[csglinux]

Is it possible to join this tour? Like in a couple months? I've got measuring gear being shipped and I'd like to use these to calibrate them.

No problem @aldarrin - welcome aboard

We'll just need to figure out logistics. The tour kit is currently in Singapore and was due to go to Europe next. Ping us again once you get your rig up and ready to go.

 

[earfonia]

The tour kit is with me now. I've done all the FR and THD measurement, and will do the impedance measurement probably this week. I can send the kit to the next participant next week. @csglinux please advise the next participant to whom I shall send the tour kit. Thanks!

Here are the FR results comparing the result from GRAS RA0045 and my coupler, E610A-SN20229. All measurement were done at 94 dB SPL at 500 Hz, but I shifted the right channel measurement down 5 dB after measurement on REW, just to visually separate them from the left channel for easier observation. Left on Top, Right at the Bottom. GRAS RA0045 in Red, my clone coupler E610A in Blue.

Etymotic ER2SE:

BLON BL-05:

Samsung AKG IG955:

Sony MH755:

I can share the rest of the measurement as well to those we are interested (THD and impedance). Just PM me if you would like to get the .mdat file.

 

[csglinux]

Sorry for the slow reply folks - too many things going on this week.
Some bookkeeping issues:

1) I don't actually know who's up next. @McMadface had mentioned @crinacle and @oratory1990 were
interested, but that was a while ago. I've not been in touch with either of those guys directly and I know they don't hang
out much on headfi much these days. @oratory1990's headfi account doesn't even seem to be active anymore. I'll reach out on Discord, but
I believe @oratory1990 is in Austria, so @crinacle may be geographically closer?
If there's anybody in the rough proximity of Singapore who'd be interested in receiving the tour package, time to speak up!
Otherwise, the folks that are waiting in Europe and the US still need another couple of months, so, @earfonia,
I'd suggest if nobody else speaks up, just hang on to the kit for now.

2) I think we've probably answered most of the initial burning questions about the performance of the various couplers.
I haven't pushed all the data yet, but we have measurements for those tour IEMs for all types of coupler
(clone and otherwise, including GRAS and B&K). When I find the time, I'll try to do a meta-analysis of the data we
have and write up some conclusions. Going forward, I think the most logical and useful continuation of this
tour is exactly what @earfonia has just done, i.e., to take one of the better-quality clone couplers and calibrate it
to match a known yardstick like the GRAS RA0045. (Some clone couplers do come with their own calibration curves,
but even then, those calibration curves seem to be of questionable quality.)

3) I have a confession to make... Before sending the tour package on to @earfonia, I re-measured all the IEMs.
I wanted to check that I could perfectly match the measurements made a year ago. That was a fun learning
exercise. No issues matching the Sony MH755. I also managed to perfectly match the ER2SE, but matching the treble
was a LOT of work. I must have done about two dozen measurements, all with slightly varying tip-on-stem fit
and insertion depths. I finally got an excellent match, but it does show what a giant crapshoot the treble is.
There's no right or wrong - it's all just a physical consequence of different depths and volumes.
One of the IEMs (Samsung/AKG EO-IG955) gave me a lot more grief and I couldn't quite match the levels
of sub-bass I'd previously measured. I don't think this is burn-in though I have to offer profuse apologies
for slightly tampering with that IEM. When I was experimenting with the various mems mics, I went through a stage
where I thought it would be a good idea to use cyanoacrylate glue on the stem of the IEM to hold the mems mics
in a consistent position and orientation. That worked great - cyanoacrylate glue sticks to plastic really, really well.
The problem is/was getting it off afterwards. If you look carefully at the nozzle stems on that IEM, they're not
perfectly smooth and round anymore. I think I may have inadvertently created a very small front-vent mod. It's not
a large effect, but it's probably large enough that I should upload newer measurements for that particular IEM.

@McMadface - is your shared Google drive still active? If so, can you please pm me and remind me how to access it?
I can then push the REW mdat files for all these IEM/coupler combinations.

 

[aldarrin]

80 dB test has it's own problem. For 0.1% harmonic distortion for example, that means the harmonic distortion is 60 dB below the signal. If the signal is at 80 dB SPL, that means the distortion is in the 20 dB SPL range. The problem with this measurement is that the room ambient noise where the IEM measurement is conducted has to be lower than 20 dB SPL. This is practically impossible for many of us. And some IEM like the BLON BL-05 and BL-05S, Beyerdynamic Xelento, and recently I measure TFZ Tx Bear Monica, they can have distortion down to less than 0.01 % in some area of the frequency range. That means 80 dB below the measurement signal. If we measure at 80 dB SPL, the harmonic distortion is at around 0 dB SPL. So the ambient noise around the test equipment should be less than 0 dB SPL. We need anechoic chamber to do the measurement properly. IMHO 94 dB SPL is just the right loudness for IEM measurement looking at all practical aspects and limitations. THD is usually correlated with the SPL anyway. I'm not so interested to see how low can it goes by lowering the signal SPL to 80 dB, but I'm more interested to see IEM that can maintain low THD in higher SPL. If an IEM can performs well in higher SPL, it will usually perform well in low SPL. So THD measurement in higher SPL is a better representation of the IEM performance.

