Topic: How accurate are published specs of professional microphones?

[page]

So I see that many companies publish their frequency range, 40hz-20khz being a common one. If a company publishes a number, what happens to (for instance) the information at 39hz, is it partially recorded/rendered, or how does that work? I see information on the resulting frequency analysis both above and below the designated range, but I'm wondering how valid a representation of the sound the mics received this information is (completely discounting how audible it may be, thats a different question).

edit: fixed typo in title.

[page]

Outside of that range you will still get response; it will roll off at some slope.  It would be unusual to have a slope much greater than third-order (18dB/octave) though, and that would usually only be if a low cut filter was engaged (stated separately in the specs, if at all).  Often you'll simply have a 6dB/octave rolloff.

Ok, so I have a set of beyerdynamic 930s. They are rated publicly as 40-20khz. So based on the above would it be safe to assume that they are capturing information around the 30hz range, even if it's attenuated? At what point would one assume if a manufacture (as you point out, an honest one) states a given number, that a ball park underneath that is where you start losing accurate information (regardless of roll off attenuation). For example, if my 930s are rated at 40-20khz, then they could still capture information (attenuated) down to lets say 25hz, but under 25hz it's just some form of noise regardless of the source material?

Another way to phrase the question would be; at what point does it go from being an accurate representation of the sound regardless of volume level to a form of noise (or noise/representation hybrid)?

[JasonSobel]

Page,

Some marketing also comes into play with the published specs.  For example, with the DPA 4021, the official specs are 40Hz-20kHz, +/- 2dB.  now, the DPA's have great bass response, and almost certainly pickup sounds down to 20Hz (or below?).  But DPA chooses to market it's microphones as having a very flat and accurate frequency response.  So it was more important to them to stop the published specs at 40 Hz and stay within +/- 2dB, then to publish a spec that looked like 20Hz-20kHz +/- 10dB (or whatever the actual value is at 20 Hz, I don't really know it).

conversely, the official specs for Neumann km140's are 20Hz-20kHz, without stating how much the frequency response varies within that range.  however, if you look at the frequency response graph, it looks like the response is -12dB at 20Hz.  So Neumann made marketing decision that it was more important to state a 20Hz-20kHz frequency range, but then downplayed the fact that the mics are at -12dB at 20Hz.

Now, the beyerdynamic CK930 mics, like the DPA's, have an official spec of 40Hz-20kHz.  if you look at the frequency response graph, it looks like the CK930's are around -5dB at 40Hz, which is about where the Neumann km140's are at 40Hz as well.  If I had to guess, from looking at the graph (and from my own experience with the mics), I would say that the CK930's are in the -10 to -15 dB range at 20 Hz.  But I certainly don't have the appropriate equipment needed to test the mics myself.  Though I would be very surprised if the response of the mics down at 20Hz, even though it's attenuated by a good deal, is less than the self-noise of the mic itself.  Therefore, I think it's a safe assumption that you're picking up actual, useful sound information down in the 20Hz-40Hz range.

- Jason

[page]

Though I would be very surprised if the response of the mics down at 20Hz, even though it's attenuated by a good deal, is less than the self-noise of the mic itself.  Therefore, I think it's a safe assumption that you're picking up actual, useful sound information down in the 20Hz-40Hz range.

I understand where your headed with it. Thanks.

Actual vs Useful becomes the argument there. Is there anyone of real worth capturing at 20hz. I'm not going to open that can of worms in this thread. I was just curious how much of the stuff under the published threshold of 40hz was signal vs noise (regardless of how much that signal may actually end up being noise when interpreted by humans).

[DSatz]

I've long felt that Beyer's microphone specifications can be relied upon more than those of many other manufacturers. They're a serious engineering organization with a long history of building high-quality professional products. And their spec sheet for the model 930 cardioid shows that it's already 5 dB down at 40 Hz where the printed curve stops. That curve has a stated ±2 dB tolerance (a fact which most manufacturers don't reveal--so again, good on Beyer).

Unfortunately, the measurement conditions aren't stated, and for pressure-gradient microphones the distance from the measurement sound source is crucial. Officially a 0° frequency reponse curve should give the free-field response under "plane wave" conditions (i.e. with none of the bass boost that would be caused by proximity effect). But most central European manufacturers have been publishing curves for many years that are actually based on a one-meter measurement distance. For pressure-gradient transducers, that makes them appear to have somewhat stronger low-frequency response than if the standard were followed to the letter.

