Sennheiser HD 650

REFERENCE CLASS? Sennheiser bills the HD 650 as “Reference Class”. For anyone writing reviews having solid references is important. That’s partly why I bought a Benchmark DAC 1 Pre. While my Denon AH-D2000s are respected in some circles, they’re not all that well known which reduces their usefulness as a reference. A few months ago I added a much more popular headphone to my collection: The Sennheiser HD 650. After dozens of hours of listening to a wide range of music I’m rather impressed but, I have to admit, they’re a bit of an acquired taste.

CHOICES: I wanted headphones revealing and accurate enough to expose subtle differences among different gear. They also needed to be compatible with typical headphone amps which rules out Stax and other electrostatics. References get more respect when they’re well known and well regarded. That eliminated the more obscure contenders like the Audeze LCD-2. And I wanted something that was not only revealing but I’d also enjoy listening to for hours on end. That ruled out ruthlessly analytical headphones like the AKG K701 and uncomfortable ones like the popular Audio Technicas. I ended up narrowing my choice to the Beyer DT880 and the Sennheiser HD 600/650/800. The DT880s, while quite nice, suffer from some of the same problems as my DT770s. That left the trio from Sennheiser.

THREE FLAGSHIPS: It’s hard to dispute the popularity of Sennheiser’s top headphones but there are endless debates on the internet about the relative merits of the HD 600, HD 650 and HD 800. Chronologically, the HD 600 was originally the top of the class (based on the discontinued but immensely popular HD 580). Then the HD 650 was added as a new flagship. And, most recently, Sennheiser went over the top with the HD 800. To make things more complicated, near as I can tell, not all HD 600s and HD 650s sound the same. Sennheiser has apparently made various production changes along the way. Some even claim HD 600 drivers have been used in HD 650s and visa versa. See the Tech Section for some further evidence. Perhaps the 600s and 650s had more obvious differences in the past, but listening carefully to the current crop, they’re three peas from the same pod. Published measurements back that up.

MAKING A CHOICE: I ultimately chose the HD 650 over the HD 600 because, to my ears, it’s the more pleasant of the two for extended listening with typical source material. And the HD 800, while generally more impressive, it’s hard to justify the roughly three times higher price. The 600 and 650 are already approaching the point of diminishing returns.

BASICS: The HD 650s are full size circumaural (around-the-ear) headphones with an open back. They don’t provide much isolation from the outside world, and the outside world may hear whatever you’re listening to. And the HD 650’s earcups are tilted in a way that makes wearing the headphones “backwards” immediately apparently if you accidentally put them on wrong. You also get a nice hinged permanent storage box with a foam insert.

SENNHEISER PRICING POLICIES: The street price for the HD 650 has gone up with Sennheiser cracking down on dealers excessively discounting their products. They’re now refusing warranty work unless you can prove you purchased from an authorized dealer. And authorized dealers are restricted to what’s known as MAP—a Minimum Advertised Price. The MAP appears to be around $500 while the MSRP (list price) is $650. If you find a price much lower than $450 it’s probably either for reconditioned headphones or it’s not an authorized dealer. If you don’t care about the warranty, there are better deals to be found although even used prices have gone up. Be cautious, however, of possible counterfits—especially for the HD600. Chinese HD600 fakes may also explain some of the larger than life differences some claim between the 600 and 650.

COMFORT: If you read my DT770 review, you know I’m big on headphone comfort. A lot of headphones sit more on the ears rather than over them. And even headphones that sit over the ears can still be uncomfortable because they’re too heavy, have a poor headband design, get sweaty, etc. The 650s are relatively light and have breathable velour covered earpads to combat any sweat problems. Likewise the headband pads seem to work well over the long haul. The 650s, as delivered, had a bit more clamping pressure than I’d prefer. Sennheiser increased the force over the HD 600 supposedly in the interest of better sound. But to my head, after a few hours, they became somewhat uncomfortable. But there’s a fix…

PRESSURE TWEAK: Wanting a bit less head squeezing, I very carefully bent each of the metal portions of the headband (one at a time) just enough to lessen the pressure a bit. And now they’re very comfortable although still not quite a match for the even lighter HD 590s. If you try this, be extremely careful not to put much pressure on the plastic portion of the headband or earcup supports as the plastic can break. The foam in the earpads also seems “break in” and soften making them more comfortable with time. I didn’t notice a huge change in the sound, but if you press lightly while listening to simulate the tighter clamping you can hear some subtle differences. So I’d try the 650s “as is” first. And if they’re uncomfortable, only then consider trying to bend them.

