the O2 board, the easiest ground point to use is the negative battery terminal for BT1 right below the gain switch. It’s easiest to use an alligator lead, or clip, to attach the black DMM probe to this terminal. That will free up one of your hands.
  
• Two Point Voltages (measured across two points) – These are measurements made between the two points specified in the instructions. They’re not referenced to ground and generally require both hands so it’s best if the board is held in place somehow.

AVOID PROBE SHORTS! – It’s easy to “slip” with a DMM probe and accidently connect two pins, leads, or solder pads together with the metal probe tip. Such a slip could create new problems by damaging components. Be very careful to touch the probe tip to only one thing at a time. And, as mentioned above, it’s best to have the board secured (even by taping it down) so it can’t slide around while you’re “probing”.

MEASUREMENT TOLERANCES: Unless a range or different tolerance is given, all measurements are OK if they are within +/- 0.2 volts of the reading given. So if the step calls for 11.7 volts anything from 11.5 to 11.9 is OK. If your meter has manual ranges (i.e. 2V, 20V, etc.) use the lowest range possible that won’t overload the meter. Also make sure the probes are making good contact if a reading is not as expected. For resistance measurements, try reversing the leads when measuring resistors in circuit.

IMPORTANT TEST EQUIPMENT CAUTION: Be careful about creating ground loops and connecting external equipment to the O2. One O2 builder tried to use a USB oscilloscope’s built in signal generator to drive the O2 while measuring the output with the same device. One of the op amps in the O2 literally exploded as he apparently created a severe “loop” between the input and output. It’s safest to play a sine wave WAV file on a portable battery powered player if you want to use an Oscilloscope.

DETECTIVE WORK: If you get stuck trying to find why something is wrong, find the parts in question on the schematic and trace out what they’re wired to. Generally work “upstream” from where the problem is found. If, for example, there’s a problem with the output of IC U3, is the input to U3 as expected? With the board unpowered (see caution below) you can use the resistance (ohms) range of the DMM to check many of the resistors. Note the caution about harmful voltages and resistance measurements in the DIY Testing section. A resistance diagram is available in the DIY Testing section.

STEP-BY-STEP DC VOLTAGES: Follow each of these steps in order unless a step instructs otherwise. If you find something that doesn’t measure correctly, you’ve found a problem that needs to be corrected. Always stop there and try to find the problem based on the suggestions. The steps are numbered making it easier to reference them on the diyAudio O2 thread if you need further advice:

1. First of all, make sure you conduct all the tests in DIY Testing up to the point you find a problem. That includes checking the AC wall transformer and resistance values. The next several steps check the power supply including the power control circuit. The voltage checks can be mostly be found on the diagram below (click for larger or right click to open in a new window).
   
2. Anything Getting Hot? - With no batteries connected, and the power switch off (out), connect the AC wall transformer to the O2 board. Note if anything is getting hot, especially U5 and/or U6. Nothing should get hot at this point. If it does with the power switch off disconnect the power and follow the advice in the next step regarding excessive current draw.
   
3. Negative Unregulated Supply - Connect the negative lead of the DMM to the battery terminal ground as described about in Single Point Voltages above. Then touch the positive probe to the non-banded end (closest to C3) of D4 as shown in the diagram. The voltage should be negative 17.5 – 32 volts DC depending on your wall transformer. If it’s in that range go to step 3. If the voltage is over 32 volts, you need a different wall transformer. If it’s much under 17.5 volts, set your DMM for AC volts and measure the banded end of D4 closest to the power switch. It should be the same as you measured in step 1 above. If it’s more than 0.2 less, something is drawing too much current. Make sure U6 and U5 are in correctly with the metal tab lined up with the stripe on the board. Make sure they’re the correct parts in the correct locations. And make sure there are no solder bridges between the very closely spaced pads on the top and bottom of the board. If you’re certain U5 and U6 are OK, is either one more than slightly warm? If so, something after the regulators is drawing too much current. If the power switch is still off (out), it could be a short (solder bridge) or other problem with C6, C7, D1, D2, D5, D6, or the power switch S1.
   
4. Positive Unregulated Supply - Test the banded end of D3 closest to C2. The voltage should be 17.5 – 32 volts. If it’s not, check the same things outlined in the step above.
   
5. Balanced Power - The voltage measured in the last two steps should be within 0.1 volts of being the same (one negative and one positive). If they’re much different, something is wrong and likely drawing too much current from the whichever supply has the lower voltage. Again, assuming the power switch is still off, the problem should be confined to only the circuitry listed above leading up to the power switch (see the schematic in the PDF).
   
6. Regulated Positive Supply – With the power switch still off, check the voltage at the un-banded end of D1 closest to U6. It should be very close to +12 volts. If it’s not, recheck the un-banded end of D4. If that’s still over 17 volts, and U6 isn’t hot, something is wrong with U6 or the solder connections in that area. If U6 is hot, something is drawing too much current. See step 2.
   
7. Regulated Negative Supply – Check the banded end of D5 closest to U6. It should read –12 volts volts. If it’s low, and U5 is not hot, something is wrong with U5 or the solder connections in that area. If U5 is hot, see the suggestions in step 3.
   
