Details

Parent Category: Products

Category: Headphone Measurements

Created: 20 June 2020

Reviewed on: SoundStage! Solo, June 2020

I measured the Focal Utopia headphones using laboratory-grade equipment: a G.R.A.S. Model 43AG ear/cheek simulator/RA0402 ear simulator with KB5000/KB5001 simulated pinnae, and a Clio 10 FW audio analyzer. For isolation measurements, I used a laptop computer running TrueRTA software with an M-Audio MobilePre USB audio interface. The headphones were amplified using a Musical Fidelity V-CAN. These are “flat” measurements; no diffuse-field or free-field compensation curve was employed. If you’d like to learn more about what our measurements mean, click here.

The above chart shows the Utopias’ frequency response. This is pretty standard stuff for audiophile headphones. Most interesting is that little peak at about 1.4kHz, which is unusual, and which will give the Utopias a slight emphasis in the upper midrange.

This chart shows how the Utopias’ tonal balance changes when they’re used with a high-impedance (75 ohms) source, such as a cheap laptop or some cheap professional headphone amps, or some tube amps. There’s a pretty big bass boost with the high-impedance source -- about +3.5dB -- so these will likely sound especially warm when used with an amp that has a tube output stage.

This chart shows the Utopias’ right-channel response compared with the Focal Stellia closed-back headphones, the Meze Empyrean open-back planar-magnetic headphones -- and the AKG K371s, which are said to be the passive headphones that best conform to the Harman curve. You can see that the Utopias fall pretty much in line with general norms of headphone voicings, but they’ll have less bass than Harman-curve headphones do.

If I didn’t know the Utopias use dynamic drivers, I’d swear their spectral-decay plot came from a planar-magnetic headphone, because of all those very high-Q resonances between about 1 and 5kHz. I’ve been told these are comb-filter effects caused by reflections between a ~100mm, flat planar driver and the flat surface of the ear/cheek simulator -- but how does that explain what we see here, where we have a relatively small, 40mm M-shaped diaphragm suspended in an open frame? Someday I will figure this out. Anyway, I’ve found that headphones with these high-Q resonances tend to sound very open and spacious.

The distortion of the Utopias is nearly non-existent even at extremely loud listening levels, rising to audibility only at 20Hz with a signal calibrated for 100dBA with pink noise. That’s louder than I can stand to listen for more than a couple of seconds, yet the Utopias remained unruffled.

In this chart, the external noise level is 85dB SPL, and numbers below that indicate the degree of attenuation of outside sounds. The lower the lines, the better the isolation. The isolation of the Utopias, like that of most open-back headphones, is negligible. I added the Stellias so you could see how a high-end closed-back model performs on this test.

The Utopias have a huge impedance swing centered at 56Hz, where impedance magnitude reaches a whopping 655 ohms, although the impedance is mostly flat, measuring 84.5 ohms at 1kHz. There’s also a big phase swing in the bass, corresponding with the impedance peak. You won’t necessarily hear this, but it does manifest itself in the substantially changing frequency response with high-impedance sources seen above.

Sensitivity of the Utopias, measured between 300Hz and 3kHz, using a 1mW signal calculated for 80 ohms rated impedance, is 105.6dB -- outstanding for high-end headphones, and no problem for almost any source device to drive.

. . . Brent Butterworth
brentb@soundstagenetwork.com

Details

Parent Category: Products

Category: Headphone Measurements

Created: 10 June 2020

Reviewed on: SoundStage! Solo, June 2020

I measured the effects of the Dekoni earpads using a G.R.A.S. Model 43AG ear/cheek simulator (including the RA040X high-resolution ear simulator and KB5000 pinna), a laptop computer running TrueRTA software with an M-Audio MobilePre USB audio interface, and Musical Fidelity V-CAN amplifier. These are “flat” measurements; no diffuse-field or free-field compensation curve was employed.

