IEC 60318-4 Clone Coupler, is it worth it?

[Earfonia]

My journey in In-Ear Monitor (IEM) measurement started with some common measurement microphones and DIY shrink tube couplers.

I've been interested in IEM measurement for quite a long time, but not ready to spend big bucks on GRAS or B&K couplers. The idea is to search for a more affordable measurement setup that can produce meaningful result. What I mean by meaningful is that the measurement result is at least consistent, so we can compare different IEMs measurements and get consistent result from the setup. Besides that I hope the measurement result is, to a certain degree, more or less comparable to the measurement result from those expensive couplers.

For the last 2 years or more I've been using China clone IEC 60318-4 couplers and I would like to share what I've learned here.
Big thanks to @JohnYang1997 for some info he shared with me that helped me in making the right decision in purchasing my latest IEC coupler from Taobao that has been my main IEC coupler for now.

Now I have 3 clone (read: compliant) IEC 60318-4 couplers:

1. IEC 60318-4 plinth coupler with ½ inch ICP prepolarized microphone. Model E610A SN: 20229 (latest purchase)
2. IEC 60318-4 plinth coupler with plug-in-powered (±5V) microphone. Model E610A SN: 20201
3. IEC 60318-4 coupler without microphone, model E610A coupler SN: 30168

The coupler without mic can be attached to measurement mic with nozzle diameter around 12.7 mm. For example MiniDSP UMIK-1.
The 1st and 2nd couplers in the list come with a build in microphone and calibration files.

A few lessons that I've learned from using the clone couplers:
1. The calibration files were poorly done, and therefore useless. So we need a way to calibrate the clone couplers with better coupler like the GRAS or B&K couplers. This is quite a challenge if we don't have access to exactly the same IEM measured by GRAS or B&K couplers and compare the measurement with their measurement result. Fortunately the 1st coupler in the list above produce pretty close measurement result with GRAS RA0045-S1, even without any calibration file.
2. When calibrated, some clone couplers can produce measurement result pretty close to the better and more expensive couplers like the GRAS coupler.
3. Although frequency response can be calibrated, harmonic distortion measurement cannot be calibrated. So poor microphone usually resulting in higher THD measurement than THD measurement result from a good branded coupler.

Thanks to a Head-Fi member, csglinux from https://www.hypethesonics.com/, he started a tour of Etymotic ER2SE measurement, so that I (and other participating members) could compare measurements of ER2SE and few other IEMs in the tour package with his GRAS RA0045-S1 coupler (RA0045 coupler + 40AO mic + 26CA preamp).
https://www.head-fi.org/threads/ety...phone-for-your-ears-and-your-couplers.908512/

After about 2 weeks of non-stop measurements and drafting the calibration files for my latest coupler (SN: 20229), the following are the comparisons of frequency response measurement result from GRAS RA0045-S1 and my coupler, E610A-SN20229. All measurements were done at 94 dB SPL at 500 Hz, but I shifted the right channel measurement down 5 dB after measurement in 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-SN20229 in Blue.

Etymotic ER2SE:

Samsung earphones EO-IG955 (Tuned by AKG)

Sony MH755:

BLON BL-05:

The reason I was looking for a better coupler and finally bought the ICP mic based coupler, is to get better THD measurement. The THD measurement from the coupler with plug-in-powered microphone (SN: 20201) is always higher than expected. Among the 4 IEMs in the tour kit, BLON BL-05 is the most impressive in THD measurement. It has very low THD to test the capability of the coupler. The coupler with plug-in-powered microphone showing THD measurement in the range of 0.5% - 1.0% THD for BLON BL-05, while the ICP mic based coupler (SN: 20229) goes down to lower than 0.01%, showing similar curve to the GRAS RA0045 with only slightly higher overall THD. Pretty impressive for a coupler that cost probably 1/6 - 1/7 of the GRAS RA0045-S1.

BLON BL-05 - Left THD - GRAS RA0045-S1:

BLON BL-05 - Left THD - E610A-20229 (Clone Coupler):

Here is my current measurement setup:

Both RME Babyface Pro FS and the B&K 1704 are battery powered, because I put everything inside a Pelican case that function as noise isolation container. It is very difficult to measure IEM with low THD like the BLON BL-05 without noise isolation container, because the ambient noise will be higher than the harmonic distortion signal that will be measured. I cut a hole on the Pelican case to install USB 3.0 coupler for USB connection of the Babyface Pro FS to my computer. I added 1 inch high density rubber mat and sound isolation seal to improve the noise isolation.