8. Power Switch – With U1, U3 and U4 removed (assuming they’re in sockets) turn on the power switch and quickly re-check the diode measurements in steps 5 and 6 above to make sure they haven’t changed. If either or both have dropped below 11.8 volts, there is something in the amp drawing too much current and it’s most likely a solder bridge somewhere. If both the positive and negative rails are low U2 may be installed backwards (and may be permanently damaged if it is), there’s a problem with C15, a short at the power switch, or possibly a problem with C10. If only one supply is low, check C8, C9, C17, C18, R5, R9, R25, and the 3 IC sockets for solder bridges, etc.
   
9. Positive Battery Voltage – Check the “+” terminal of BT1 (closest to the edge of the board). It should be 11.8V and slightly less than what you measured in step 4. If it’s too low, check R1 and D2. If it’s too high check D1.
   
10. Negative Battery Voltage – Check the “-“ terminal of BT2 (closest to U3). It should be –11.8 V and slightly less that what you measured in step 3. If it’s wrong, check D5, D6 and R2.
    
11. U2 Pin 2 (voltage sense) – Carefully, without shorting any pins together, touch the probe to pin 2 of U2. Pin 1 is the lower left corner by the dot on the IC and the mark on the board by D6. Pin 2 is next to pin 1 as shown in the diagram above. The voltage should be –8.4V. If not, verify the side of R5 closest to the batteries is 11.8V. Then measure the right side of R9 closest to C6. It should be –11.8V. If both of those are correct, then either R5 and/or R6 are the wrong resistor(s), or U2 is in backwards, the wrong IC, or damaged.
    
12. U2 Pin 3 (voltage reference) – One more pin over is pin 3 and it should be –10V. If not, is the LED on? If the LED isn’t on, check R6 by the power connector. The end at the edge of the board should be same as pin 3 of U2. The other end should be –11.8V. Make sure R6 and the LED are the correct parts. If pin 3 is close to –11.8 volts, check the LED for a solder bridge or it may be damaged. If pin 3 is a positive voltage, make sure the LED is the correct part and installed correctly. The LED is simply across the power rails right after the power switch with R6 in series to limit the current. The LED has to be working for the power controller to work properly (i.e. –10V at pin 3).
    
13. U2 Pin 1 (Q1 gate pull down) – Should be –11.8. If this is wrong but pins 2 and 3 above are correct U2 may be bad, the wrong part, or not installed correctly. Or there’s a nearby solder problem.
    
14. U2 Pin 4 (negative supply) – Should be –11.8
    
15. U2 Pin 5 (voltage reference) – Should be –10 (same as pin 3)
    
16. U2 Pin 6 (same as pin 1) – Should be –11.8 (same as pin 1)
    
17. U2 Pin 7 (Q2 gate pull down) – Should be around 6 to 11.8 volts (9,8 on my board). If it’s much below 6 positive volts, or worse a negative voltage, but pins 5 and 6 are correct, most likely Q2 has been damaged and the gate is pulling pin 7 too far negative. The other possibility is R8 or U2 (see pin 1).
    
18. U2 Pin 8 (positive supply) – Should be 11.8
    
19. U4 Socket Pin 4 – Should be –11.8. If all the steps above are OK but pin 4 isn’t correct, it’s likely Q2 is damaged, the wrong part, soldered wrong, in backwards, etc.
    
20. U4 socket Pin 8 – should be 11.8. If everything above is correct, but pin 8 is wrong, it’s likely Q1. See the step above.
    
21. Add Op Amps - If everything above checks out, the power supply is at least correctly supplying positive and negative 11.8 volts to the audio circuits. Unplug the AC power but leave the switch on and wait for the LED to go out. Then turn the switch off and insert U1, U3, and U4 paying attention to the correct orientation and being careful not to bend any pins. With no headphones connected, reconnect the AC power and verify the correct voltages at D1 and D5 again. If either is now low, something is drawing far too much current. It could be U1, U3, or U4 are defective, damaged, the wrong part, or installed wrong. Unplug the power and remove U3 and U4 leaving just U1 and U2. Recheck the voltages at D1 and D5. If either voltage is still low the problem is likely associated with U1. If the voltages are OK, repeat the same sequence adding first U3 and then U4 until you find which IC is causing the high current drain. Be sure to wait for the board to power down before adding or removing ICs.
    
22. Re-check Supplies - With all IC’s in place, if D1 and D5 measure normally, the voltages on pins 4 and 8 of U4 should also be OK (see the diagram). If not, check D1 an D5.
    
23. DC Offset - If there’s excessive DC offset at P2 (over 0.008 volts) check pin 1 of U3 and U4. If only one is reading high (or negative) power the amp off and swap U3 and U4 and see if the problem reverses. If so the op amp is bad. If not, look for solder problems around U3 and/or U4. Make sure R12 and R13 are correctly soldered and the correct value.
    
24. DC Conclusion - The above concludes most of the DC voltage checks. If all of the above pass, and there’s no DC offset at P2 (the output) the amp should work OK. It’s at least safe to try with some junk headphones.
    

