[bxxk]

Re: Cross-Spectrum Microphone Calibration Service

Hi Herb,

I've built some mics before, including an ambisonic tetrahedral mic, and used a B&K reference mic to manage the calibration. But I would not mind my own for my home DRC work... I'm sure I'm not alone in wondering what you would recommend in various price ranges for a cheap or inexpensive reference mic.

The reason I'm asking is because the ECM8000 is not all that good (eg the bass roll off, which is not in the original Panasonic WM51a capsules, so is possibly in their preamp), even for the price - I believe the similar NADY is better and I've seen home made units which are better. Whatever mic is used will still need calibration, but using a better (flatter / quieter) mic to start with is usually a good idea.

Many thanks for your input, it is much appreciated.
BK

 

[Anechoic]

Re: Cross-Spectrum Microphone Calibration Service

Hi Herb,

I've built some mics before, including an ambisonic tetrahedral mic, and used a B&K reference mic to manage the calibration. But I would not mind my own for my home DRC work... I'm sure I'm not alone in wondering what you would recommend in various price ranges for a cheap or inexpensive reference mic.

It's kind of a broad question that's difficult to answer, but I'll try.

In the same (rough) price range as the ECM8000 I've measured the Nady CM 100 and the dbx RTA-M. The Nady is flatter to 20 kHz although I didn't look much lower. The RTA-M rolls off a bit above 20 Hz.

If you're willing to spend a couple of hundred dollars, the Josephson and Earthworks lines are great mics for the price, but pricewise are geared for the serious enthusiast. The Beyerdynamic MM 1 has a great reputation and can sometimes be found for under $200, but I've never personally handled one. There's the APEX220 which falls into the ECM8000 price range and may or may not be a rebadged RTA-M.

One microphone that I really like is the Radio Shack lapel microphone. They're cheap, but they are amazingly flat up to 20 kHz and much more consistent from unit to unit. The downside is that the noise floor is pretty high (around 60 dBA IIRC) which limits its use for low SPL testing. You also have to find a way to mount it such that the mounting doesn't muddy up the signal. Caveat: it's been several years since I last measured one so it's entirely possible that there has been a revision and the mic is now.

 

[Dennis H]

Re: Cross-Spectrum Microphone Calibration Service

I think the Behringer and the Nady are both OEM'd from Superlux in China -- probably the same mic with a different label. They seem to have lost their flat response when Panasonic discontinued the WM-60 and everyone switched to the WM-61. I thought it was a Superlux problem but I recently read a paper where someone built 8 mics using the WM-61, 4 with the stock wiring and 4 with the 'Linkwitz mod'. They all had the 10K hump although the ones with the mod didn't hump quite as high. The ones with the mod also had lower distortion at high SPL.

Herb, the Panasonic electret capsules don't really require a preamp like you describe for a traditional condenser mic. They work fine with just a battery and a resistor as long as they are feeding a fairly high impedance preamp or sound card. Superlux is including more of a preamp built in to make them compatible with a pro mic preamp with low impedance and 48V.

I was looking at some measured 5-48K curves for the Behringer and it looks like the tails are more like 24dB/oct. The tails don't really matter for someone measuring a room but they could make a difference for someone designing a crossover because of errors in the calculated phase using a Hilbert transform. As an experiment, I built a fake cal file extending down to 1Hz and ran a Hilbert transform to get the phase. Then I truncated the file to 20Hz and ran the HBT again. The phase error at 20Hz was about 70 degrees with the truncated file.

 

[Anechoic]

Re: Cross-Spectrum Microphone Calibration Service

I think the Behringer and the Nady are both OEM'd from Superlux in China -- probably the same mic with a different label.

There was a discussion a few years back on Usenet (Google groups search is completely fubar'd right now so I can't find it) where there was some dispute over that - I think the mic element was probably the same but the electronics are different. The ECM8000 electronics have changed a couple of times over the years so if may be that they're the same now.

I was looking at some measured 5-48K curves for the Behringer and it looks like the tails are more like 24dB/oct. The tails don't really matter for someone measuring a room but they could make a difference for someone designing a crossover because of errors in the calculated phase using a Hilbert transform. As an experiment, I built a fake cal file extending down to 1Hz and ran a Hilbert transform to get the phase. Then I truncated the file to 20Hz and ran the HBT again. The phase error at 20Hz was about 70 degrees with the truncated file.

