A poor (wo)man's microphone measurement equipment

[MicUlli]

Hi All,
measuring microphone performance is very meaningful for the hobbyist for several reasons. The most important issue is that you can get information far beyond your subjective "golden ear" perception.

This thread shall
1. collect your requirements regarding microphone measurement and
2. give some advice how basic measurements can be done.

From time to time i will post some tips (partially guided by the normative reference IEC 60268-4)..

I personally prefer REW as basic test software, therefore all information i will post is related to it.

I am very curious to receive your comments / thoughts
MicUlli

EDIT: I have finished my document series and post it at the beginning of this thread for convenience:

audio interface calibration_r2 was modified, CMRR measurement of the interface was appended.
mic_measurements_1 is focused on electrical measurements
mic_measurements_2 is related to the acoustical measurements

 

[rock soderstrom]

This thread comes at the right time, I am still at the very beginning of my microphone measurements attempts and am curious what information will be mentioned here.

I think it is very important in microphone DIY to have a reliable objective measurement basis in addition to subjective perceptions.

 

[gyraf]

clio is probably the standard for this - the chirp technology makes it possible to get usable responses without having to go anechoic

apart from that, I use AudioTester and our roof at night, flying the mic and having speaker pointing upwards. Works only in calm weather.

 

[soliloqueen]

you can "hopscotch to anechoic" if you're only doing comparative measurements. if you're testing variations on one capsule, like different k67s, all you need to know is the difference between a reference k67 in your environment and an anechoic chamber and you can subtract that out instead of trying to use a flat omni mic in your environment and potentially introducing reflections to your reference curve. because so many of the differences in performance between designs are due to external dimensions of the capsule and the remaining differences happen mostly off-axis, as long as you're only taking 0deg measurements at far enough away, the differences are negligible if you have a good reference for a capsule of the same size and similar construction

 

[kingkorg]

Hi All,
measuring microphone performance is very meaningful for the hobbyist for several reasons. The most important issue is that you can get information far beyond your subjective "golden ear" perception.

This thread shall
1. collect your requirements regarding microphone measurement and
2. give some advice how basic measurements can be done.

From time to time i will post some tips (partially guided by the normative reference IEC 60268-4)..

I personally prefer REW as basic test software, therefore all information i will post is related to it.

I am very curious to receive your comments / thoughts
MicUlli

Why not share your findings and approaches regarding measurement to start with? This way, it seems to me, you are out fishing for other’s knowledge for personal (potentially commercial?) use.

To this day i'm not really certain about your intentions regarding mic stuff. You have shared some projects which i am very thankful for, but you also mentioned developing some commercial products, or am i wrong? You also categorically rejected some factual stuff presented by me which were not in line with your current understanding regarding this topic.

I personally am hesitant about sharing my proprietary stuff, as most of the good, unique, innovative things here end up in the wrong hands being sold, or as part of a commercial product, or in hands of Chinese developers. Once posted here the material is owned by this site's owner, the way i understand it.

Everything one needs for basic measurement is already on the web, readily available.

 

[kingkorg]

you can "hopscotch to anechoic" if you're only doing comparative measurements. if you're testing variations on one capsule, like different k67s, all you need to know is the difference between a reference k67 in your environment and an anechoic chamber and you can subtract that out instead of trying to use a flat omni mic in your environment and potentially introducing reflections to your reference curve. because so many of the differences in performance between designs are due to external dimensions of the capsule and the remaining differences happen mostly off-axis, as long as you're only taking 0deg measurements at far enough away, the differences are negligible if you have a good reference for a capsule of the same size and similar construction

To your point:
https://repforums.prosoundweb.com/index.php?topic=25011.0

[quote author="maxwall"]Would like to know a little more about the base or body containing the electronics that these heads were installed on. I'm assuming all stock U47 components and power supply.

Martin
[/QUOTE]

 

[soliloqueen]

great graphs. i would peg the gefell as closest here, mount differences not withstanding. wonder what's causing the high end peaks in all of them?

 

[kingkorg]

great graphs. i would peg the gefell as closest here, mount differences not withstanding. wonder what's causing the high end peaks in all of them?