OSCILLOSCOPE TESTS: First, see Important Test Equipment Caution above. If you have an oscilloscope you can observe the amp trying to reproduce a sine wave. Especially for problems in just one channel, a scope can allow working backwards from the output and comparing the two channels. It will also easily show serious DC offset (assuming a DC coupled scope), clipping, oscillation (assuming it’s a fast enough scope), etc. I don’t suggest running out and buying a scope just to troubleshoot your O2 unless you plan on getting more serious with DIY electronics. You typically have to spend at least around $200 to get something that’s minimally usable.

FINAL THOUGHTS ON TROUBLESHOOTING: If all of the above have failed to solve your problem, here are some additional options:

• Ask for help on the diyAudio O2 thread 
  
• Find someone with more test equipment and/or experience to check out your O2 board
  
• Consider buying an assembled O2 board or completed amp (see O2 Resources)

 

---

Circuit Description (geek stuff)

---

INTRO: This section is optional and intended for those who understand basic electronics and want to know more about how the O2 works. The previous article covered the design process, while this section describes how the O2 works down to the component level. The schematic is available in a high resolution PDF in the Resources section.

AC POWER SUPPLY: The wall adapter provides 13.5 – 20 VAC to J1.  D3 is a half-wave rectifier for the positive power supply and D4 does the same for the negative supply. It’s called half-wave because only half the AC cycle is used to alternately charge each of the two supplies. C2 – C5 filter the rough half-wave DC into reasonably constant DC. As long as the bottom of the AC ripple is significantly above the dropout voltage of the regulators, it behaves roughly the same as a more conventional full-wave supply but doesn’t require a center tapped or dual winding transformer. There are four filter capacitors instead of two to further reduce the ripple, provide lower ESR and higher ripple current capability. A classic 7812 and 7912 (U5 & U6), properly grounded, regulate the filtered DC to +/- 12 volts and remove 99.9% of the ripple in the process. C6 and C7 improve the transient response of the regulators by lowering the output impedance at high frequencies. This allows the regulators to “reject” more noise and anomalies making the entire power supply quieter. D1 and D5 block the battery voltage from feeding back into the regulators which would otherwise drain the batteries even when the amp is off. They are Schottky types and only drop about 0.25 - 0.4 volts drop in this application.

AC WALL TRANSFORMER: The power supply, with a 13.5 VAC (no load) wall transformer, is right on the edge of letting some ripple though under worst case conditions. If the O2 is used with low line voltage, and for sine wave testing, or using very power hungry low impedance headphones, a higher voltage transformer is recommended (14 – 20 VAC). Some 12 VAC transformers and/or a lower AC line voltage may cause the amp to exhibit higher distortion during low impedance sine wave bench testing as the power supply will fall out of regulation and let AC ripple through the regulators. A 14 - 20 VAC transformer solves the problem (like the WAU16-400 in the parts list). In real world use, the WAU12-200 is the least expensive option and works fine at normal line voltages playing music (not sine waves) into 99% of headphones. The On Semi regulators reach their drop out voltage at about 11.5 VAC loaded. If the amp is driving a low impedance load at high current levels, you need at least 12 VAC loaded (200 ma or about a 60 ohm 5 watt resistor if you want to test a transformer). This doesn’t leave much room for using a 12 VAC transformer if the line voltage is low (below about 117 VAC) hence the primary 14 – 20 VAC recommendation.

AC CURRENT CONSUMPTION: While the DC idle current is 20 mA – 24 mA the AC current is higher for several reasons. Two supplies are being generated from a single AC input and the peak charging currents are higher in a half wave supply. There’s also the quiescent current of the regulators, losses in the filter caps, and the battery charging current. Under normal use a 200 mA wall transformer works great. For driving low impedance headphones to high power levels, a 400+ mA transformer is better. Full power worst-case sine wave testing while charging the batteries requires nearly 480 mA RMS of AC current.

BATTERY POWER: The batteries are diode isolated by D2 and D6 so they don’t see the full 12 volt supply rails. These are Schottky diodes to maximize the operating voltage when running on battery power. R1 and R2 are 1 watt power resistors. They’re sized to limit the charge current to well under 1/20 of the battery capacity when the batteries are fully charged (around 8 mA with 220 ohms). This prevents overcharging the batteries. And even with the batteries at zero volts, they dissipate well under 1 watt each and limit the maximum current to a safe value (around 50 mA). R1 and R2 can be anywhere in the range of 150 – 270 ohms depending on battery capacity and the intended use. With high capacity batteries and mostly DC operation, use the lower end of that range for faster charging. For mostly AC power, use the higher end to extend battery life on constant charge.

POWER SWITCH: S1 is a two pole (DPDT) power switch. You need two poles with two batteries. A simple push on/push off type is specified with a round button to make the front panel more DIY friendly.

POWER LED: You might think this doesn’t need a mention but there are some special requriements. First, the normal forward current for most LEDs is 20 mA. That’s as much power as the entire rest of the amplifier needs! So a 20 mA LED would cut the battery life in half. The O2 uses a “HE” high efficiency red LED that is sufficiently visible with only about 0.5 mA. Second, it’s powered symmetrically from the rails (18 – 24 V) on purpose even if that seems less efficient. Otherwise one battery will drain slightly faster than the other. That gets you nothing except mismatched batteries. Finally, the LED’s forward voltage is a critical element of the power management circuit. You can’t change to a different color (especially white or blue) without making other changes as that will require more current and the different forward voltage means the power management circuit has to be altered.