My conclusion so far, I'm quite happy with my latest clone coupler (ICP Mic - SN: 20229). IMHO it is worth it.
One day I might end up with a GRAS coupler, but for now, this clone coupler is good enough.

I recently ordered the Panasonic RP-TCM125 IEM for measurement, so I can compare it with Amir's review of that IEM:
https://www.audiosciencereview.com/.../panasonic-rp-tcm125-review-budget-iem.22797/

If @amirm has the time, I would love to send over a few IEMs to be compared to Amir's GRAS 45CA

Anyone who has experience with IEC 60318-4 clone couplers, please share your experience so we can learn which clone coupler is worth the money and which are not

 

[BrEpBrEpBrEpBrEp]

I recently got one as well - type E610A from "Sounds Good Store" on AliExpress. I'm curious - which vendor on Taobao did you buy from?

I got pretty close to Crinacle's measurements on my Blon BL-03s using the vendor's calibration when I did a quick check, but I've been planning on checking all my IEMs against Crinacle & Oratory1990 measurements to see how it does.

 

[Earfonia]

The one with ICP microphone, SN: 20229, I bought from here:

Taobao link:
https://item.taobao.com/item.htm?spm=a1z0d.6639537.1997196601.15.4c197484slCmuU&id=542357654840
Choose option / type : 4

the seller has the 2nd hand B&K 1704, but not listed in his shop. Need to talk / chat with him if we want to include it.

If getting from Taobao is difficult for some, here is the same item on AliExpress:
https://www.aliexpress.com/item/4000072763864.html?spm=a2g0s.8937460.0.0.96d52e0exzIpC8
Choose type: 4

95.2US $ |Iec711 318-4 Artificial Ear Armature Tuning Test Headphone Tuning Test Soundcheck - Air Conditioner Parts - AliExpress

Smarter Shopping, Better Living! Aliexpress.com

www.aliexpress.com

Choose type: 4

B&K 1704:

177.9US $ |Second-hand Not Brand-new Danish B&k1704-c-102 Signal Amplifier Signal Conditioner - Air Conditioner Parts - AliExpress

Smarter Shopping, Better Living! Aliexpress.com

www.aliexpress.com

The one with 5V plug-in powered mic (SN: 20201), I also bought from AliExpress 'Sounds Good Store':
https://www.aliexpress.com/item/4000073069259.html?spm=a2g0s.9042311.0.0.27424c4dEWb0my
Type: 1

 

[Earfonia]

I recently tested those couplers and another 2x DIY couplers to measure ATH-CKN50 and BLON BL-05S.

The couplers I used for testing:

1. IEC 60318-4 plinth coupler with ½ inch ICP prepolarized microphone. Model E610A SN: 20229
2. IEC 60318-4 plinth coupler with plug-in-powered (±5V) microphone. Model E610A SN: 20201
3. Earthworks M23 + E610A coupler SN: 20494
4. MiniDSP UMIK-1 + E610A coupler SN: 30168

The coupler no 3 is a DIY Earthworks M23 + E610A coupler belong to csglinux. I borrow from him for testing. Coupler no 4 is mine. The coupler 1 and 2 using calibration files that I created to match those to GRAS RA0045 coupler. The M23 calibration file created by csglinux, and the MiniDSP UMIK-1 using the calibration file from the factory (not made or suitable for pressure mic application like IEM measurement).

Here is the measurement result:

Frequency Response Measurement of Audio-Technica CKN-50 (using small large bore ear tip):

Frequency Response Measurement of BLON BL-05S (using small generic 3.8mm bore ear tip):

Audio-Technica CKN-50 THD Measurements:

BLON BL-05S THD Measurements:

 

[BrEpBrEpBrEpBrEp]

Wow, that's quite the setup. I mainly bought mine for FR measurement & EQing, so I'm not too bothered about inability to measure low levels. It's good to see differences between your calibration results and csglinux's seem pretty minimal with the exact same IEM - seems this method works even assuming some small unit variation in the IEMs used.

Can you post the original supplier's calibrations as well as your calculated ones? I'm curious as to how off exactly they are. Crinacle is pretty close for most of the 20 Hz - 10 kHz range, but I'm not sure what kind of calibration he used - I'd previously assumed the one provided by the supplier.