You have to be careful with using Hilbert Transforms on condenser mic freq response data since condenser mics aren't minimum phase.

 

[Dennis H]

Re: Cross-Spectrum Microphone Calibration Service

You have to be careful with using Hilbert Transforms on condenser mic freq response data since condenser mics aren't minimum phase.

Hmmm, interesting. I'm no expert but I can't imagine a mechanism where a mechanical transducer and analog electronics wouldn't be minimum phase. There have been a lot of discussions about that with speaker drivers, cone breakup modes, baffle diffraction, etc. and it always turns out (as far as I know) that it is behaving as a minimum phase system even though it doesn't appear that way at first. BUT.... the HBT is making assumptions and it's subject to computation errors if you don't do things like extending the tails beyond the measured values. You have to do that with speaker drivers and I'd imagine you have to do it with mics as well.

 

[Anechoic]

Re: Cross-Spectrum Microphone Calibration Service

I've heard it repeated by extremely trustworthy sources (including people who have written some of the books on my shelf and have written ANSI/IEC microphone standards) that it is indeed the case, but have never seen a detailed explanation as to how/why.

The condenser mic is a fundamentlally different transducer than a loudspeaker driver- the loudspeaker is a linear motor that can be modeled by 2-D array of point sources on a plane while a condenser mic is a capacitor of varying charge and their mathematical models exhibit very different behavior (for example the response of the condenser mic is independent of frequency below the diaphragm resonance). It is my understanding that dynamic mics and ribbon mics, which basically act as loudspeakers in reverse, are minimum phase but condensers are not.

 

[Anechoic]

Re: Cross-Spectrum Microphone Calibration Service

For anyone interested in purchasing a calibrated ECM8000 or who wants their own mic calibrated:

• Most of my suppliers are out of stock of ECM8000 and won't have a reliable supply until mid-March. I am able to get them one at a time and I am filling back orders but please be aware it may take a couple of weeks to fill new orders. As always, I do take pre-orders and I ship them on a first-come-first-served basis.
• I normally turn around mic calibrations in 48 hours. Please be aware that I will be traveling on business from March 16 to March 23 so any units I receive for calibration after March 11 won't be returned until the week of the 23rd.

 

[Dennis H]

Re: Cross-Spectrum Microphone Calibration Service

I've heard it repeated by extremely trustworthy sources (including people who have written some of the books on my shelf and have written ANSI/IEC microphone standards) that it is indeed the case, but have never seen a detailed explanation as to how/why.

The condenser mic is a fundamentlally different transducer than a loudspeaker driver- the loudspeaker is a linear motor that can be modeled by 2-D array of point sources on a plane while a condenser mic is a capacitor of varying charge and their mathematical models exhibit very different behavior (for example the response of the condenser mic is independent of frequency below the diaphragm resonance). It is my understanding that dynamic mics and ribbon mics, which basically act as loudspeakers in reverse, are minimum phase but condensers are not.

Thanks, Herb. I guess I'll have to take your word for that. Just curious, how could you ever use a condenser mic to design a crossover if it's introducing excess-phase errors? Seems like all the measurement and crossover programs assume the cal file is minimum phase and experience shows they work pretty well.

 

[Anechoic]

Re: Cross-Spectrum Microphone Calibration Service

Thanks, Herb. I guess I'll have to take your word for that. Just curious, how could you ever use a condenser mic to design a crossover if it's introducing excess-phase errors?

About all you can do is trust in your cross-over design theory. That said most pro measurement mics are designed to have a relatively flat phase response up to 15-20 kHz, but that doesn't help you with an ECM8000 unfortunately.

Over the past week I made another attempt to get phase data for my measurement mics with no success. In the case of BSWA, in what I ca only guess is a language-barrier problem, no matter how many times I ask for "phase response" and "phase data" for the MP201, they keep sending me frequency response curves.

Seems like all the measurement and crossover programs assume the cal file is minimum phase and experience shows they work pretty well.

I'll have to try and dig it up, but when I first started doing microphone measurements, I used to give phase results generated by the Hilbert transform. When I learned that condenser mics weren't minimum phase devices, I found a generic freq and phase response curve for a measurement mic. I scanned in the frequency response, ran the Hilbert alogorthim on the data and compared the resulting phase response to the measured curve - and they weren't the same. that little exercise scared me straight.