My guess is that original Neumann used as a reference here has some aging going on, possibly some stiffening, debris, which causes this very hf roll off on it. Maybe even very thick diaphragm. I don't think that difference would be present on a straight out of factory capsule.

Differences below say 10k would point to difference in assembly/construction. Gefell 7-8k peak is certainly due to mounting/position issue. Mess 800hz-1k with Dale's is not the capsule, but either noise that slipped through, or resonance of the body/tube/headbasket.

Thiersch PVC and Wagner are spot on, i wouldn't care about that very high frequency boost, it is not really a boost, just less roll off compared to the original, which would go away with time. Thiersch is a bit tilted, so probably overall "warmer" sounding. This is small deviation, well within production tolerances for a M7.

Edit:
Kudos to @Martin B. Kantola for not smoothing the responses, otherwise these small intricacies might have been lost. There is no reason to smooth out the measurements no matter how rough they might seem they are useful just the way they are. Unless you want to sell or hide something, or present your measurements looking like the big guys. Which are totally misleading, and why people don't find them usefull.

 

[Андрій Чубака]

I don't know about other people. But when i am comparing some sound from different mic i have my own opinion about quality of sound. Coloration of sound based not only on frequency response graph of mic. But also there is available some frequency of fet to every part of sound on graph. So for example 2 mic with about same graphic of frequency response can sound different with some different coloration of sound from fets or tubes. Some sound is smooth some is not so. And for every frequency on graph this smoothness can be different.

 

[kingkorg]

I don't know about other people. But when i am comparing some sound from different mic i have my own opinion about quality of sound. Coloration of sound based not only on frequency response graph of mic. But also there is available some frequency of fet to every part of sound on graph. So for example to mic with about same graphic of frequency response can sound different with some Coloration of sound from fets or tubes.

Which is why there are other measurements not related to frequency response.

 

[Андрій Чубака]

I am still don't feels like a pro with this mics and mics building. But i hope in some day i will do some good mics. And i don't feel like is measuring equipment is really important for me with my building and using possibilities. All my projects is low costs. And i don't think like i need something really more here.

 

[kingkorg]

I am still don't feels like a pro with this mics and mics building. But i hope in some day i will do some good mics. And i don't feel like is measuring equipment is really important for me with my building and using possibilities. All my projects is low costs. And i don't think like i need something really more here.

You can absolutely build marvelous sounding microphones without making a single measurement. If it sounds good, it's good.

But if you go for cloning certain mics, i have no idea how anyone would do it. Maybe build kits that come with pre made, certified measurements. Of which there are none as far as i know. All focus mostly on esthetic.

 

[MicUlli]

Why not share your findings and approaches regarding measurement to start with? This way, it seems to me, you are out fishing for other’s knowledge for personal (potentially commercial?) use.

To this day i'm not really certain about your intentions regarding mic stuff. You have shared some projects which i am very thankful for, but you also mentioned developing some commercial products, or am i wrong? You also categorically rejected some factual stuff presented by me which were not in line with your current understanding regarding this topic.

I personally am hesitant about sharing my proprietary stuff, as most of the good, unique, innovative things here end up in the wrong hands being sold, or as part of a commercial product, or in hands of Chinese developers. Once posted here the material is owned by this site's owner, the way i understand it.

Everything one needs for basic measurement is already on the web, readily available.

As i mentioned in the thread start i will do. And yes, i am interested in widening my knowledge, but not for commercial use. Believe me, there are a lot more attractive opportunities to earn money for a development engineer..
My intention is to help hobbyists making their microphone baby better. And sometimes it is better to leave someone's oppinion uncommented instead of to convince him.
I fully understand that you dont want to share your deep knowledge, but nobody forces you.
If everything regarding measurement is already on the web, what is wrong to collect and comment it in this thread?
BR MicUlli

 

[Roman Beilharz]

In addition to measurements I am re-recording identical, self recorded HQ audio signals that I play back over a mastering grade speaker system in my acoustically treated/optimized control room using different mics and build configurations in identical setups (height, axis/angle, distance) to select my build's components. I consider this to be much more insightful (is that a word?) for me as a sound engineer/producer and musician than any measurement I know of.