 

[Earfonia]

For about 2 years I used that 5V coupler 20201 with supplier calibration file. Only the last few months since I got in contact with csglinux and the IEMs from the tour package then I realized the calibration file was a bit off, especially on the bass region below 100 Hz, where the difference is almost 3 dB at 20Hz. The rest actually not bad. I haven't measured the IEMs from the tour package using the 20201, but here you can see the difference of measurement of the same IEMs above, the CKN50 and BL-05S, between the 20229 that already calibrated to GRAS RA0045, and the 20201 (5V mic) using their original factory calibration file:

ATH-CKN50:

BLON BL-05S:

 

[Skeptischism]

Thanks so much for this. I was just noodling around having a look before posting a thread to ask if any members had experience in this area. I lost one of my JH16 customs a couple of months ago (I nearly cried) and instead of buying a new set for a few grand, I decided last week I would make my own. I can buy the required materials (resin, impression/mould material, connectors, drivers) and equipment (coupler/fixture, UV chamber) for considerably less even for the first set and from then on, they will be even cheaper.

I already have the minidsp mic and i'm building an ADC that should match the babyface for after that. i'm already designing a multichannel dac/dsp for digital speaker crossovers and it will drive a balanced set of headphones on each pair of outputs, meaning it can drive 4 pairs of balanced headphones, or .... a single 3 or 4 way crossed balanced IEM. so a multichannel headphone amp and DSP just kind of falls out of my DSP design requirements that was a nice realization to offset the sadness of losing my JH16. Not sure if I need the B&K, as I should be able to rig up a mic-pre with phantom power for the mic, although perhaps it would be better for consistency with other peoples measurements, if they are standard?

So basically the rig will be used to measure each driver in the IEM individually and correct, then cross and measure the combined output and perhaps see what FIR/convolution can do for multi-driver IEMs . I know a couple of attempts have been made at digital XO for IEMs before, by the big players, but neither were wildly successful and neither were able to bring the sort of performance and DSP that is possible today.

My first sets of IEMs may be utter failures, but I think this sort of setup could do for IEMs what digital has done for speaker development. If nothing else, it'll be fun.

 

[Skeptischism]

So my friends can expect IEMs for xmas

 

[Earfonia]

Thanks so much for this. I was just noodling around having a look before posting a thread to ask if any members had experience in this area. I lost one of my JH16 customs a couple of months ago (I nearly cried) and instead of buying a new set for a few grand, I decided last week I would make my own. I can buy the required materials (resin, impression/mould material, connectors, drivers) and equipment (coupler/fixture, UV chamber) for considerably less even for the first set and from then on, they will be even cheaper.

I already have the minidsp mic and i'm building an ADC that should match the babyface for after that. i'm already designing a multichannel dac/dsp for digital speaker crossovers and it will drive a balanced set of headphones on each pair of outputs, meaning it can drive 4 pairs of balanced headphones, or .... a single 3 or 4 way crossed balanced IEM. so a multichannel headphone amp and DSP just kind of falls out of my DSP design requirements that was a nice realization to offset the sadness of losing my JH16. Not sure if I need the B&K, as I should be able to rig up a mic-pre with phantom power for the mic, although perhaps it would be better for consistency with other peoples measurements, if they are standard?

So basically the rig will be used to measure each driver in the IEM individually and correct, then cross and measure the combined output and perhaps see what FIR/convolution can do for multi-driver IEMs . I know a couple of attempts have been made at digital XO for IEMs before, by the big players, but neither were wildly successful and neither were able to bring the sort of performance and DSP that is possible today.

My first sets of IEMs may be utter failures, but I think this sort of setup could do for IEMs what digital has done for speaker development. If nothing else, it'll be fun.

Wow that's a very interesting approach to DIY your own IEM!
The B&K1704 unit is just to drive the ICP / CCLD microphone, but we can easily DIY a circuit to provide constant current to the microphone, if not mistaken the ICP mic requirement is around 22 V ± 2 V (typ. 21 V) with constant current around 3 to 4.1 mA (nom. 3.55 mA). So depending on the microphone that you use. I bought the B&K unit because I got a very good price for a 2nd hand unit.

If we just need FR comparisons between drivers, I guess simple solution like MiniDSP mic with DIY coupler is sufficient. But we know the result won't be accurate as to be compared with measurement result from other measurement equipment. So if you plan to build more IEMs and will use the measurement gear frequently, I guess the ICP based E610A coupler is a good invenstment. And please take note that MiniDSP UMIK-1 mic is not air tight between the mic capsule and the mic barrel where it is attached. I have 2 of UMIK-1 both the same, so it will produce measurement with lot's of bass rolls off, and basically inaccurate bass measurement. What I did with UMIK-1, I opened the top part of the mic barrel and apply some glue around the mic capsule to seal it.

Btw, do you have any thread / post about DIY ADC? I'm a noob on that subject, but interested to learn. Thanks!