POWER MANAGEMENT: The idea is to shut down the amp when the the 9 volt (8.4 volt nominal) batteries drop to somewhere in the 6 – 7 volt range. U2 is a low power comparator. The circuit prevents potentially damaging headphones with DC. The “A” section compares the total power supply voltage to the LED voltage. The LED in this application has a relatively constant drop as the batteries discharge so it forms a “free” voltage reference without any added power consumption or other components. It’s plenty accurate for this application although if you substitute a different LED it may alter the shutdown voltage. C1 removes noise from the battery voltage and also provides a slight turn on delay. When the voltage at pin 2 is higher than pin 3 the comparator output pulls to the negative rail which turns on power MOSFET Q1 and also causes the output of U2B, operating as an inverter, to go high turning on MOSFET Q2. C16 and C21 provide a controlled turn on reducing the transient “click” at power on. If the battery voltage falls too low, or one battery is disconnected, pin 3 will be higher than pin 2 and Q1 and Q2 rapidly switch off shutting down the amp. R25 provides hysteresis to help prevent the amp from turning right back on again as the now unloaded battery voltage rises. Power consumption in shutdown is under 1 mA.

Q1 AND Q2: These parts are worthy of a special note. Fairchild makes several suitable MOSFETs for both Q1 and Q2 but many are currently out of stock. I’ve provided substitutes in the parts list that are in stock but some are in a TO220F package rather than the desired, and much smaller, IPAK configuration. If TO220 parts are used they must be the insulated variety (the “F” in TO220F) as they will likely touch each other and possibly touch the adjacent metal battery as well. This isn’t a problem with the smaller IPAK versions the board is designed for. It’s also worth noting these parts must have a Vgs(max) spec of at least 25 volts which is less common and restricts the choices.

BYPASSING THE POWER MANAGEMENT: Some might be tempted to bypass the power management circuit entirely and just leave it out for an AC powered desktop O2. This is not a good idea. The power on transient is about ten times greater without the circuit in place. And the on resistance of Q1 and Q2 is so low the drop across them in operation is insignificant (typically < 0.1 V). While the bigger transient may be safe for some headphones, I don’t think it’s worth taking the chance. And if you switch the AC power the transients can be even worse due to the regulators not powering up at exactly the same time. So please don’t blame me if you bypass the MOSFETs and ruin your headphones.

TURN ON TRANSIENT: With the power management circuit the turn on transient measures about 800 mV peak (560 mV RMS) into 600 ohms but is very brief at only about 1.5 mS (equivalent to half of one cycle at 330 hz). So it’s a brief click and within even what an iPod Touch can produce playing music and should be harmless. Here you can see the rails (yellow and green) and the output (purple):

 

TURN OFF TRANSIENT: On power off the transient is well under half the above value at only 350 mV peak (250 mV RMS) but much longer at about 20 mS which is the same as a half cycle at 25 hz creating more of a soft “thump”. Into lower impedance loads it’s even less as the remaining energy in the caps discharges sooner. This is also something you can easily get from an iPod playing music and should be harmless:

 

POWER SUPPLY FILTERING AND DECOUPLING: C8 and C9 are low ESR types and provide additional power supply filtering as well as a local low impedance power reserve for musical peaks. Don’t make these caps larger. They’re the optimum value already. They also form the star ground for the amplifier. C17 and C18 are ground referenced 0.22 uF decoupling (bypass) capacitors for the output stage while C10 serves the input stage and is not ground referenced. This is intentional and was the result of testing with the dScope on the first prototype. It eases the ground routing.

INPUT JACK: A switched 3.5mm jack (J2) is used to ground the inputs when nothing is plugged in. This isn’t strictly required but is handy to show the O2 itself is silent. Most sources have an output impedance of 330 ohms or less. This limits the input Johnson Noise of the O2, in use, to < 600 ohms. But with the input open circuit, the Johnson Noise of the 10K input resistors (R14 and R20) unrealistically dominates the noise performance and there is audible noise at higher volume settings with sensitive headphones. The input is also more prone to EMI-based hum pickup with an open circuit.

OFF BOARD INPUTS: If desired, there’s a 3 pin header (P1) for off board input wiring. If the B3-080 enclosure is used, the input jacks can be located above or below the PCB and the wires can enter P1 from the top or bottom. If off board connectors are used it’s essential to cut the small ground tracks from pins 3 and 4 of J2. Otherwise the inputs will be shorted to ground unless something is plugged into J2. Alternately, you can leave J2 off the board entirely (and delete the hole from the panel). Ideally off board jacks should be isolated from the chassis and only grounded at the P1 header using twisted pair or shielded wiring kept away from the power supply and output stage.