CABLE & PLUG: The HD650 uses a split “Y” style replaceable cable. It’s a 4 wire cable to better isolate the left and right drivers. Other headphones, like the HD590, share a common conductor in a 3 wire cable. Some prefer “side entry” like the 590 and DT770 but that only works well if it’s on the side your source is. The cable on the 650 is notably higher quality, softer, and less microphonic than the thinner and stiffer cable on the 600. If you like to spend money on snake oil (see Subjective Debate) there are several third party cables available. Unlike nearly every other headphone I own, the HD 650 has a 1/4” large plug and comes with a pigtail adapter for use with 3.5mm mini jacks.

AESTHETICS & BUILD QUALITY: I’m not much into bling and tend to emphasize function over form. For example, it’s frustrating when products are compromised in the interest of looking better or different—like the flat keyed “island” keyboard pioneered by Apple that slows down typing. And the new HD 598 with its Pimp My Ride treatment is less comfortable than the older HD 590 it replaces. But, that said, I have to point out the HD 650s don’t look and feel like $500 headphones. They’re almost entirely plastic, the glossy somewhat mottled paint looks tacky, “Sennheiser” is cheaply silkscreened onto the headband in giant letters, and the HD 650 badges are seriously cheesy. Mine were made in Ireland and whoever painted them probably had a few too many pints at lunch. As a point of reference, the AKG K701 and Denon AH-D2000 both cost less but use higher quality materials and seem more high-end.

DRIVER BREAK IN: A few readers challenged the idea of breaking in my Beyer DT770s as being audiophile myth. So with the HD650 I decided to see if I could measure a change. I tested them right out of the box and again after 48 hours of break in. There was a small but measureable change in the bass resonance. I don’t know if it’s audible but at least the concept has some basis in objective reality. See the Tech Section for more.

INITIAL SUBJECTIVE IMPRESSIONS: I thought the HD 650 had a pleasant honesty and openness upon first listen but otherwise it didn’t make a strong first impression. To me, they’re warm, mellow, and rather laid back. It’s neither sterile and analytical like say the best Etymotics, nor exciting and enhanced like the Beyer DT770 or some of the Ultrasones. But there’s more going on than is initially obvious. And, in some ways, I think the HD 650’s main strength lies in what’s missing from what you hear. Here’s the best analogy I can think of…

POLARIZED SUNGLASSES (seriously!): Have you ever compared regular and polarized sunglasses while driving a car or out on a lake? The polarized variety dramatically cut glare and reflections. The best sunglasses remove more of what you don’t want (glare) letting you see more of what you do want. In harsh conditions, they actually enhance your vision. Believe it or not, that describes what I like most about the Sennheiser HD 650. They seem to remove glare, sibilance, harshness and other obvious flaws that are so common in many (especially pop) recordings and somehow reveal more of the music. At first, the lack of harshness can leave music sounding a bit lifeless in comparison to other headphones. But, for me, much of what’s missing usually turns out to be unwanted anyway. And once you’ve adjusted, it translates into an amazing lack of long term listening fatigue. I find myself hearing things I never noticed before.

LONG TERM LISTENING: I’ve spent hours at a time listening to the HD 650s driven from my Benchmark DAC1 Pre on all sorts of music and they’re really easy to live with. After listening to them for an hour, switching to any other headphone sounds just plain wrong. That’s partly human nature and how our brains work. But the interesting thing is the reverse isn’t nearly as true. Listening to the Beyer DT770 for an hour and then switching to the 650, instead of sounding wrong, the 650s might sound a bit dull in comparison but their strengths are immediately apparent. There’s an honesty to the 650’s sound my brain seems to immediately recognize and latch onto. And, after just a few minutes with the 650, I have no desire to go back to the Beyer.