For soundproofing my test setup, I modified a waterproof Pelican case and added 1 inch high density rubber mat inside it to improve soundproofing and added weight to the whole case to reduce low frequency vibration. The rubber (gym) mat is pasted to the case walls using double sided tape. With the soundproof case my setup can measure down to 0.05% THD without affected by room ambient noise from 20Hz - 20kHz, and some area after 1kHz can go as low as 0.005% before masked by room noise. The most difficult to block is the low frequency noise.

This is the ambient noise measured when the lid of the case closed:


The hard case for the measurement equipment:

Crazy thought, but can we just get a calibrated omnidirectional mic added to capture ambient noise and substract that from the measurements? Otherwise it seems like you'd need to go down endless rabbit holes trying to remove every source of environmental noise.

 

[earfonia]

Crazy thought, but can we just get a calibrated omnidirectional mic added to capture ambient noise and substract that from the measurements? Otherwise it seems like you'd need to go down endless rabbit holes trying to remove every source of environmental noise.

I'm afraid it is not that simple. the noise reduction mic would have to be coincident in placement with the measurement mic to be able to capture the exact phase of the ambient noise, so the subtraction is accurate, while at the same time isolated from the sound coming from the IEM that is being measured.

That's the reason I use the Pelican case reinforced with 1 inch of dense rubber mat covering all over the internal sides of the case, to block outside noise. So I could do THD measurement more accurately. For frequency response measurement at 94 dB SPL (@ 500 Hz), sound isolation box is not necessary as long as the room is not noisy. But THD measurement requires low (very low) ambient noise for accurate measurement.

That is also the reason that I think THD measurement at 80 dB SPL is not realistic and not accurate in most cases.

 

[aldarrin]

Yeah, figured it'd be a long shot (or at the very least require another coupler / mic located near the measurement rig). But it sounds like measuring FR it isn't as important? I'm not super worried about THD other than avoiding problem areas for the driver (e.g. resonances) when EQ'ing.

 

[earfonia]

Yeah, figured it'd be a long shot (or at the very least require another coupler / mic located near the measurement rig). But it sounds like measuring FR it isn't as important? I'm not super worried about THD other than avoiding problem areas for the driver (e.g. resonances) when EQ'ing.

I guess what important and what less important are still in learning process. What I'm interested to do now is to learn how to relate / interpret measurement to what I hear. How important are those parameters that we measure, we will probably have better understanding about it later. All in progress. Trying to make sense of what we measure, and to discover new things.

For example, below is measurement if Tingker TK200, huge sudden THD bump at 5.6 kHz to around 2-3% distortion. If I don't see the THD measurement, by listening test alone, I won't notice that 2-3% distortion.

 

[castleofargh]

It's really hard to match THD with subjective impressions. In my case there is another issue added on top of that fact, and that's how I almost never listen to music at the levels used for THD measurements. For critical listening and blind tests, I find that I usually get the best results slightly above 60dB SPL. And as strange as it might be, casual listening is done at even lower levels. For some reason that's how I like it best. The lower the ambient noise, the lower I end up playing music or the TV. With strongly isolating IEMs in a quiet place, I often end at a hard limit imposed by the noises I get from my own body covering some of the music. That's when I realize I’ve gone too low and I’m a weirdo.

So the actual THD values could be different when I’m using the IEM. I also end up with a good deal of ambient noise at levels where the extra signal is born. All in all, THD graphs are of little personal use. With that said, if I measure something like 3% in the midrange like your example, my bias will make me look for another IEM. Doesn't matter if I notice it or not by ear, my brain goes: ”bad IEM scary! Me no touch” or something completely rational like that.

 

[csglinux]

CE1 tour is now on! I created a separate thread for it for easier reference:

https://www.head-fi.org/showcase/introducing-the-ce1-the-worlds-most-somethingist-headphone.25259/

Feel free to do your bit to help audio science

 

[castleofargh]

That will get even fewer people than this tour, but I applaud the will to act and personal efforts.

BTW, I'm back home at last, so put me back in the ER2SE tour for whenever it's convenient, plizz.

 

[Raketen]

CE1 tour is now on! I created a separate thread for it for easier reference:

https://www.head-fi.org/showcase/introducing-the-ce1-the-worlds-most-somethingist-headphone.25259/

Feel free to do your bit to help audio science

Cool idea- Is that a community project? (Is the data going to be public domain? etc...)

 

[csglinux]

Cool idea- Is that a community project? (Is the data going to be public domain? etc...)

That's the goal. We'll try to post the data somewhere where we have tools to compare measurements with things like the Harman/Olive metrics, e.g.: https://www.hypethesonics.com/er2se-database/

 

[kmmm]

Hi.
is it possible to join this tour? I saw the kit might be heading to Europe? Or am I too late?

 

[csglinux]

Hi.
is it possible to join this tour? I saw the kit might be heading to Europe? Or am I too late?

It's never too late Where are you located?

 

[kmmm]

Thank you!
I’m in Norway