INPUT TERMINATION & FILTERING: R7 and R3 serve multiple purposes and can be any value from 100 ohms to about 330 ohms (normally they can be the same value as the high gain resistors R19 and R23—i.e. 274 ohms). They provide series input current limiting protection for U1 for when the amp is powered off or the inputs are overloaded. They also form an RC filter with C11 and C12 to prevent significant RF from making it to the op amp inputs where it could be demodulated and create excessive DC and noises. And finally they enhance the ESD capability of the amp. R14 and R20 set the input impedance to 10K which, as discussed in the last article, is optimal.

GAIN STAGE: U1 operates as a non-inverting amplifier for each channel. The crosstalk between the sections is better than –110 dB (I measured it) so there’s no benefit to using single op amps. R16 and R22 were chosen to keep the impedances low to reduce Johnson Noise but not excessively load the amplifier increasing THD. This is one area where things get tricky if you’re “op amp rolling” and testing many op amps in a given circuit. Some low power op amps would work better with higher values for R16 and R22 (and the other 4 gain resistors). For the Low Power version of the O2 using the OPA2277 feedback resistors of 1.5K still work but doubling all the gain resistor values (i.e. ~3K for R16/R22, 2K for R17/R21, etc.) to reduce the load on the low power op amp at the expensive of slightly higher noise. Ideally C19 and C20 should be reduced to 100 pF if R16/R22 are 3K to maintaint roughly the same compensation pole frequency.

GAIN SWITCH: S2 switches the ground resistors in the gain stage feedback loop. That way if the switch is briefly open circuit the amp won’t slam into a supply rail from opening the feedback loop. It just drops to 1X gain. All the routing is kept as short as possible and by switching the ground resistors the parasitic values and crosstalk are less of an issue.

GAIN STAGE LIMITATIONS: The overload level of the input stage is directly related to the power supply voltage. When running from AC power it’s very unlikely to be a problem at realistic gain settings. But for portable use you have to pick a reasonable battery voltage to use for the calculations. Looking at the curve below at 0.1C (27 mA for this battery which is very close to the typical consumption of the O2 in use) even at 8 hours the battery voltage is still about 8.6 volts (graph courtesy accupower-usa.com):

The battery finally starts to seriously die at about 9 hours when it hits 8.5 volts. Using that value, the input stage starts to clip at 4.6 V RMS. Here’s it’s shown at 4.55 V RMS with the actual rails to the NJM2068 shown in yellow and green at 8.25V. The exact measurements are shown below the traces and the rails are lower due to the ~0.25V drop across D2 and D6:

The maximum input gain would be 4.5/Vinmax. So for a 0.5 V iPOD LOD you get 4.5/0.5 = 9X. For the 1.4 V RMS AMB gamma USB DAC it’s 4.5/1.4 = 3.2X. Here’s the O2 input stage on AC power:

Above you can see 7.1V RMS. So the max gain is 7/Vinmax. Now the max gain with the 1.4V RMS gamma DAC is 7/1.4 = 5X (14 dB).

EXCESS GAIN: On AC power you can figure out the excess gain available using 20*log( 7/V_110db ). So if your headphones hit 110 dB at 0.5 volts it’s 20*log(7/0.5) = 23 dB. For the HD600/650 it’s 20*log(7/2.3) = 9.6 dB. On battery power it’s 20*log( 4.5/V_110db ) and the HD600 still has 6 dB of excess gain available (half the volume control’s range with the “3B” taper). The HD600 is about at the limit of what most portable amps, like the Mini3, can even drive and the O2 still offers 6 dB of excess gain beyond that running from nearly dead batteries.

GAIN STAGE OP AMP: I dScope tested nearly two dozen op amps in developing the O2 (for the gain and output stages). See: Op Amp Measurements. One of the biggest surprises was a $0.39 part essentially matched the most expensive op amps I tested! And it may surprise some to know it’s made by JRC and isn’t their version of the venerable 5532 (which also performed well but costs more, is noisier, and uses more power). The NJM2068 is dead quiet, has vanishingly low distortion, and uses less power than lots of other parts I tried including the popular OPA2134. At moderate gains and with typical headphone loads the output stage dominates the distortion not the NJM2068. So improving the NJM2068 doesn’t significantly change the total performance. And even measuring just the gain stage the NJM2068 challenges the measurement limits of the dScope in some tests. At higher gains (> 4X) there’s a small reduction in high frequency distortion using the NJM5532 instead of the NJM2068 but it costs more, has 3 dB more noise, and raises the total idle current by 13% to around 25 – 27 mA. For those who are worried about 0.0018% vs 0.0010% at 10 Khz, and don’t care about battery life, the NJM5532 is the better choice. Anything more expensive is a waste of money and might perform worse due to needing different compensation, impedances, power supply treatment, etc.

OP AMP “SOUND”: This is discussed more in Op Amp Myths, the Design Process article, and the first article. In a proper application where the op amp is being used as intended it’s been widely shown op amps that measure well in circuit don’t have a detectable “sound”. The “sound” others claim to hear is either from incorrectly using the op amps (which will show up with the right measurements) or it’s the usual subjective bias of sighted listening—i.e. their brains are deceiving them. The NJM2068 measures extremely well in this application, so I wasn’t surprised when none of us could tell the O2’s sound quality from the Benchmark DAC1’s headphone output in blind listening..