This chart shows the frequency response of the Audeze LCD-X headphones measured with the stock earpads and the Dekoni Choice Suede pads. The difference is subtle -- the Dekoni pads kick in a few dB more bass below 50Hz, and reduce the LCD-X headphones’ peak at 2.4kHz by about 3dB. This is the only case where my listening notes don’t seem to correspond well with the measurements; I noted a little less bass, but clearly there’s more, and the reduction in output in the 2 to 3kHz band should have created the impression of even more bass. There seems to be something going on psychoacoustically here that I don’t quite grasp.

This chart shows the frequency response of the HiFiMan Sundara headphones with the stock earpads and the Dekoni Elite Sheepskin pads. This is, by far, the biggest difference I measured with any of the Dekoni earpads. With these headphones (and probably most compatible HiFiMan models, which tend to measure fairly similarly), the Dekoni pads tame most of the bass by about 1dB; they also boost it substantially below 30Hz, but there’s so little musical content that low that this shouldn’t have a big impact on the sound. Most importantly, they fill out the midrange between 600Hz and 2.5kHz, but reduce the lower treble between 2 and 4kHz. So it’s a substantially different sound, yet still true to the original vibe of the Sundaras in many ways.

This chart shows the frequency response of the Sony MDR-7506 headphones measured with the stock earpads and the Dekoni Platinum Protein pads. The results aren’t radically different; the important thing is that roughly 3dB reduction in energy between about 150 and 220Hz. This is the upper-bass/lower-midrange region, and too much energy here results in bloated, dull-sounding bass -- which, in my opinion, is the only major sonic flaw with the stock 7506 headphones.

I didn’t test isolation in my review, but I thought it’d be interesting to measure the isolation effects of the Dekoni pads with the Sony MDR-7506 headphones. (The other headphones I used are open-back models, so their isolation is negligible no matter what pads you use.) Except in a one-octave band between 600Hz and 1.2kHz, the Dekoni Platinum Protein pads improve isolation through most of the audioband. From about 2.5 to 13kHz, the isolation with the Dekoni pads improves by about 10dB on average, which means these will be a great choice for those who wear their MDR-7506 headphones on public transit, or who use them for field audio production.

. . . Brent Butterworth
brentb@soundstagenetwork.com

Details

Parent Category: Products

Category: Headphone Measurements

Created: 20 May 2020

Reviewed on: SoundStage! Solo, May 2020

I measured the TMA-2 HD Wireless headphones using laboratory-grade equipment: a G.R.A.S. Model 43AG ear/cheek simulator/RA0402 ear simulator with KB5000/KB5001 simulated pinnae, and a Clio 10 FW audio analyzer. For isolation measurements, I used a laptop computer running TrueRTA software with an M-Audio MobilePre USB audio interface. The headphones were amplified using a Musical Fidelity V-CAN. These are “flat” measurements; no diffuse-field or free-field compensation curve was employed. If you’d like to learn more about what our measurements mean, click here.

The above chart shows the TMA-2 HD Wirelesses’ frequency response with a wired connection. This curve is unusual in that the usual peak centered somewhere between 2kHz and 3kHz has moved up to span the octave from 4kHz to 8kHz, and also the entire treble range is a little low relative to the midrange and bass. There’s also a little extra energy in the upper bass and lower mids, although that’s fairly common.

This measurement shows the difference in frequency response between a wired connection and using the Bluetooth headband. (I used my Samsung Galaxy S10 phone with white noise as a source.) The minor differences you see are the result of the inaccuracy of the noise-based measurement; there’s effectively no difference in response between the two modes.

This chart shows how the TMA-2 HD Wirelesses’ tonal balance changes when they’re used with a high-impedance (75 ohms) source, such as a cheap laptop or some cheap professional headphone amps, or some exotic tube amps. There’s no significant difference in response.

