[csglinux]

After some discussion with fellow headfi measurement enthusiasts, I thought it would be helpful to send a pair of Etymotic ER2SE dynamic driver IEMs out on a world tour. The aim isn't to produce a set of reviews for these headphones (and I'm not going to give them marks out of 5 here), but rather to provide a reference data point for our mics and couplers so that we can come to some common agreement. Several of us have 711-compliant couplers which still may have their own small unit variations. Others are using tube couplers that might benefit from a calibration file derived from a common reference measurement (at least for a sealed/non-ported IEM).

Here's my logic in choosing the ER2SE with stock small silicone (frost) flange tips as a reference:

1) They should easily fit and seal well in any coupler.
2) The stock silicone triple flange tips will hold the IEM in place reasonably well even without the additional use of mounting putty (though putty is still a good idea because it stops any vibration of the IEM in the coupler).
3) There can be no ambiguities from variations in preparing or rolling foam tips. (Ok, eartip effects turned out to be more subtle than I thought... More on this later...!)
4) They're relatively neutral, with no wild FR swings.
5) They have very low distortion, which makes for an interesting challenge in trying to compare THD measurements.
6) They have a flat impedance curve, so no wild swings in FR, even if higher z-out sources are used to drive them.
7) We got them at a reasonable price, many thanks to @wquiles

I suggest using the stock small-sized (frost) triple flange tips, at least as a starting point. If you use the larger-size tri-flange tips, you'll see a definite change in your measurements. It should be possible to find an insertion depth where the resonance peak sits at (or similar to) those shown in the figures below (see also first spoiler link for a photo). Ideally, to get the best agreement, you'd want to try and match the primary resonance peak that I ended up with (~10.5 kHz). Also, note that the left channel is unit number 002527; the right is unit number 002528. The ER2SE uses MMXC connectors, so be careful here, because the left and right buds CAN be swapped over. The only way of knowing consistently which is which is by those serial numbers printed on the IEM body.

SPLs at 200 mV, 1 kHz sine wave (N.B. This may be coupler-dependent):
Left channel (SN 002527): 104.4 dB
Right channel (SN 002528): 104.3 dB

SPLs at 200 mV, 500 Hz sine wave (Hopefully this is less coupler-dependent):
Left channel (SN 002527): 102.2 dB
Right channel (SN 002528): 102.0 dB

Following some useful input from @yuriv, the suggestion is for us all to try and measure frequency response using a volume level set to 94 dB for a 500 Hz sine wave. (Some variation around this SPL won't matter much for these IEMs.) For THD measurements, I've used 80 dB @ 500 Hz and 94 dB @ 500 Hz. Feel free to use something else for THD if you prefer, but please report what SPL you used, and please be careful and don't fry our headphones!

Unfortunately, somewhere along the tour it looks like somebody's dog ate the packaging. This is what it used to look like:



A sample set of initial results are given below to get us started. (If anybody would like any of this raw data in ASCII or REW/mdat format, please pm me.)

Spoiler: Recommended Setup

This is roughly the insertion depth you're looking for (ER4XR shown below, but the body housing is identical to that of the ER2SE):

In my couplers, these Etymotic triple flange tips finish pretty much level with the lip of the coupler opening for an ~10.4 kHz primary resonance peak. (N.B. If you push the IEMs in much farther than this, there's a risk of the eartip touching the grill in front of the coupler mic. This is something I noticed after having made my initial measurements and this is why I've now dropped that primary resonance peak from ~11.5 khz to ~10.5 kHz, i.e., a shallower insertion.) The mounting putty stops the IEM vibrating during measurements (which can show up as glitches in the FR). The coupler here is resting on a sorbothane pad, but this isn't necessary for basic FR measurements. I was driving the ER2SE from a QP1R (0.15 Ohm z-out), but higher z-out sources shouldn't substantially change the results as the ER2SE impedance curve is flat (see below).

Spoiler: Basic Frequency Response (Amplitude vs Frequency)

Measurements here were made at an SPL of 94 dB @ 500 Hz:

Additional measurements using Comply Ts100 foam and SpinFit Cp800 silicone tips can be found in the following post: https://www.head-fi.org/threads/general-iem-measurements-discussions.903455/#post-14869632

Spoiler: Impulse Response

Impulse Response

Spoiler: 300 Hz Square-Wave Response

Spoiler: Total Harmonic Distortion

The dynamic driver in the ER2SE has even lower total harmonic distortion than that in the ER2XR and much lower that that in the (more expensive) ER4 balanced-armature series of IEMs. All THD measurements here were made via stepped-sine sweeps at 80 dB:

Comparisons of percentage THD at 80 dB and 94 dB (both measured at 500 Hz):

Here are more detailed breakdowns of the distortion and noise components at 80 dB (@ 500 Hz):

and 94 dB (@ 500 Hz):

Spoiler: Impedance

Pretty much flat at 15 Ohm:

If anybody would like to join this tour, please jump on the bottom of this thread with a request to the last person with the headphones (they're currently with @yuriv and the plan is, I think, for them then to go to @hakuzen, then @castleofargh). I'm not worried about the financial loss if they go missing, but bear in mind they're basically irreplaceable, because no other set of ER2SE is going to measure exactly the same. So I'm trusting you folks to package them carefully and post them on with a reputable shipping company!