COMPROMISES: Using yet another car analogy, on a brand new glass smooth road even a hardcore sports car can deliver a fairly smooth ride. But on more typical roads that are worn, patched, potholed, etc. a big luxury car delivers a much smoother ride. You can’t have razor sharp handling and an ultra plush ride. It might not be as exciting, but if you want to cover 500 miles of choppy worn freeway, most would consider a good luxury car the more pleasant choice. The Sennheiser HD 650 is the Mercedes S Class of headphones. The big Mercedes makes bad roads seem better than they really are while still delivering an accurate and precise driving experience and the 650 does the same with typical flawed recordings. Few roads are glass smooth and most music is is not very well recorded, mixed and mastered. Both are a great solution to a very common problem. But a Mercedes S Class isn’t a Porsche 911. And, on the right roads, the Porsche can put a bigger smile on a car enthusiast’s face. But stuck in traffic or spending hours on the freeway the Mercedes is the better choice. The same is true here. For the best recordings there are more exciting headphones than the HD 650. But in the real world, with a large variety of typically recorded music, the 650 is hard to beat if you like an honest, relatively accurate sound.

THE FLAWED TUBE ANALOGY: Some have said the 650’s sound is “tube like” but I can only partly agree and it’s a critical distinction. Some tube gear has a similar “warm” sound often caused by the frequency response of the output transformers and/or a higher output impedance. But tube gear also typically has much higher distortion (euphonic or otherwise) and that distortion, if anything, masks details in the music along with the flaws. The 650 does the opposite and isn’t masking anything with added grunge. It has extremely low distortion and reveals more detail in the music.

A HINT OF OBJECTIVITY: I used my failed attempt at a poor man’s fake measurement “head” (a box with an embedded microphone) to level match the headphones using pink noise driving two at a time from essentially identical sources with a near zero output impedance. In short, I tried to level the playing field as much as possible. I listened to the same track on the 650 and the comparison headphone, switching back and forth as necessary, noted the differences, and moved on to the next track and repeated the process for a total of five very different pieces of music. Rinse, lather and repeat for the other headphones and you have at least a relatively fair subjective comparison.

SENNHEISER HD 590 vs HD 650: The HD 590 was my previous open-backed reference headphone. The HD 650 is slightly less efficient needing a few dB more volume. The first few times I swapped back and forth the 590 seemed a bit more “alive” due to a brighter high-end and an increased “presence” through part of  the midrange. The 650 sounded a bit boring in comparison. But focusing on some female vocals I quickly decided the midrange on the 650 is much closer to what a female voice is supposed to sound like. And then I noticed the 590’s brighter high end is mostly added harshness in the high frequencies rather than actual HF content from the recording. The HD590 shares this trait with the Beyer DT770 but it’s less obvious without a proper reference like the 650. The bass is very similar between the two headphones on most recordings. But if there’s really deep bass down low the HD 650 does a better job. And the 650 is more open and spacious—which is saying quite a bit as the 590 is no slouch in that area. The 650 ultimately comes across as more honest, with a vastly better midrange, and significantly less fatiguing. But the 590 is still the most comfortable headphone I own being lighter than the 650. Sennheiser has gone backwards in the comfort department.