This chart shows the TMA-2 HD Wireless headphones’ right-channel response in wired mode compared with several closed-back headphones, including the AKG K371s, which are the headphones said to come closest to the Harman curve. The TMA-2 HD Wirelesses look a lot like the Drop + THX Panda headphones; they have the same unusual lack of energy at 2kHz, although they have a little less overall treble output. The peak around 150Hz makes them a little heavy in the upper bass.

The TMA-2 HD Wireless headphones’ spectral decay plot shows just one minor resonance at about 1.5kHz, but it’s well-damped.

Distortion is fairly low in the TMA-2s, at least in wired mode. At the very loud level of 90dBA, there’s a bit of distortion in the bass, but nothing noteworthy above 100Hz. At the crazy-loud level of 100dBA, distortion rises to between 1% and 2% above 100Hz, and gets higher in the bass, but I think unless you are listening to hip-hop or EDM cranked insanely loud, you’re not going to notice distortion with these headphones.

In this chart, the external noise level is 85dB SPL, and numbers below that indicate the degree of attenuation of outside sounds. The lower the lines, the better the isolation. The isolation of the TMA-2 HD Wirelesses is comparable to that of competing open-back models. I added the NAD Viso HP70s so you could see how a typical noise-canceling model performs on this test.

The TMA-2 HD Wirelesses’ impedance magnitude is fairly flat, running between 36 and 42 ohms with a resonant peak at 75Hz, and the phase is just about dead-flat.

Sensitivity of the TMA-2 HD Wireless headphones in wired mode, measured between 300Hz and 3kHz, using a 1mW signal calculated for 32 ohms rated impedance, is 102.8dB, which means they’ll play pretty loud straight off a typical smartphone.

. . . Brent Butterworth
brentb@soundstagenetwork.com

Details

Parent Category: Products

Category: Headphone Measurements

Created: 10 May 2020

Reviewed on: SoundStage! Solo, May 2020

I measured the HiFiMan Deva headphones using laboratory-grade equipment: a G.R.A.S. Model 43AG ear/cheek simulator/RA0402 ear simulator with KB5000/KB5001 simulated pinnae, and a Clio 10 FW audio analyzer. For isolation measurements, I used a laptop computer running TrueRTA software with an M-Audio MobilePre USB audio interface. The headphones were amplified using a Musical Fidelity V-CAN. These are “flat” measurements; no diffuse-field or free-field compensation curve was employed. If you’d like to learn more about what our measurements mean, click here.

I also used the Clio 10 FW to run some quick output measurements on the Bluemini. It put out 139mW at 0.5% THD, and 145mW at 1% THD. Even given the Devas’ low sensitivity (see below), that’s enough power to get you to about 110dB with 0.5% THD, which is a very loud listening level.

The above chart shows the Devas’ frequency response with a wired connection. This curve is very typical for open-back planar-magnetic headphones, with essentially flat response from 50Hz to 1.8kHz, and an octave-wide peak centered at about 3.6kHz.

This measurement shows the difference in frequency response between a wired connection and using the Bluemini. (I used pink noise and the Clio’s real-time analyzer function for this measurement, which is why it shows a -3dB/octave low-pass function.) The Bluemini seems to have two-pole high-pass filtering, starting at about -1.5dB/octave below 300Hz, and increasing to about -8dB/octave below 70Hz. Note that this is not a function of Bluetooth; it seems to be intentional tuning performed in the Bluemini’s DSP. I noticed this first using time-windowed log chirp tones, and confirmed it with the pink noise measurements. So the Deva headphones should sound brighter with the Bluemini.

This chart shows how the Devas’ tonal balance changes when they’re used with a high-impedance (75 ohms) source, such as a cheap laptop or some cheap professional headphone amps, or some exotic tube amps. There’s a tiny, probably imperceptible boost of about 1dB at 20Hz with the high-impedance source.

This chart shows the Devas’ right-channel response compared with several open-back planar-magnetic headphones -- and also with the Dan Clark Audio Æon Flow 2 Closed headphones using their perforated earpads, as those are the planar headphones that I’ve found come closest to the Harman curve. Obviously, the Devas are very much in line with typical open-back planars, and they’ll have less bass than Harman curve headphones do.