ADDITIONAL COMPENSATION: C19 and C20 work with the 2068’s dominant pole compensation to optimize the stability and transient response. Their value (220 pF) might seem large to may who are used to seeing 22 pF or less in other designs but that’s mostly related to the 1.5K value of R16 and R22. If R16 and R22 are increased to 3K in the lower power version C19 and C20 should be reduced to 100 pF. If the resistors were ten times greater (and they often sadly are) C19 and C20 would be ten times smaller. Compensation is discussed more in the Design Process and Op Amp Measurements articles.

VOLUME CONTROL: U1 directly drives VR1 which is a 10K Alps RK097 audio taper pot designed for volume control duty. C13 and C14 isolate the input DC bias current of U3 and U4 from the volume control eliminating the “rustling” noise you get when changing the volume in many amplifiers like the Mini3 and FiiO E9. If you remote mount the volume control use twisted pair or shielded wiring with the ground(s)/shield(s) terminated at the volume control pads on the PCB. The twisted pairs can be removed from CAT5 wire (leaving them tightly twisted) and work well for such wiring.

VOLUME CONTROL LOCATION & GAIN STRUCTURE: The volume control was intentionally placed “in the middle” between the two stages for some very good reasons. The primary benefit is it dramatically improves the noise performance of the O2. See Section 2-11 in the Design article.

DC BLOCKING: C13 and C14 are high quality 2.2 uF film caps rated for coupling applications. I’ve tested 4 different caps and found no significant difference between them including analyzing the “ differential signal” (or lack thereof) directly across the capacitor during operation as well as more usual tests like THD sweeps. The default Kemet MMK caps are designed for signal coupling and do an excellent job. There’s also a couple Wima part numbers listed for those who are “brand name” conscious and don’t mind spending a bit more. I also tested an Epcos cap that also performed very similarly. The fourth was an expensive audiophile grade Wima MKP polypropylene that would never fit in the case that I tested only for comparison. But, just as Doug Self and others have documented, it didn’t perform any better in this application. All the caps had only a really tiny linear differential signal related to their ESR. Another myth busted and discussed more in the Design Process article.

R12 & R13 TRADE-OFF: R12 and R13 provide DC bias for the output op amps. Their value wants to be large to push the low frequency roll off from the C13 and C14 down close to DC. But their value wants to be small to minimize DC offset at the headphones. So it’s a classic trade off. Larger value film caps also take up more space and are more expensive. So 1 uF caps with 100K would be cheaper but the amp would have over twice the DC offset. I opted to go the more expensive route at around 40K and 2.2 uF. This means – 3 dB at only 1.8 hz and results in a typical DC offset of under 4 mV—both excellent specs.

OUTPUT STAGE: To me, this is the best part of the O2. If you would have asked me if a $0.60 op amp could crank out well over 140 mA of peak current at seriously low levels of distortion into the lowest impedance loads I would have had some serious doubts. For that kind of performance you’re supposed to buy a $24 pair of big TO220 buffers like the TI/Burr Brown BUF634. But those clever folks at JRC had other ideas with the NJM4556 (also called the JRC4556 or NJR4556). They took a better than decent audio op amp and seriously upgraded the output stage. The result works way better than most would ever guess. The trick here was to parallel the two devices in each IC and use a dedicated IC for each channel. This has numerous benefits including twice the peak current capability, better channel isolation from the high currents involved, lower distortion, twice the thermal dissipation and more. More on the dissipation can be found a few paragraphs down but it turns out you really do need two ICs.You can just compare the O2 measurements to the Cmoy measurements and you’ll see the difference (although there’s more going on between those two). There’s also more information about the output stage in the O2 Design Process article starting around section 2-12.

CURRENT LIMITING & HEADPHONE PROTECTION: A welcome bonus with the NJM4556 is the current limiting is about perfect. Two sections in parallel current limit about 20% above the 166 mA max current goal calculated in the Design Process article. This helps prevent accidental headphone damage into low impedance loads compared to amps with current limits more fitting for an arc welder. And some amps without any current limiting blow up if their outputs are even briefly shorted (which happens just plugging a pair of headphones in). Not so with the O2. The NJM4556 seems rather tolerant of at least brief abuse and the current limit might save some low impedance headphones from being destroyed if someone accidentally uses the wrong gain setting, etc. Other amps with higher current limits can deliver more than 1000 mW into low impedance loads which is far in excess of the safe limits specified for most headphones.

PARALLEL OP AMPS: Op amps are paralleled all the time. It can lower their noise under some circumstances and increases their current and power dissipation capability. I’ve seen lower current op amps paralleled for headphone duty, but I’ve never seen the 4556 in parallel. If someone knows of a published design that predates this one, please let me know?

CURRENT SHARING & OUTPUT IMPEDANCE: The two op amp sections on a given silicon die within a single package tend to be fairly well matched. But it’s still a good idea to provide some isolation and series resistance so they will more equally share the load current and not fight each other due to differences in DC offset. This series resistance also improves stability. I did the math, and tried a half dozen different 4556’s, and concluded 1 ohm works well giving an output impedance of around 0.5 ohms. If you don’t mind a slightly higher output impedance, you can go up to 4 ohms if you want even more isolation, a bit more short circuit protection, and less DC offset current (see the next paragraph) and still stay under the “1/8 rule” for 16 ohm headphones. For 32 ohm headphones you could go up to 8 ohms each. See: Output Impedance. Because the NJM4556 op amps have beefy output stages and are operating with maximum feedback their output impedance, without the resistors, is under 0.1 ohms.