DENON AH-D2000 vs HD 650: The Denons are my closed back reference and are significantly more sensitive (efficient) than the HD 650. The D2000 has a more forward and intimate sound than the 650. Instead of sitting 10 rows back, it’s like you’re on stage and the musicians have huddled around you. It’s a different perspective but it’s otherwise tough to fault the Denons when listening to clean recordings. I’d say the Denon’s lower midrange and upper bass is subjectively more accurate than the 650. Like the 590, the Denon is brighter in the highs but it’s a cleaner sort of bright. But it’s still a double edged sword and, in my opinion, the Achilles heel of the Denon. On less than ideal source material the D2000 shines a relentless spotlight on high frequency flaws. The sound is more strident and can become fatiguing much quicker. The lower bass is a bit more exciting on the Denon while the upper bass is more accurate but its overall similar to the 650. So, for me, it mainly comes down to the Denon’s slightly “cooler” midrange, brighter highs, more intimate soundstage and much less forgiving nature. On clean source material they’re both very pleasing headphones in their own ways. For evaluating gear and recordings, I think the Denons are the better choice. But for enjoying typical mass produced music, the Sennheisers get my vote for their ability to smooth over the rough edges. The Denons are also significantly less comfortable because they rest more on, rather than over, the ear with their oddly shaped ear pads. They’re also heavier.

BEYERDYNAMIC DT770 PRO 80 vs HD 650: The sensitivity is close but the HD 650 is slightly more power hungry. Much like with the 590, a quick switch with the DT770 makes the 650 seem flat and boring. The punchy deep bass and “enhanced” highs of the Beyer provides an even greater contrast. And, while there’s some initial appeal to the Beyer’s lively sound, the emphasized highs mean much greater fatigue. The highs suffer from the same problem as the HD590’s only worse. It’s mostly a fake sense of detail and, to my ears, is more distortion than music. The Beyer’s upper midrange isn’t nearly as natural as the 650’s. And when I listen to the 650 for a while and go back to the DT770 its like viewing an oversaturated photo that’s been heavily tweaked in Photoshop. Some might like the enhancements, but I find myself preferring the more realistic original. I will confess, on certain music, I do enjoy the DT770’s pyrotechnic deep bass. And, when driven from a low impedance source, I think the 770’s upper bass and lower midrange are probably more accurate than the 650’s. Subjectively, the 770’s bass emphasis is much lower down the spectrum avoiding the “warm” bias of the HD 650. In terms of soundstage and openness the 770 is at least in the same league but not as transparent. Without the exaggerated thump, sizzle, and grunge, the 650 is easily more revealing. I notice many more subtle details in recordings than with the DT770. The comfort is very similar.

SENNHEISER SIBLINGS: I don’t own the HD 600 or HD 800 but I did get to spend some time with each and compare them to the HD 650. There are probably a million words already written on the 600 vs 650--complicated by the fact Sennheiser has apparently changed both along the way—but here’s my two cents worth along with an HD 800 comparison:

HD 650 vs HD 600: To my ears, the current HD 600 has noticeably less deep bass, a slightly “cooler” balance through the midrange, and a bit more sparkle (and perhaps a bit of harshness) in the highs. But the difference is not nearly as dramatic as many have argued. More significant to me, on less than pristine source material, the HD 650 is more forgiving with significantly less listening fatigue. I also prefer the slight increase in bass although, ideally, I wish the emphasis was confined to the deepest bass more like with the DT770. Despite their more forgiving nature, the 650 doesn’t give up much in the way of resolution or detail over the 600—everything is still there, it’s just presented in a slightly warmer and more flattering way. In terms of pure accuracy, however, the 600 is probably the winner. If I only listened to audiophile recordings I probably would have chosen the cheaper HD 600 instead. The comfort is the same as my headband tweaked 650.

HD 650 vs HD 800: To me, the HD 800’s highs sound more like the the HD 600 while the deep bass sounds more like the HD 650. So, in some ways, it’s the best of both. The 800 is also better made and has a much higher bling-factor for those who care about such things. The 800 seems even more revealing than the 600 while the angled drivers and larger earcup volume contribute to a more spacious sound. For an ultimate reference headphone, the 800 easily wins and I would love to own a pair. But if I have to choose just one for listening to typical real world music, I think the 650 is a better compromise. Comfort is close, but to my head the esoteric 800 is actually a bit less comfortable.