The Deva headphones’ spectral decay plot looks typical for planar-magnetic headphones, with lots of very high-Q resonances between about 1 and 5kHz. I’m told these are comb-filter effects caused by reflections between the large, flat planar driver and the flat surface of the ear/cheek simulator. However, the Devas have even less bass resonance than I’m used to seeing from planars, which suggests they’ll have nice, tight-sounding bass.

As is usually the case with planar-magnetic headphones, the distortion of the Devas is nearly non-existent even at extremely loud listening levels.

In this chart, the external noise level is 85dB SPL, and numbers below that indicate the degree of attenuation of outside sounds. The lower the lines, the better the isolation. The isolation of the Devas is comparable to that of competing open-back models. I added the Audeze LCD-2 Closed Backs so you could see how a planar-magnetic closed-back model performs on this test.

Like most of the planar-magnetic headphones I measure, the Devas’ impedance magnitude and phase are just about dead-flat; it’s about 18.8 ohms across the audioband.

Sensitivity of the Deva headphones, measured between 300Hz and 3kHz, using a 1mW signal calculated for 18 ohms rated impedance, is 88.6dB, which means you won’t get loud volumes when driving them directly with a smartphone.

. . . Brent Butterworth
brentb@soundstagenetwork.com

Details

Parent Category: Products

Category: Headphone Measurements

Created: 01 May 2020

Reviewed on: SoundStage! Solo, May 2020

I measured the FS-HR280 headphones using laboratory-grade equipment: a G.R.A.S. Model 43AG ear/cheek simulator/RA0402 ear simulator with KB5000/KB5001 simulated pinnae, and a Clio 10 FW audio analyzer. For isolation measurements, I used a laptop computer running TrueRTA software with an M-Audio MobilePre USB audio interface. The headphones were amplified using a Musical Fidelity V-CAN. These are “flat” measurements; no diffuse-field or free-field compensation curve was employed. If you’d like to learn more about what our measurements mean, click here.

The above chart shows the FS-HR280 headphones’ frequency response. This curve is unusual; it would look a lot closer to normal if the peak at 3kHz were 6dB higher and the one at 6kHz were 6dB lower. The balance of bass to treble seems fairly even, but I would speculate the lower midrange will sound overly full, and the upper midrange and lower treble will sound rather lean.

This chart shows how the FS-HR280s’ tonal balance changes when they’re used with a high-impedance (75 ohms) source, such as a cheap laptop or some cheap professional headphone amps, or some exotic tube amps. There’s a boost of about 1dB in the bass and a reduction of about 1dB in the treble.

This chart shows the FS-HR280s’ right-channel response compared with several open-back planar-magnetic headphones -- including the AKG K371s, which are said to be within ±1dB of the Harman curve. This comparison looks weirder because the curves are normalized to 94dB at 500Hz, and the FS-HR280s have a peak right there so their curve is shifted down by a couple of dB. Still, it’s easy to see how much less energy the FS-HR280s have at 3kHz and how much more they have at 6kHz.

The FS-HR280s’ spectral-decay plot looks pretty clean -- there are no significant high-frequency resonances, and the resonance at roughly 650Hz lasts only about 6ms, or about four cycles at that frequency.

Distortion in the FS-HR280s is pretty mild at the loud listening level of 90dBA. At the crazy-loud level of 100dBA, they have a lot of distortion below about 50Hz, but considering that’s such a loud level, and that most music doesn’t have a lot of content below 50Hz, I don’t expect this distortion would intrude on your listening except in very rare cases.

In this chart, the external noise level is 85dB SPL, and numbers below that indicate the degree of attenuation of outside sounds. The lower the lines, the better the isolation. The isolation of the FS-HR280 headphones is comparable to that of similarly sized closed-back models. I added the NAD Viso HP70 headphones so you could see how a model with noise canceling performs in this test.