DC OFFSET DIFFERENCES & QUIESCENT CURRENT: Because they’re on the same die, the two op amps in each 4556 tend to be fairly well matched for DC offset. But if one is say 2mV and one is 4 mV, that creates (0.004 – 0.002) / 2 ohms = 1 mA DC current between the two amplifiers. Such a low value value of current doesn’t significantly change the distortion performance of the 4556 but it does increase the idle current of the O2 by 1 mA. This is most significant in the Low Power version where it’s about a 15% change. Hence I suggest increasing the isolation resistors (R10, R11, R15, R18) to 6.8 ohms in the lower power version to minimize this current. Because the Low Power version is only recommended for loads of 32 ohms or higher  the output impedance can safely be as high as 4 ohms with no ill effect. 6.8 ohm isolation resistors result in a 3.2 ohm output impedance which is safely below 4 ohms. This will significantly extend battery life if there are significant DC offset differences.

SHORT CIRCUITS: Headphone plugs create a brief short when inserted and removed. That’s why you read everywhere to always turn down the volume before doing either. I’ve tested the O2 for damage with such shorts and I haven’t had any problems. But a sustained short, or one at a very high level, could damage the 4556 op amp. But even the expensive output buffers have cautions on their datasheets about short circuit damage as their thermal protection may not be fast enough to save them. It’s a common trade-off in headphone amp design. Many amps are far more fragile than the O2. But if it’s something you’re worried about, I would suggest increasing R10, R11, R15 and R18 to 1/4 your minimum headphone impedance. So for 300 ohm headphones, use approximately 75 ohm resistors. This will improve the odds the O2 will survive more severe short circuits.

JRC: For those into designer labels JRC deserves some explanation. JRC doesn’t get much respect in audiophile circles. Based on the O2’s performance, however, those audiophiles might be foolish to dismiss this audio-centric company. For what it’s worth, JRC is more focused on analog audio applications than any of the other mainstream suppliers. Even Analog Devices tends to focus more on digital, high speed stuff, A/D, D/A, etc. Some interesting JRC trivia:

• Lower Cost Industry Standard ICs – They make their own versions of a lot of bread and butter analog ICs. For example they have their own version of the most popular audio op amp on the planet—the 5532 originally developed by Signetics. And the NJM2903 low power comparator in the O2 is their less expensive version of the LM2903 from National. In my experience these parts work very similarly to the ones they’re based on but almost always cost less. This might help explain why some audiophiles see JRC as the “WalMart” of chips (lower cost versions). But… keep reading!
  
• Application-Specific ICs – JRC’s strength is offering ICs nobody else offers. A big company, like say Denon, might approach JRC to make a very specific IC that Denon needs in large quantities. Such chips are common in the digital world, but much less so in the analog world. And even better, JRC often adds these application-specific parts to their regular offerings and anyone can buy them.
  
• NJM4556 – I don’t know this for a fact, but rumor has it the 4556 used here was originally an application-specific part designed for, ta da, driving headphones! It’s a unique op amp. Nobody else makes anything much like it. JRC is not the best at marketing, their datasheets are kind of ho-hum (probably partly because of the Japanese-to-English language issues), but the O2 proves their parts, including this one for under $0.60, can work very well and fill critical niche markets.

OUTPUT CIRCUIT & STABILITY: As explained in the last article, the goal was not to use any distortion inducing output inductors or ferrites. The O2 achieves that goal five ways: Dominant pole compensation, additional compensation, a two stage design (see the Design Process article), proper board layout and power supply/ground scheme, and the 1+ ohm output resistors on each of the four op amps. I tested the stability in the first article with several reactive loads including added capacitance without even a trace of ringing or instability.

POWER CONSUMPTION & DISSIPATION: If you do the math (see Burr Brown's SBOA-022 appnote) for the target goals of voltage and current (see the Design Process article) the power dissipation playing music under worst case conditions is around 700 mW per channel. If you tried to use a single IC, like the AD8397, for both channels it would get too hot and likely fail with 1.4 watts total dissipation. But two DIP8 devices can handle 1.6 – 2.0 watts. I've monitored the O2's power consumption versus time using a differential scope connection across a current shunt playing real music with various loads. Worst case clipping music into 15 ohms, the entire amplifier only consumes about 60 mA of average power or about 1.5 watts when running on AC. And it's rare the 4556 ICs rise above 40C case temp. If you subtract what's going into the load and used by other parts of the circuit, each NJM4556 is dissipating less than 700 mW even under the most extreme conditions using music. I've run the O2 at full power into 15 ohms with sine waves for several seconds at a time and haven't noticed any changes in its measured performance let alone had any failures despite the fact the dissipation per channel is (briefly) over 2 watts. I suspect this exceeds the ideal SOA (Safe Operating Area) of the output stages in the chip. That's why there's a warning in the Cautions section about sustained sine wave testing into low impedance loads. The O2 is designed for real world use. If you sufficiently abuse it with test signals into loads below 150 ohms you might damage it.