Have fun and let's all (hopefully!) learn something

 

[castleofargh]

@hakuzen already has my address from sending me his coupler ^_^.
I'll add one or 2 crappy IEMs to the package for the next guys, just to diversify the calibration approach if they want to.

 

[antdroid]

I'd be interested in joining up on this tour. I'm in Seattle USA, so I can wait til it reaches the US again. Currently using taobao IEC711 coupler and mic.

 

[CoiL]

If I had better rig, I would join. But atm I think it is no point taking part of this with iMM-6 and Audiotool(android). But I will keep eye on this thread.

Btw, if any of You has unused Sony MH755 and willing to sacrifice those, maybe throw them in the pot? Would be interesting to see how different target curves measure.

 

[McMadface]

If I had better rig, I would join. But atm I think it is no point taking part of this with iMM-6 and Audiotool(android). But I will keep eye on this thread.

Btw, if any of You has unused Sony MH755 and willing to sacrifice those, maybe throw them in the pot? Would be interesting to see how different target curves measure.

That's a great idea. I have a MH755 that's just sitting in my desk drawer. I can send it out along with the ER2SE to @hakuzen. Alternatively, I can send it to @csglinux for him to measure, and then we can send our respective sets to @antdroid for consolidation before sending both sets off to @hakuzen in Europe.

 

[hakuzen]

That's a great idea. I have a MH755 that's just sitting in my desk drawer. I can send it out along with the ER2SE to @hakuzen. Alternatively, I can send it to @csglinux for him to measure, and then we can send our respective sets to @antdroid for consolidation before sending both sets off to @hakuzen in Europe.

agree to do the america tour before sending to europe. not in a rush here, and i need time to set up a new silent mini pc in order to get lower noise and disto (so sending to @castleofargh before me is also ok).
check my inventory (profile), if you want i add any of my iems to the list.

thought there were some problems when inserting tri-flange tip into a 711 coupler..

 

[castleofargh]

same here, I say go with common sense and geography instead of pecking order. @csglinux "a girl in your area wants to meet and measure your rig. click to join!". of course this only matters if many people join. if it's the usual 6 or 7 of us for now, no need for a clever algorithm finding the shortest route to reach each points on the map.

here are 2 options to consider IMO, maybe, or not:
-you, as the first man to land on that pair(one small step response for man...), should do measurements at various insertions, associated with maybe pics of the IEM or some easy to understand distance value in reference to ... whatever. so that people with different gears and preferences can have options to try and match their way with your own results. like if they for example want to do the crinacle thing at 8kHz, or something like I do, more aligned to actual insertion of each IEM in my ears so the Ety would go pretty much as far as possible when I measure for myself. or one of many other options available to them including simply what works well for them as far as stable seal is involved on their "coupler".
seems to me like it could help, and maybe help you redo your own reference if you notice that you'd like to measure more like someone else with a super duper rig in the future.

-or complete opposite, we decide right now that your graph in the first post is the universal reference CSG 1337 or something. everybody just agrees to adjust to that and we're done. when we post stuff, we mention that as compensation instead of RAW, DF or Harman,etc. TBH I'm also fine with that too, we wish for a reference, what that reference is doesn't matter all that much to me.

 

[hakuzen]

After some discussion with fellow headfi measurement enthusiasts, I thought it would be helpful to send a pair of Etymotic ER2SE dynamic driver IEMs out on a world tour. The aim isn't to produce a set of reviews for these headphones (and I'm not going to give them marks out of 5 here), but rather to provide a reference data point for our mics and couplers so that we can come to some common agreement. Several of us have 711-compliant couplers which still may have their own small unit variations. Others are using tube couplers that might benefit from a calibration file derived from a common reference measurement (at least for a sealed/non-ported IEM).