The FS-HR280 headphones’ impedance curve shows a big peak at 60Hz -- something I used to see often in dynamic-driver headphone measurements, although it’s not so common now. That peak and the big phase swing that accompanies it are the reason the headphones’ response changes somewhat with high-impedance sources.

Sensitivity of the FS-HR280s, measured between 300Hz and 3kHz, using a 1mW signal calculated for 32 ohms rated impedance, is 106.9dB, which means your smartphone won’t have any problem driving them.

. . . Brent Butterworth
brentb@soundstagenetwork.com

Details

Parent Category: Products

Category: Headphone Measurements

Created: 20 April 2020

Reviewed on: SoundStage! Solo, April 2020

I measured the Monoprice Monolith M1570 headphones using laboratory-grade equipment: a G.R.A.S. Model 43AG ear/cheek simulator/RA0402 ear simulator with KB5000/KB5001 simulated pinnae, and a Clio 10 FW audio analyzer. For isolation measurements, I used a laptop computer running TrueRTA software with an M-Audio MobilePre USB audio interface. The headphones were amplified using a Musical Fidelity V-CAN, with an Audio-gd NFB1-AMP used for distortion measurements. Except as noted, these measurements were made with the supplied lambskin earpads. These are “flat” measurements; no diffuse-field or free-field compensation curve was employed. If you’d like to learn more about what our measurements mean, click here.

The above chart shows the M1570s’ frequency response. The flat response below 1kHz is typical for large planar-magnetic headphones, but the largely flat response above 1kHz isn’t. The peak at 3kHz is only about +4dB, where we might normally see something like +10dB.

This measurement reveals the difference in frequency response with the lambskin and velour pads. As I found in my listening, the character of the headphones remains largely the same no matter which pads are used; the difference is that with the velour pads there’s a little more treble and a bit less bass.

This chart shows how the M1570s’ tonal balance changes when they’re used with a high-impedance (75 ohms) source, such as a cheap laptop or some cheap professional headphone amps, or some exotic tube amps. There’s no significant difference, indicating that the tonal balance of the headphones won’t change when you switch source devices (unless one of those source devices a has non-flat response).

This chart shows the M1570s’ right-channel response compared with several planar-magnetic headphones -- including the Dan Clark Audio Æon 2 Closed headphones with perfed earpads, which measure very close to the Harman curve. You can see that the M1570s measure very close to the Audeze LCD-X headphones, which share a somewhat similar form factor, and that the HiFiMan HE6se headphones (which I consider fairly typical of audiophile headphones) have a lot more treble.

It’s common to see many high-Q (narrow) resonances between about 2 and 6kHz with planar-magnetic headphones; these appear to be due to reflections between the large, flat planar driver and the flat cheek plate of the G.R.A.S. Model 43AG ear/cheek simulator. What’s unusual here is that the reflections/resonances extend to lower frequencies than I’m used to seeing (down to about 750Hz), and they’re stronger than usual, some hanging in there at -20dB for more than 10ms. I am still not sure if or how this affects subjective performance -- that’s something I really need to dig into.

As is often the case with planar-magnetic headphones, the total harmonic distortion of the M1570s is inaudible even at extremely loud levels.

In this chart, the external noise level is 85dB SPL, and numbers below that indicate the degree of attenuation of outside sounds. The lower the lines, the better the isolation. The isolation of the M1570s is comparable to that of competing open-back models. I added the Audeze LCD2 Closed-Back headphones so you could see how a closed-back planar-magnetic design fares in this test.

The impedance magnitude of the M1570s is basically flat at about 58 ohms, and the phase response is very flat.

Sensitivity of the M1570 headphones, measured between 300Hz and 3kHz, using a 1mW signal calculated for 60 ohms rated impedance, is 101.1dB, much higher than the rated 96dB. No wonder they sound pretty good plugged straight into a smartphone.