Fact of life: By 40 Hz, the response of any cardioid condenser microphone will be in a definite state of decline at miking distances greater than a few inches. At very low frequencies with pressure-gradient transducers you start to get more noise and rumble and energy from mechanical vibration and moving air currents in proportion to the actual acoustic energy that may be present. Thus most high-quality condenser microphones include some low-frequency filtering, which may have its -3 dB point somewhere between 20 and 40 Hz depending on the manufacturer and model.

Finally, the circuitry of any DC-polarized condenser microphone has an inherent 6 dB/octave rolloff below some given frequency which is determined by the capsule capacitance and the value of the resistor through which the charge is being delivered.

--best regards

[page]

But most central European manufacturers have been publishing curves for many years that are actually based on a one-meter measurement distance.

For some reason I thought I found that beyerdynamic was doing their measurements at 1m. That thought's probably lost to history.

Push a 20Hz sine wave through your speakers.  Does that sound like noise?

Fun exercise. I'd done it before, but revisited it with new equipment. I get just about squat at 20hz, a Death Star rumble at 30, a midget Death Star rumble at 35, and then a subsonic kick drum effect around 40hz. I didn't get anything at 19khz, even after slowly turning the gain up near the limits of what I thought was safe/sane for amateur testing.

[page]

It takes a pretty good sized driver to get low lows, 10" is good, 12" is better.  And a big enclosure to suit them.

It doesn't help that I live in a place that sort of requires me to use headphones and is realistically too small for one of these:



 

On a serious note, I do appreciate everyone's insight.

[Will_S]

I didn't get anything at 19khz, even after slowly turning the gain up near the limits of what I thought was safe/sane for amateur testing.

It's quite possible your system was reproducing it, but you couldn't hear it.  Got a dog?  Mine will perk up when I playback 19 kHz (or 22kHz for that matter) even though I can't hear much past 17k.

[Giant_Rick]

A friend of mine once told me he bought a pair of Behringer dynamic mikes.. those one for the meetings and speeches, so nothing really 'technical'.
Outside the box the specifics were outstanding (they were promising some frequencies you cannot even think of! The sticker was put by some clever seller, not by Behringer), but inside there was a paper with the real specs.. worse than the other.

So my humble answer is: it depends on the honesty of the seller. And I think if the seller is smart enough he won't fake anything.

[page]

I didn't get anything at 19khz, even after slowly turning the gain up near the limits of what I thought was safe/sane for amateur testing.

It's quite possible your system was reproducing it, but you couldn't hear it.  Got a dog?  Mine will perk up when I playback 19 kHz (or 22kHz for that matter) even though I can't hear much past 17k.

Yeah, my hearing drops off after about 16.5 or 17k, somewhere in there. My IEMs drop off after 16, but I've got a new set of cans (and a DAC/amp) that supposidly can go to around 21khz. I've also got a scoop in one ear around the 1khz mark for about 4-5db though...  It does explain, or at least provide a convenient cover as to why I don't listen to my S.O. 

side bar: Get your hearing tested and protect it!

[Church-Audio]

I didn't get anything at 19khz, even after slowly turning the gain up near the limits of what I thought was safe/sane for amateur testing.

It's quite possible your system was reproducing it, but you couldn't hear it.  Got a dog?  Mine will perk up when I playback 19 kHz (or 22kHz for that matter) even though I can't hear much past 17k.

Yeah, my hearing drops off after about 16.5 or 17k, somewhere in there. My IEMs drop off after 16, but I've got a new set of cans (and a DAC/amp) that supposidly can go to around 21khz. I've also got a scoop in one ear around the 1khz mark for about 4-5db though...  It does explain, or at least provide a convenient cover as to why I don't listen to my S.O. 

side bar: Get your hearing tested and protect it!

There are no industry standard tests that have been adopted by all manufacturers for testing frequency response of microphones... Its very easy to measure the response of a speaker.. Its extremely difficult to measure the response of a microphone. And do it with accuracy. Companies like BEYER, DPA can be trusted for the response charts they publish others no so much. In the end the ears are always going to be the best judge of how a microphone sounds not a spec sheet. The measurement conditions in a lab will never be replicated in the real world these are ideal conditions guarenteed to produce the best results things like Humidity, distance temperature, altitude.. ALL adversely effect the response of a microphone so in the end the spec sheet does indicate a level of quality but does not reveal how the product will work in the field. That's why some of you like AKG, SHURE, NEUMANN ect. Because you all hear differently and you use mics in different locations that make the use of one mic over another useful. When I build mics I take into consideration where it will be used and try to find the frequency response that suits the situation that's not always flat and perfect.

Chris


Chris