Here's my logic in choosing the ER2SE with stock medium silicone flange tips as a reference:

1) They should easily fit and seal well in any coupler.
2) The stock silicone triple flange tips will hold the IEM in place reasonably well even without the additional use of mounting putty (though putty is still a good idea because it stops any vibration of the IEM in the coupler).
3) There can be no ambiguities from variations in preparing or rolling foam tips.
4) They're relatively neutral, with no wild FR swings.
5) They have very low distortion, which makes for an interesting challenge in trying to compare THD measurements.
6) They have a flat impedance curve, so no wild swings in FR, even if higher z-out sources are used to drive them.
7) We got them at a reasonable price, many thanks to @wquiles

I suggest using the stock medium triple flange tips, at least as a starting point. It should be possible to find an insertion depth where the resonance peak sits at (or similar to) those shown in the figures below (see also first spoiler link for a photo). Ideally, to get the best agreement, you'd want to try and match the primary resonance peak that I ended up with (~11.5 kHz). Also, note that the left channel is unit number 002527; the right is unit number 002528. The ER2SE uses MMXC connectors, so be careful here, because the left and right buds CAN be swapped over. The only way of knowing consistently which is which is by those serial numbers printed on the IEM body.

SPLs at 200 mV, 1 kHz sine wave:
Left channel (SN 002527): 105.6 dB
Right channel (SN 002528): 105.5 dB



A sample set of initial results are given below to get us started. (If anybody would like any of this raw data in ASCII or REW/mdat format, please pm me.)

Spoiler: Recommended Setup

This is roughly the insertion depth you're looking for (ER4XR shown below, but the body housing is identical to that of the ER2SE):


In my coupler, these Etymotic triple flange tips finish just inside the lip of the coupler opening for an ~11.5 kHz primary resonance peak. The mounting putty stops the IEM vibrating during measurements (which can show up as glitches in the FR). The coupler here is resting on a sorbothane pad, but this isn't necessary for basic FR measurements. I was driving the ER2SE from a QP1R (0.15 Ohm z-out), but higher z-out sources shouldn't substantially change the results as the ER2SE impedance curve is flat (see below).

Spoiler: Basic Frequency Response (Amplitude vs Frequency)

These are averaged data from two sets of 711 couplers, both of which were well within 1 dB of each other (raw data available upon request). Measurements were made at 95 dB.

Additional measurements using Comply Ts100 foam and SpinFit Cp800 silicone tips can be found in the following post: https://www.head-fi.org/threads/general-iem-measurements-discussions.903455/#post-14869632

Spoiler: Impulse Response

Impulse Response

Spoiler: 300 Hz Square-Wave Response

Spoiler: Total Harmonic Distortion

The dynamic driver in the ER2SE has even lower total harmonic distortion than that in the ER2XR and much lower that that in the (more expensive) ER4 balanced-armature series of IEMs. All THD measurements here were made via stepped-sine sweeps at 80 dB:

Spoiler: Impedance

Pretty much flat at 15 Ohm:

If anybody would like to join this tour, please jump on the bottom of this thread with a request to the last person with the headphones (they're currently on their way to @McMadface, but @hakuzen is already next on the list). I'm not worried about the financial loss if they go missing, but bear in mind they're basically irreplaceable, because no other set of ER2SE is going to measure exactly the same. So I'm trusting you folks to package them carefully and post them on with a reputable shipping company!

Have fun and let's all (hopefully!) learn something

nice measurements, btw. well done!

 

[csglinux]

It sounds like you folks have this all figured out - great! I agree it makes sense to do a North American tour before sending anything off to Europe. I knew @hakuzen was busy building his fanless PC, so probably wasn't in a hurry. Also, @castleofargh, I want you to know that if any foxy young chick in the LA area would like to measure these ER2SE's first, I would - in the name of science - be willing to bump you one slot further down the queue

I'm with @castleofargh on the issue of insertion depth. I just use whatever's natural for my ears and/or couplers. With the Ety line of IEMs, you're typically shooting for the deepest insertion you can achieve and you wouldn't normally start slowly pulling out your ER_whatever until you'd achieved an 8 kHz resonance peak. There probably is merit in measuring multiple insertion depths - if only to show that there are variations with this and that one measurement isn't definitive. However... for us all to compare (and exchange data files?) for multiple insertion depths sounds like a lot of work and I'm lazy by nature. So I'm sort of sympathetic to the bind that @crinacle is in here - I don't want to measure a large number of possible insertion depths, so I did in fact match the L to R resonance peaks in those ER2SE measurements (they initially ended up at about 11.5 kHz and 11.9 kHz). I'll let you folks decide on this. (BTW, I've had no issues inserting any Etyomic IEMs, with any type of eartip, into my couplers.)

The MH750 might actually add something new and interesting, because I believe it's vented(?). Again, I'll let you folks decide. The ER2SE is already in the post to @McMadface, but I had a thought here - @McMadface - why not just bring your MH755 to the SoCal CanJam and we can measure them there before sending everything off to @antdroid?

If I had better rig, I would join. But atm I think it is no point taking part of this with iMM-6 and Audiotool(android). But I will keep eye on this thread.