. . . Brent Butterworth
brentb@soundstagenetwork.com

Details

Parent Category: Products

Category: Headphone Measurements

Created: 10 April 2020

Reviewed on: SoundStage! Solo, April 2020

I measured the Æon 2 Closed headphones using laboratory-grade equipment: a G.R.A.S. Model 43AG ear/cheek simulator/RA0402 ear simulator with KB5000/KB5001 simulated pinnae, and a Clio 10 FW audio analyzer. For isolation measurements, I used a laptop computer running TrueRTA software with an M-Audio MobilePre USB audio interface. The headphones were amplified using a Musical Fidelity V-CAN, with an Audio-gd NFB1-AMP used for distortion measurements. Except as noted, the measurements were made with the stock earpads and no filter. These are “flat” measurements; no diffuse-field or free-field compensation curve was employed. If you’d like to learn more about what our measurements mean, click here.

The above chart shows the Æon 2 Closeds’ frequency response with the stock earpads and no filter. What’s unusual about this response is that there’s only a relatively small peak around 2 to 3kHz -- which means the listener may perceive a dip at these frequencies.

This chart shows the effects of the different filters provided with the Æon 2 Closeds, all referenced to 94dB at 500Hz. Even with the Y-axis resolution on this chart doubled to 5dB per division, the differences are subtle -- they’re basically serving as a treble control.

This chart shows the difference in frequency response with the stock earpads and the perforated earpads, referenced to the AKG K371s, the headphones said to be closest to the Harman curve. The perforated pads do, as Dan Clark suggests, get the Æon 2 Closeds fairly close to the Harman curve, at least above 1kHz, and the large increase in energy around 3kHz definitely pushes the headphones much closer to the kind of response I typically measure. Same goes for the extra little bump around 100Hz, although there’s 3 to 7dB less bass than the Harman curve calls for.

This chart shows how the Æon 2 Closed headphones’ tonal balance changes when they’re used with a high-impedance (75 ohms) source, such as a cheap laptop or some cheap professional headphone amps, or some exotic tube amps. With the high-impedance source, the bass appears to rise by about 1.5dB at 20Hz -- not an audible difference with probably 99% of music recordings.

This chart shows the Æon 2 Closeds’ right-channel response with the stock pads and no filter, compared with two planar-magnetic headphones priced in the mid-three figures, as well as with the AKG K371s, the headphones said to be closest to the Harman curve. You can see that the Æon 2 Closeds’ response is flatter overall than the other models.

Well, this is weird. With the 36 planar-magnetic headphones I’ve previously measured, I’ve seen a bunch of hashy-looking, very high-Q (i.e., narrow) resonances between about 1.5 and 5kHz, which I’m told represent reflections between the ear/cheek simulator and the large, flat planar diaphragm. To the best of my memory, the Æon 2 Closeds are the first planar-magnetic headphones I’ve measured that don’t show this effect. (Incidentally, the filters could conceivably have damped the reflections, but this measurement was done with no filters in place.) I repeated the measurement a day later with a fresh setup and got a similar result. There is one strong resonance around 1kHz, but it’s so high-Q that I expect it would be only fleetingly audible. I’m not sure if or how this lack of high-Q upper-midrange/lower-treble resonances would affect the sound, but clearly there’s something different in the way Dan Clark Audio engineers its planar-magnetic headphones.

For planar-magnetic headphones, the total harmonic distortion of the Æon 2 Closeds at loud levels is relatively high, averaging around 3% between 300Hz and 1.3kHz. I repeated this measurement with different amps, and with the headphones’ left channel, and got basically the same result every time; other headphones measured in the same session didn’t show this problem. In headphones and speakers, distortion below about 5% is rarely audible, so I don’t necessarily see (or hear) this as an issue, but it’s more evidence that Dan Clark Audio is doing something different with its drivers.