The problem with real acoustic performance miking/recording comparisons on the web is, that they are never standardized in terms of mic positioning, performance differences, (unknown?) circuit/component/cabling/preamp/ADC differences and so they are never repeatable. Forget about the classic we-put-5-mics-in-front-of-same-singer-approach: the inevitable different angles/distances/heights render the audio results unusable. The closest 0-degree-axis mic will usually sound best. Yay!

I am aware of the fact, that even the best speaker system doesn't generate sound the same way as a singer/instrument would. But as my results are based on a standardized setup, I can much better draw helpful real-life conclusions. I know I have used expensive high-end equipment, but I can imagine that it might not even be such a bad idea to replicate my notion with whatever-you-own equipment, because it should be able to tell you pretty well, how differences in builds will be noticed within your own overall environment.

That's what I did to achieve this:

1. I recorded female and male vocals, whispering/speech, a viola and a Djembe with the most natural high-end front-end chain in my studio in 96kHz/24 Bit (Josephson C617 SDC, JZ pop filter, Vovox Sonorus direct-S cable, Gordon Electronics Model5 preamp, RME Octamic XTC ADC).

2. I chose a default distance to one of my Sky Audio Verdade speakers (KS digital DSP corrected B8 sub, 2 PSI AVAA active bass traps, SPL Performer amp), that is far away enough for any musical signal to sound "complete" while still being close enough to minimize room reflections and induce a noticeable proximity effect - like a real singer/instrument would. I ended up at precisely 30 cm between tweeter and capsule diaphragm at a well defined height. I then verified the linearity and checked the phase response of this position with a calibrated measurement mic (Sonarworks omni, REW). Levels were set in a way, that 1KHz sine presented @70 dB SPL created -18dB at the ADC output. This measurement served as reference to divide any testing mic measurement by and pretty much nulls out the effects of room and speaker this way.

3. I then recorded all pre-recorded audio signals through all mics* and test builds to compare to each other using a slightly higher leveling (1KHz sine@70dB SPL creating -16dB@ADC), because this seemed closer to levels I would use in an actual recording session.

*Microtech Gefell M930, SONY C100, Neumann U87, Neumann KM86, Warm Audio 47jr, AKG C414, Rode NTR, Josephson C617, U47 style build (M7, EF86) with Replica Microphones (RM) and Thiersch M7 (STW7 red line), Telefunken NOS EF86, EF806S and JJ EF86 tube, RM and Haufe BV08 transformer (to A.Grosser's specs), ERO MKT cap vs. Mundorf cap etc.

Here are some of my findings:
- testing my U47 style build (M7, EF86) the Haufe/Grosser BV08 transformer and the Thiersch mylar M7 sound superior to their Replica Microphone clones, but the differences are smaller than expected
- at first I could not distinguish between the JJ EF806S, NOS Telefunken EF86 and EF806S - later on I found the JJ to sound softer and "sweeter", while the Telefunken gave me a bit more "whack" and faster transient response
- my U47 style build blows away all other mics when it comes to female vocals and is great for any vocal, while newer capsule designs have more whack when it comes to drums (like M930/K47 style capsules).
- for the Viola, the Neumann KM86 is hard to beat
- the SONY C100 is the only modern low-noise mic, that comes close to that classic Thiersch M7/U47 sound, with a little less air/"satin" in the top end.
- at least when you have a decent room (or you are using an absorbing rear screen/filter): close miking in Omni rocks: rounder, fuller, less strained, don't need proximity bass for most of the sources. My build sounds best in omni or hypercardioid and so do most of the other multi pattern mics in the race (C100, U87, KM86).
- a frequency response graph doesn't tell you anything musically useful about a microphone; time-domain and distortion are much more relevant, but hard to measure and even harder to interpret (IMO a waste of time).

When I am done with all testing, I will structure and render my audio to share with the community, if you like.

Cheers!