HIGHER VOLTAGE SUPPLY RAILS: There have been several suggesting higher voltage supply rails for various reasons. Some have dredged up rare discontinued headphones that need more than 7 Vrms to hit 115 dB SPL and others have suggested it as a way to increase the input stage overload point. The problem is in the above paragraph. Output stage dissipation goes up as the square of voltage. Increasing the rails from 12V to 15V increases the 700 mW dissipation with music to about 1100 mW which is more than any DIP-8 (or SOIC-8) device can safely handle unless you like to live dangerously. And please keep in mind if the op amp does fail it will likely dump massive DC into your headphones potentially taking them with it. I don’t believe that’s a risk worth taking. My suggestion for the 0.1% of cases that really need more voltage to simply use another amplifier that has higher dissipation ability.

PCB BOARD LAYOUT: The O2 is a bit cramped and getting a decent layout was challenging. It’s very much function over form and designed to work well not look nice. The tests show it’s reasonably “electrically symmetrical” meaning both channels perform very similarly. The Design Process article went into a lot of things to consider when laying out a PC board. In a perfect world I wouldn’t have warm parts like the regulators near electrolytic caps for example but that only means the caps might last 20 instead of 25 years with typical use. The alternative would have meant compromising the power and ground distribution. The ground reference for the regulators is essentially part of their feedback loop and important to the performance of the power supply and entire amp. Many similar decisions were made with the PCB design including the front panel layout.

LOW POWER VERSION: Part of testing the nearly two dozen op amps was measuring the power consumption of the O2 with the various op amps installed. The datasheets often supply a range of quiescent currents and often at the wrong power supply voltage. Real world conditions (9 volt battery power, loading on the amp, etc.) often change the value. All but ~1 mA of the O2’s 20 - 24 mA idle current is the audio circuitry and 95% of that is the quiescent current of the 6 individual op amps. So focusing on the op amps was clearly the main key to improving battery life. Here’s what I came up with that results in about triple the battery life by dropping total idle current to 6 – 8 mA:

• OPA2277 - This Burr Brown/TI part for the gain stage requires only 0.8 mA per op amp, has low noise and the slew rate is plenty fast. It has a bit more high frequency distortion than the NJM2068 but it’s still low enough for gains of 7X and lower. The main drawback is the price: $4.65 vs $0.39 for the NJM2068.
  
• TLE2062 - This one was tougher. The requirements: High current output (> 40 mA), 24+ volt power supplies, through hole (not surface mount), low noise, low distortion, short circuit protected, and lower quiescent power than the NJM4556. Amazingly, the TI TLE2062 meets all of these requirements with less then 0.4 mA of quiescent current per amp. And they’re only $3 each which isn’t too bad. It’s rated at an impressive 80 mA max current, but as you’ll see, it doesn’t quite manage that at low distortion.
  
• Mix & Match? - You could use just one or the other. But the OPA2277 won’t buy you much more battery life by itself. And the TLE2062s dominate the amp’s distortion performance so using the NJM2068 in the gain stage doesn’t improve things. So they’re much better as a pair or not at all.

LOW POWER DISTORTION & OUTPUT: As mentioned above, there are some trade offs. I don’t recommend the low power version for loads less than about 32 - 50 ohms. The graph below is running from battery power for all tests. The TLE2062 degrades the Low Power (LP) O2 to roughly Mini3 performance into 15 ohms (red vs pink below). Into 150 ohms, however, the LP O2 blows the Mini3 away in max power with 167 mW into 150 ohms vs 38 mW at similar levels of distortion. The standard O2 (light blue bottom trace) is still significantly better. And finally, the brown trace is the LP O2 into 600 ohms showing the OPA2277/TLE2062 pair has respectable performance (0.001% at 1 volt) when the output stage isn’t dominating the overall distortion (And for the eagle-eyed geeks, the LP plots started at 10 mV which is they the look much better at the far left—I’m doing this on all future plots):

 

LOW POWER VERSION NOISE: Low power op amps are usually more noisy, and that’s true here. This result is about 6 dB worse than the default version but still better than almost every other headphone amp I’ve measured. As explained elsewhere this is barely audible at full volume with the UE SF5s. The two stage design helps a lot regardless of the op amps used, and the OPA2277 has very impressive noise specs considering it’s very low power. Here’s the noise referenced to the usual 400 mV at a worst case full volume:

 

---

What’s Next

---

DESKTOP VERSION: As mentioned elsewhere, I’m working on an enhanced desktop-only version of the O2 that I hope to have done before winter. See: O2 Summary

BOTTOM LINE: The O2 isn’t quite as anvil simple as the original Cmoy but it’s not a complex design when you break it down. The only things somewhat unusual are using a half wave power supply, a power controller for safe use of 2 batteries, and the parallel 4556 output stage. It’s solid evidence proper engineering can easily outperform far more expensive “designer” parts and mythical topologies. As I’ve been saying since I started this blog, the implementation matters most.