 

[Skeptischism]

No, sorry, although I will probably do that when its ready. it is being built primarily for DAC measurements, since I cannot afford/justify an audio precision system, but I need better performance than the regular crop of commercially available ADCs. Combined with a passive notch filter and measurement amplifier, I should be able to measure -115-120dB THD, so a couple orders of magnitude better than the 0.01% the best capsule/mic on your list is capable of. I managed to score a couple NOS AK5394AVS and will parallel those 2, for better SNR (its already capable of better than the datasheet specifies)

regarding the IEMs; there is a very long and informative thread on head-fi, which started many years ago. plenty of info. i'm pretty good with my hands; somehow still pretty steady.

In nearly 900 pages, active crossovers have only been mentioned a handful of times and nobody actually did anything. I plan to change that. as far as I know, the only people that did anything are JHAUDIO and Unique melody. Ultimate Ears apparently had a DSP they used for demos at shows that allowed them to mimic the sound of any and all of their IEMs at shows, but AFAIK all of these were using IIR filters only, no FIR.

 

[somebodyelse]

No, sorry, although I will probably do that when its ready. it is being built primarily for DAC measurements, since I cannot afford/justify an audio precision system, but I need better performance than the regular crop of commercially available ADCs. Combined with a passive notch filter and measurement amplifier, I should be able to measure -115-120dB THD, so a couple orders of magnitude better than the 0.01% the best capsule/mic on your list is capable of. I managed to score a couple NOS AK5394AVS and will parallel those 2, for better SNR (its already capable of better than the datasheet specifies)

In case you're not aware, see this thread for details of @IVX's soon to be released Cosmos ADC and analog processing unit which aim to do pretty much what you've described.

 

[Skeptischism]

No I hadnt seen that, but given we already have our passive notch (the most expensive part, given component count and quality)., built using Samuel Groners PCBs, there is no reason to buy it. i'm actually hoping to do significantly better than -120db (the amps are specced WAAAAY better than the ADC of course), just setting goals that are easily achievable. measurement at these levels is tricky and I dont like the use of minijacks in the slightest. for reliable and repeatable measurements at these levels, rigid SMA connectors and rigid 'cables' are preferred. otherwise just moving the cable around is enough to throw the results out. We (hochopepper actually) also already has an ESS ADC using their top chip. the 2x AKM is just one i'm building for this project, as it will outperform the ESS in a few areas. the autoranging and scripting will be where the magic lies, to do sweeps etc. at this point it would cost me more, for a product that doesnt fit my needs, vs one that exactly fits our needs.

 

[Skeptischism]

SMA. and short, rigid 'cables'

 

[Skeptischism]

. also, the lack of 50Ω termination option on IVX is enough to rule it out for me, if it's intended for measurements. we are also looking to add a much wider bandwidth SAR type industrial ADC and neither XLR or mini will cut it.

 

[IVX]

Guys, to get THD -115-120db level with a notch, any laptop built-in audio is more than enough. Cosmos ADC reading THD at -130db level and THD+N 123db without a notch filter just a cigarette-pack-sized ADC.

 

[Skeptischism]

Like I said, I expect to do significantly better, but i'm not going to put an exact number on it before I have even preliminary builds. you are correct though, that was a brain fart, what I mean is to be able to measure with confidence, devices that produce at least -115-120 THD+N. All going to plan, I would hope to get to around ~ -150db if using the full 40dB gain in the LNA. A laptop built in audio ADC is not close to suitable for my purposes, notch or not. My apologies for the confusion. THD+N not vanilla THD. I havent used the AK before and its an old part that does quite well by today's standards. Will see how we go. Wont be for sale, the chips are unavailable anyway. actually I notice the AK chip specs SINAD of -110, rather than straight THD+N. we shall see.

anyway, I dont want to continue Off topic here. my apologies for the diversion @Earfonia.

 

[IVX]

just as an experiment, I tried to use my desktop audio-in(residual THD -94db@1kHz) with Cosmos APU notch, and got THD from AP SYS2522 analog_generator -124db. At the same time, AP reads from that notch -127db or THD. Hence, an extremely cheap device as an analog notch-filter(OPA1612+ a few caps/resistors), solves the problem of precise THD measurement nearly for free but only at a fixed frequency.
The main unsolvable lack of AKM and Ti ADCs is the drop of linearity near to full scale. That lack doesn't let to use these ADCs for decent levels of THD+N measurements without a notch.
PS: no one practical engineer will measure uVolts and sub uVolts levels without differential inputs aka balanced inputs. That's why SMA, BNC etc is the wrong choice - In+/In-/GND are needed.

 

[Skeptischism]

Perhaps this is better left for a paid for thread?

did I say without balanced inputs?

 

[Skeptischism]

no need to be so defensive.

 

[IVX]

Custom SMA?