SOURCE REQUIREMENTS: The HD 650 has a high impedance (300 ohms) and needs more voltage than most portable devices and PC headphone outputs can provide. Generally you need about 2 volts RMS (5.7 volts peak-to-peak) for wide dynamic range music at realistic levels. PC and portable audio gear is typically limited to around 1 volt or less with many managing only about 0.5 volts. So you might need a headphone amp. While the 650 may deserve better, even the $20 FiiO E5 can manage about 1.3 volts which might be enough for many tastes. The FiiO E9 can do a great job with the 650 as its 10 ohm output impedance is relatively insignificant compared to 650’s 300 ohms and the E9 has a very healthy 7 volts of output. For more, see my blog articles on amps and impedance.

CAUTIONS:  If you prefer a more exciting and/or a brighter sound, the 650 might not be a great choice. Serious thump and sizzle fans are even more likely to be disappointed. And if you’re looking to reveal flaws (say for mixing work) the 650 is probably too forgiving and polite. The HD 650 needs more than an iPod to play loud enough for most tastes. The Denon D2000 is much more efficient and easier for most gear to drive. Also remember the 650 has a large 1/4 inch plug and requires a bulky adapter to use with portable gear.

MEASUREMENTS: See the Tech Section for an impedance plot and a few other details. But, lacking an artificial head, I don’t have a way to measure full size headphones properly. So I present measurements from HeadRoom instead.

FIRST CLASS:

• Very low listening fatigue
• Makes marginal recordings much more enjoyable
• Detailed and revealing sound
• Relatively accurate (leaning towards warmth)
• Very comfortable

ECONOMY:

• Might be too laid back for some tastes
• Some excitement lost with the best recordings
• Aesthetics and build quality don’t match the price
• Relatively expensive
• May need a dedicated headphone amp

BOTTOM LINE: If a lack of audible fatigue with nearly all music and physical comfort are important, the HD 650 should be on your short list of headphones under $500. If you listen to them, give yourself some time to adjust as it’s easy to confuse a lack of high frequency distortion with a lack of high frequency musical content. Once acclimated, my ears often hear more of the music with the 650 but that wasn’t immediately obvious on the first listen. Their ability to improve the sound of typical recordings is, in my opinion, their greatest strength. But just like a luxury car’s smooth ride comes with a price, the HD 650 is probably not quite as much fun as some other headphones, including the Denon AH-D2000, on pristine source material. I’m ultimately in this hobby for the music. And I don’t want to be restricted to the very limited selection of audiophile-grade recordings to thoroughly enjoy an expensive pair of headphones. So I value the HD 650’s ability to make vast quantities of margin recordings more enjoyable without masking musical details. But as an ultimate reference headphone, I think it falls a bit short.

---

TECH SECTION:

SPECS: The Sennheiser provide specs are:

• Impedance : 300 ohms
• Cable: 3 Meters (10 feet) with 1/4” plug
• Sensitivity/Efficiency: 103 dB at 1 V RMS
• THD: 0.05%
• Weight: 9.2 oz (260 grams)
• Frequency Response: 10 – 39,500 hz (darn, I was hoping for 40,000 hz!)

IMPEDANCE & PHASE ON SIMULATED HEAD: Here’s the impedance and phase on my “simulated head” which is really just a head-sized multi-layer damped box with a microphone embedded in one side. In this case, it’s just closing off the earcup and you can see the bass resonance is around 90 hz. The minimum impedance is a bit above 300 ohms and peaks around 500 ohms. As proof something is going on over the years with Sennheiser making changes, HeadRoom measured a peak of about 450 ohms lower down at 70 hz. Some of the difference could be due to HeadRoom using a true HATS (head) but even wearing the HD 650 on my own head and running the plot it looked a lot like it does here so I suspect there have been changes to the drivers:

BREAK IN: To see if there are measurable differences before and after break-in I made several measurements when the 650 was new right out of the box. And then I ran them at moderately loud volume for 48 hours and repeated the measurements. I went so far as to leave them undisturbed on my “simulated head” the entire time. I also rechecked the measurements in open air and with them on my head and the difference was consistent. The bass resonance shifted slightly lower by a few hertz and that’s what you might expect as the driver’s suspension loosens up a bit. Imagine a new pair of shoes being stiff and after you’ve worn them for a while they break-in and flex easier. That’s roughly what’s likely going on here. That said, the difference in resonance frequency is so slight it’s hard to imagine it’s audible. But it’s possible other characteristics, like distortion, also improve with break-in. Collectively all the changes together could be audible. And it’s very possible other headphones, with different driver designs and suspensions, would show much bigger changes after break-in:

DRIVE REQUIREMENTS: I used an oscilloscope with a very fast update rate to capture peak signal values playing various music while listening to the headphones. As mentioned in the main article above, I think most people would be very satisfied with about 2 volts RMS of drive capability (5.7 volts peak-to-peak). This exceeds what you can get from typical 3.7 volt Li-Ion battery powered gear, or 5 volt USB/PC powered audio devices. The FiiO E5 and E7 use a DC-DC charge pump to generate a negative power supply rail to increase their output but still manage only around 1.3 volts RMS which is likely enough for typical compressed pop music, but might clip on wide dynamic range music like classical or audiophile jazz recordings. The HD 650 ideally deserves an amp with a proper split power supply like the FiiO E9 or even a good dual battery Cmoy. But it will need enough gain to get up to at least 2 volts from whatever your source can manage (typically at least 4X or about 12 dB).

HEADROOM FREQUENCY MEASUREMENTS: The Frequency Plots at HeadRoom show the HD650 has extremely low distortion and very similar frequency response to the HD600 and HD800 and even the Denon D2000. But you can see about 5 dB of enhanced bass compared to the HD600 and a significant dip above 10 Khz. A lot of the bumps and dips from 2 khz to 10 Khz are related to reflections and standing waves in the earcup. In my tests, just moving the headphones around slightly can make a big difference in this range. The brain seems to largely filter these reflections out while the measurement microphone cannot. So it’s best to draw an imaginary line through those peaks and valleys yielding an average trend. If you do that, you can see the Denon D2000 is probably the most accurate of the 4 headphones shown in the link followed by the HD800.

HEADROOM DISTORTION PLOT: Headroom’s Harmonic Distortion Spectrum is a bit odd in how the frequency axis is scaled. But it shows, for a low frequency signal, the distortion is below –90 dB which is an amazing result as that’s about 0.003%! I’ve seen similarly low distortion measurements of the 650 elsewhere. This is lower distortion than many amplifiers and helps bust the myth that headphones always have far more distortion than electronics. Some proponents of tube gear, singled ended amps, and zero feedback amps, like to believe the relatively high distortion of their amp is OK because headphones always distortion even more. But it’s far from true.

HEADROOM SQUARE WAVE: The 500 hz Square Wave Response shows typical overshoot and subsequent ringing but is a fairly decent result. If you draw an averaged line through the ringing, the square wave is fairly flat which implies accurate midrange frequency response. For comparison purposes, see the popular Grado SR 80. The Grado looks much less like a square wave.

AN EQ EXPERIMENT: I tried using highly configurable parametric EQ to tweak my Denon D2000s to be similarly forgiving and I failed. When I filter out equivalent amounts of sibilance and harshness with the Denons, more of the music is gone and the result sounds veiled and artificial. The HD 650, to my ears, somehow works more like polarized sunglasses. They largely remove what I don’t want while revealing more of what’s left. It’s not just a simple frequency response difference. My objective side wants to pin down what’s going on but I don’t have a full explanation besides the HD 650’s lack of distortion.

MEASUREMENT SUMMARY: As with speakers, headphone measurements only tell part of the story. In this case they show you may well need an amp to properly drive the HD 650, they share frequency characteristics with the HD 800/600 and have very low distortion. They also help validate claims Sennheiser has changed the design over the years. And, just barely, I’ve shown a measurable difference before and after break-in. But I can’t fully explain their ability to improve poor recordings with little loss of detail. I might do better if I had a proper artificial head and conducted more acoustic measurements comparing the HD 650 to say the Denon D2000. But, for now, I’m happy to just enjoy the HD 650.