@CoiL - I'd love to have seen a comparison with your rig Please reconsider joining the tour. The Audiotool (Android vs iOS) story is quite an interesting one and it needs more love and attention. I personally didn't have much luck with the Android version, but that may just have been because of my Android hardware and me giving up because I'm lazy. I always saw a low-frequency roll-off, whereas the iOS version on my iPad (same name, but totally different app and dev team) agreed almost perfectly with the results I got from REW. The Android AudioTool dev and I had many communications about this and it seems the app is just at the mercy of the Android system and ADC. It could partially be down to cheap mics, but I suspect most of the ADCs are intentionally designed with a high-pass filter to cut rumble noise when people are holding and bumping mics around. I tested this directly from my phone's 3.5 mm socket and also via USB-C Dongles into the phone (the Google dongle versions 1 and 2 and the new Apple USB-C dongle) and these were all even worse. I still think this app might have merit with the right hardware though. It is awesome to be able to do measurements from your phone.

nice measurements, btw. well done!

Thanks @hakuzen - although it remains to be seen how nice they are. I suspect one day somebody will measure the ER2SE with the new GRAS "Hi-Res" coupler and show that in fact there are no 10-12 kHz resonance peaks

 

[McMadface]

The MH750 might actually add something new and interesting, because I believe it's vented(?). Again, I'll let you folks decide. The ER2SE is already in the post to @McMadface, but I had a thought here - @McMadface - why not just bring your MH755 to the SoCal CanJam and we can measure them there before sending everything off to @antdroid?

Yeah, we can do that. SoCal CanJam is 2 weeks away, so I didn't know if you all want to wait that long to get things kicked off, but it certainly is more efficient. If anyone else going to CanJam wants to bring their rigs, they can get measurements before the sets start their world tour.

 

[castleofargh]

no hurry here, I first discussed doing something similar 3 or 4 years ago, the important element is to find motivated people, the more the better. how long it takes doesn't matter much for me. if I'm totally honest, I don't need that reference for my own fooling around with mics and EQ. I support this action because I think forum members will compare graphs from different sources no matter how many times we tell them not to. so this is my effort toward a more common amateur reference to help them be slightly less wrong when they keep doing it anyway.
then hopefully we enter a vicious circle, people get less confused when wrongly interpreting FR because the effective variations are now slightly minimized(hopefully), so things more often correlate with their impressions. so maybe they get more involved and learn what they should have learned before pretending to know how to interpret a FR graph. then audiophiles see more people using graphs and actually making sense of them, so some of them stop thinking like a medieval peasant and join the modern era of this hobby. then in 5 years, audiophiles are no longer mentioned to start a joke about gullibility and ignorance.

ok maybe I'm slightly optimistic, it might take 5 more generations and what we're doing here we will probably have no impact whatsoever. but beside those tiny details, I think we rock. save the cheerleader, save the world.

 

[Slater]

Maybe I missed this, but have you guys considered sending 2 different IEMs as a reference?

Because one way to tell if your rig is calibrated would be to then test it against the 2nd one. And if the graphs for both IEMs each matched the ‘standard’ versions, then you KNOW your self is perfectly dialed in.

Also, in the event that 1 of the 2 gets lost, broken, does, etc, at least there’s the other one that can keep circulating around (hopefully for a long time to come).

The cost of shipping 2 IEMs in a box instead of 1 would be a trivial difference.

Just a thought. I apologize if this has already been discussed.

 

[McMadface]

Maybe I missed this, but have you guys considered sending 2 different IEMs as a reference?

Because one way to tell if your rig is calibrated would be to then test it against the 2nd one. And if the graphs for both IEMs each matched the ‘standard’ versions, then you KNOW your self is perfectly dialed in.

Also, in the event that 1 of the 2 gets lost, broken, does, etc, at least there’s the other one that can keep circulating around (hopefully for a long time to come).

The cost of shipping 2 IEMs in a box instead of 1 would be a trivial difference.

Just a thought. I apologize if this has already been discussed.

Yup. I'm adding a MH755 into the tour.

 

[CoiL]

Maybe I missed this, but have you guys considered sending 2 different IEMs as a reference?

Because one way to tell if your rig is calibrated would be to then test it against the 2nd one. And if the graphs for both IEMs each matched the ‘standard’ versions, then you KNOW your self is perfectly dialed in.

Also, in the event that 1 of the 2 gets lost, broken, does, etc, at least there’s the other one that can keep circulating around (hopefully for a long time to come).

The cost of shipping 2 IEMs in a box instead of 1 would be a trivial difference.

Just a thought. I apologize if this has already been discussed.

That`s why I gave an idea to include MH775 and it`s done now

 

[Slater]

Yup. I'm adding a MH755 into the tour.

That`s why I gave an idea to include MH775 and it`s done now

Smart move! Nice job, gents