Ro

 

[kingkorg]

@Roman Beilharz congrats, this is exactly how i started my exploration, and is still part of my routine. Audio Test Kitchen does the same.

I would suggest recording pink noise as well, although not very musical it is very telling, and usefull if you for any reason might want to match two totally different mics.

I could not distinguish between the JJ EF806S and a NOS Telefunken EF86, so either they are both great or tubes are being overvalued when it comes to sound

This is exactly what i found, and you might continue to come to similar conclusions in future regarding many different components. What is lacking though in this approach is the high spl component which could show some differences when components start operating at their limits.

- a frequency response graph doesn't tell you anything musically useful about a microphone; time-domain and distortion are much more relevant, but hard to measure and even harder to interpret (IMO a waste of time).

This is where i would disagree just because the point you brought earlier. The publised measurements are not standardized and often smoothed. This leads you to believe they are useless simply because they can't be really compared. Well executed measurements are priceless, but rare. At this point, since i've been performing all of my measurements under same conditions for years, i dont need to hear a microphone, i know exactly how it will sound after i'm done with the measurements. Of course, this is me, other people need explanation regarding the measurements, so i do my best to provide them.

Great example of this would be well documented thread of Soliloqueen's flat K47, which i loved the second i saw the curves show up on my screen, and then the listening test confirmed it. What I didn't expect was everyone else to go crazy about it.

I do use my ears, and there are exceptions where FR is terrible by all means, but microphone finds it's use anyways. Great example would be sm57, or Ear Trumpet Labs which use cheap capsules, and esthetic focused design that gives you quirky performance. People seem to love them though.

One thing to keep in mind, performance and sound of microphones and monitors will change day to day due to temperature and humidity, and God knows what else! Often not by small margin. So it is a good idea to "shootout" and measure mics in one go.

 

[kingkorg]

As i mentioned in the thread start i will do. And yes, i am interested in widening my knowledge, but not for commercial use. Believe me, there are a lot more attractive opportunities to earn money for a development engineer..
My intention is to help hobbyists making their microphone baby better. And sometimes it is better to leave someone's oppinion uncommented instead of to convince him.
I fully understand that you dont want to share your deep knowledge, but nobody forces you.
If everything regarding measurement is already on the web, what is wrong to collect and comment it in this thread?
BR MicUlli

Everything you need regarding basic and advanced measurement techniques can be found on these two channels. I, selfishly, am hesitant compiling them in a simple thread, and servin them on a plate because of earlier mentioned reasons. Probably two most underrated YT channels ever.

https://m.youtube.com/@ListenInc
https://m.youtube.com/@AudioScienceReview

 

[godzich]

I would be interested in knowing what kind of sound source could be used? Probably a as flas as possible full range small high quality speaker? Just the manufacturer part number is missing


I don't build mics, but repair them now and then. I have an ATS-2 to perform the test with enough knowledge to build a small anechoic chamber, only the as frequency-flat as possible and affordable sound source is missing... If anyone has any good ideas or hints on what to use, please chime in

Christian

 

[kingkorg]

I would be interested in knowing what kind of sound source could be used? Probably a as flas as possible full range small high quality speaker? Just the manufacturer part number is missing


I don't build mics, but repair them now and then. I have an ATS-2 to perform the test with enough knowledge to build a small anechoic chamber, only the as frequency-flat as possible and affordable sound source is missing... If anyone has any good ideas or hints on what to use, please chime in

Christian

Some googling wouldn't hurt. You can use almost any speaker, the imperfections of the speaker are compensated by a known flat reference microphone.

 

[Roman Beilharz]

@Roman Beilharz congrats, this is exactly how i started my exploration, and is still part of my routine. Audio Test Kitchen does the same.

I would suggest recording pink noise as well, although not very musical it is very telling, and usefull if you for any reason might want to match two totally different mics.

Alright, I will include it. What exactly are you reading/concluding from the way pink noise is being transformed into AC by the mic?

This is exactly what i found, and you might continue to come to similar conclusions in future regarding many different components. What is lacking though in this approach is the high spl component which could show some differences when components start operating at their limits.