In this chart, the external noise level is 85dB SPL, and numbers below that indicate the degree of attenuation of outside sounds. The lower the lines, the better the isolation. The isolation of the Æon 2 Closed headphones, whether using the stock or perforated pads, is quite good, and an obvious improvement over open-back models such as the Audeze LCD-1 headphones (also shown). I was a little worried that some isolation might be sacrificed with the perforated pads, but it’s actually a little better than the stock pads between 100 and 800Hz.

The impedance magnitude of the Æon 2 Closed headphones is dead-flat at 12.5 ohms, and the phase response is very flat.

Sensitivity of the Æon 2 Closeds, measured between 300Hz and 3kHz, using a 1mW signal calculated for 13 ohms rated impedance, is 86.7dB. That’s low, so I strongly recommend using an external amp or a high-quality portable music player with these.

. . . Brent Butterworth
brentb@soundstagenetwork.com

Details

Parent Category: Products

Category: Headphone Measurements

Created: 01 April 2020

Reviewed on: SoundStage! Solo, April 2020

I measured the PM-50 headphones using laboratory-grade equipment: a G.R.A.S. Model 43AG ear/cheek simulator/RA0402 ear simulator with KB5000/KB5001 simulated pinnae, and a Clio 10 FW audio analyzer. For isolation measurements, I used a laptop computer running TrueRTA software with an M-Audio MobilePre USB audio interface. The headphones were amplified using a Musical Fidelity V-CAN. These are “flat” measurements; no diffuse-field or free-field compensation curve was employed. If you’d like to learn more about what our measurements mean, click here.

The above chart shows the PM-50s’ frequency response with the rounded pads. This is an unusually flat response, as we’ll see in the comparison graph below.

This chart shows the difference in frequency response with the rounded pads and the squared pads. The squared pads add an average of about 3dB more energy between about 20 and 50Hz, and again between about 1.8 and 4.2kHz, so they’re likely to have a more vivid, exciting sound (which may or may not be a good thing).

This chart shows how the PM-50s’ tonal balance (with the rounded pads) changes when they’re used with a high-impedance (75 ohms) source, such as a cheap laptop or some cheap professional headphone amps, or some exotic tube amps. There’s basically no difference, as is typically the case with planar-magnetic headphones.

This chart shows the PM-50s’ right-channel response with the rounded pads compared with several open-back planar-magnetic headphones priced in the mid-three figures. You can see that the PM-50s have a lot more energy around 1.5kHz than the others, and a lot less energy around 3kHz. I’m surprised that this doesn’t sound weird.

The only major anomaly on in the PM-50s’ spectral-decay (waterfall) chart actually isn’t an anomaly for planar-magnetic headphones -- a bunch of very high-Q (narrow) resonances between about 1.3 and 6kHz; these appear to be due to reflections between the large, flat planar driver and the flat cheek plate of the G.R.A.S. Model 43AG ear/cheek simulator. There’s also a minor, well-damped resonance at around 1.3kHz, which corresponds to the response peak in the same vicinity.

As is often the case with planar-magnetic headphones, the total harmonic distortion of the PM-50s is extremely low.

In this chart, the external noise level is 85dB SPL, and numbers below that indicate the degree of attenuation of outside sounds. The lower the lines, the better the isolation. The isolation of the PM-50s, whether using the rounded or squared pads, is about the same as that of competing open-back models. I added the Audeze LCD-2 Closed-Back headphones so you could see how a closed-back planar-magnetic model fares in this test.

The impedance magnitude of the PM-50s is flat at 32 ohms, and the phase response is very flat.

Sensitivity of the PM-50 headphones, measured between 300Hz and 3kHz using a 1mW signal calculated for 32 ohms rated impedance, is 104.7dB. That’s very high for planar-magnetic headphones, so as stated in the text, these will play plenty loud off a smartphone except perhaps with older recordings that are mastered at low levels and that you want to crank.

. . . Brent Butterworth
brentb@soundstagenetwork.com