 

[Ricci]

Re: Cross-Spectrum Microphone Calibration Service

Herb,

Do you know what the useable spl limit for the ECM8000 is? Seems to be about 120db or less?

 

[lovebohn]

Re: Cross-Spectrum Microphone Calibration Service

Herb,

Do you know what the useable spl limit for the ECM8000 is? Seems to be about 120db or less?

Good question!

 

[Anechoic]

Re: Cross-Spectrum Microphone Calibration Service

Herb,

Do you know what the useable spl limit for the ECM8000 is? Seems to be about 120db or less?

I've tested it up to 114 dB with no problems. When I get back from my trip I'll run some tests and see if I can find a definitive limit.

 

[Ricci]

Re: Cross-Spectrum Microphone Calibration Service

I've tested it up to 114 dB with no problems. When I get back from my trip I'll run some tests and see if I can find a definitive limit.

That'd be great. I seemed to have run into some output/ distortion limits with the mic while doing some close mic (2 or 3") low frequency tests. The apparent level that close up to the driver cone was somewhere between 120-125db.

 

[Anechoic]

Re: Cross-Spectrum Microphone Calibration Service

That'd be great. I seemed to have run into some output/ distortion limits with the mic while doing some close mic (2 or 3") low frequency tests. The apparent level that close up to the driver cone was somewhere between 120-125db.

I just ran a quick test comparing the distortion of my one of my reference mics (ACO Pacific 7052, rated to 144 dB) to an ECM8000 at 1 kHz.

At 115 dB, my ref mic gets 0.13% distortion, the ECM8000 reads 0.54%

At 120 dB, my ref mic gets 0.22% distortion, the ECM8000 reads 0.81%

At 125 dB, my ref mic gets 0.46% distortion, the ECM8000 reads 1.37%

 

[Ricci]

Re: Cross-Spectrum Microphone Calibration Service

Thanks Herb.

Overall that doesn't seem too bad, but it looks like you want to keep things under 120db if you can for the ECM8000. Do you think that there would be a difference if the frequency was lower at like 32hz or 40hz?

 

[Anechoic]

Re: Cross-Spectrum Microphone Calibration Service

I'm falling behind trying to cal all those mics I sold last week, but when I get a chance, I'll run the test again at a lower frequency.

 

[Ricci]

Re: Cross-Spectrum Microphone Calibration Service

Thanks again for being so forthcoming with data and help with odd requests and questions. I think this distortion info will be very interesting for a lot of people.

I'm going to go ahead and guess that there will be noticeably higher distortion on the LF test. Won't the lower frequencies cause greater deflection of the mics diaghram for a set SPL?

 

[Anechoic]

Re: Cross-Spectrum Microphone Calibration Service

I'm going to go ahead and guess that there will be noticeably higher distortion on the LF test.?

Tests at 30 Hz:

At 115 dB, my ref mic gets 1.8% distortion, the ECM8000 reads 3.1%

At 120 dB, my ref mic gets 2.95% distortion, the ECM8000 reads 5.0%

At 125 dB, my ref mic gets 4.2% distortion, the ECM8000 reads 7.4%

FWIW, this is a different ECM8000 than the one used for the 1 kHz test.

 

[Ricci]

Re: Cross-Spectrum Microphone Calibration Service

Well...That's eye opening. That could explain some of the things I was seeing. I can only assume that the distortion will get even higher the lower the frequencies get. Looks like you should keep things under 110db when using this mic to remain clean. Not that any one serious about high SPL measurements would be using an ECM8000 in the first place.

Do you think that the ECM will compress the signal somewhat above 120db?

 

[Anechoic]

Re: Cross-Spectrum Microphone Calibration Service

Well...That's eye opening. That could explain some of the things I was seeing. I can only assume that the distortion will get even higher the lower the frequencies get. Looks like you should keep things under 110db when using this mic to remain clean.

That sounds like good advice.

Do you think that the ECM will compress the signal somewhat above 120db?

I think that it depends on the mechanism that is causing the distortion, but my guess would be that you would see clipping rather than compression. I'll have to but the mic on an o-scope and check some waveforms at high amplitudes to see what's going on.