Interesting; so do you think tubes are overvalued in terms of their contribution to the sound? It might depend on the very circuit you are testing, I guess. Things tend to be influenced by many factors at the same time. There are so many pseudo-religious discussions going on on the web about some components that I wanted to hear it for myself.

Very good point, though, about the high SPL sound sources. I will think about including some higher SPL test rounds and check the differences. Well, I have to stop somewhere and it has become a tremendous amount of work, already, but...

This is where i would disagree just because the point you brought earlier. The publised measurements are not standardized and often smoothed. This leads you to believe they are useless simply because they can't be really compared. Well executed measurements are priceless, but rare. At this point, since i've been performing all of my measurements under same conditions for years, i dont need to hear a microphone, i know exactly how it will sound after i'm done with the measurements. Of course, this is me, other people need explanation regarding the measurements, so i do my best to provide them.

Well, I have precise measurements, but I can still hear differences between build configurations that look the same to me. But I am sure if someone like you can read and interpret the measurements better than I can, that will be useful. Only for the common engineer/hobbyist/musician (thread starter quote "poor (wo)mans mic...") I think going the re-recording way allows for real-life conclusions, fast. Happy to learn from you, by the way. I will check out your links.

Great example of this would be well documented thread of Soliloqueen's flat K47, which i loved the second i saw the curves show up on my screen, and then the listening test confirmed it. What I didn't expect was everyone else to go crazy about it.

I do use my ears, and there are exceptions where FR is terrible by all means, but microphone finds it's use anyways. Great example would be sm57, or Ear Trumpet Labs which use cheap capsules, and esthetic focused design that gives you quirky performance. People seem to love them though.

One thing to keep in mind, performance and sound of microphones and monitors will change day to day due to temperature and humidity, and God knows what else! Often not by small margin. So it is a good idea to "shootout" and measure mics in one go.

Interesting as well and yes, I tested in bulk, but had to take breaks and work in between. It's a great adventure in any case and I have learnt so many things, already.

Punch out

Ro

 

[kingkorg]

Alright, I will include it. What exactly are you reading/concluding from the way pink noise is being transformed into AC by the mic?

Kind of like with distorted electric guitars, it is extremely easy to hear differences in frequency response simply because all the frequencies are there the whole time and you get all of them. With vocals for example, it can be tricky because a lot of HF where most the differences are found are transient in nature, and short lasting. I like drum tracks as well as i get long tail of cymbal decay.

I use these takes with Eq match software (CurveEQ, Fab Filter) to match similarly constructed microphones if i need to. Say u87 and a 100$ k67 based mic with similar headbasket. You would be surprised how close you can come. If i suddenly need a matched pair, this helps big time.

Interesting; so do you think tubes are overvalued in terms of their contribution to the sound?

To a certain extent. A lot of mythology spills from guitar amp world. However, here we use tubes in pretty linear region, with for example vocals under 1% thd. So how a certain tube saturates is really not important. It is sometimes questionable if you could really tell the difference between a FET and tube, presuming everything else being the same.

Most of the differences in sound come when you start swapping tubes due to some differences in purely electrical properties. For example plate resistance of a tube can interact differently with following capacitor and transformer. You could get some resonances in the low end, or frequency response could change. Also bias point of the tube could be different. But that is no magic, or mojo. 220pf range capacitor in Elam 251 will shift cutoff point big time depending on internal resistance of the tube. Which can tremendously impact the response of the high end. So certain tube might sound "warmer" than the next one. However no magic here, just stupid change of the frequency response of the circuit.

Well, I have precise measurements, but I can still hear differences between build configurations that look the same to me. But I am sure if someone like you can read and interpret the measurements better than I can, that will be useful. Only for the common engineer/hobbyist/musician (thread starter quote "poor (wo)mans mic...") I think going the re-recording way allows for real-life conclusions, fast. Happy to learn from you, by the way. I will check out your links.

I am anal when it comes to empiricals when i build and discuss gear. How else would you clone a certain mic. The very second i start doing creative work i start thinking intuitively, and forget about every measurement i ever made.