Microphone sensitivity and mV/Pa to dB

[cashhewn]

Hi people of the SoS forums, hope you’re all well.

As Hugh mentioned last time I posted, so many numbers and so little understanding …

I have two different models of microphones made by the same legacy German manufacturer that I’m simultaneously “shooting out” in order to compare vocals on and to generally make sure are working properly. They are both being amplified by two separate but identical preamplifiers.

I’m noticing that in order to get the same level coming into the DAW, one preamp needs to be set to just about exactly 10dB more gain (if I’m recalling correctly).

Is there a way to calculate via the published specs on sensitivity that this discrepancy in preamp gain is just about correct? Like mV/Pa sensitivity to dB gain in the preamp?

Or is it not so simple?

Here are the specs:

Mic 1:
Max. SPL
127 dB
S/N Ratio
82 dB
Sensitivity
28 mV/Pa

Mic 2:
Max. SPL
125 db
S/N Ratio
74 db
Sensitivity
8 mV/Pa

Thanks for your help!

[Hugh Robjohns]

It's incredibly simple.

The only relevant specs for matching output levels are the sensitivity numbers. The higher the sensitivity the larger the output level for a given sound source.

When determining sensitivity the acoustic reference is normally 1 Pascal (Pa) of air pressure which equates to a sound pressure level of 94dB SPL.

With that sound pressure acting on the mic's diaphragm, the generated output voltage is measured in milliVolts, hence the sensitivity being given as mV/Pa. Your two mics are specified as 28mV/Pa and 8mV/Pa.

To calculate the difference in output level in decibels, you use the standard formula:

dB = 20 x Log (v1/v2)

So, 28/8 = 3.5

Log 3.5 = 0.544

20 x 0.544 = 10.881dB

Thus, to perfectly match the output level of the two mics, the quieter one needs to be amplified by 10.9dB more than the louder one.

Your estimate of 10dB is very close....

[cashhewn]

Hugh,

How lucky (and grateful) am I (and hopefully the rest of us) to have such an excellent and humble teacher at our service.

Thank you so much.

[cashhewn]

(and of course, I say this as a very happy subscriber to both the digital and print editions of SoS)

[Hugh Robjohns]

[UCLAPost]

OK, only because Hugh strikes me as brilliant, I'd like to know how to convert "old timey" values to modern ones. Specifically, I'm looking at the user guide (in German of course so I'm using Google translate)) for the AKG C12A (the one released in 1962 to "replace" the C12). It says:

Idle Transfer Factor at 1000Hz and 200 Ohms Impedence
0.4 mV/µbar = -68 db (related to 1 V/dyne/cm2)

Sound limit pressure (Max SPL) is 117.5dB

Any way to convert that to Sensitivity mV/Pa and derive the S/N Ratio @ 94 dBSPL?

If there's other values of relevance to this, here's the .pdf I'm using.
https://coutant.org/akgc12a/akg_c12a.pdf

Thank you oh supreme intellect.

UCLA Post

[Hugh Robjohns]

0.4 mV/µbar = -68 db (related to 1 V/dyne/cm2)

The correlation is that 1Pascal = 0.00001 bar, or 10µbar.

So 1mV/Pa is the same as 1mV/10µbar which is therefore 0.1mV/µbar.

In other words, there's a factor of ten between them. So 0.4mV/µbar is the same as 4mV/Pa.

If that sounds a low sensitivity for a capacitor mic, it is. The C12A was a low output mic by design. To put it in a real world context, it's about 10dB hotter than an SM57 would be, and about 12dB quieter than a typical modern capacitor mic. So you'd need a good high-gain preamp.

If you want to check other vintage mic specs, there's a handy page of ancient mic spec converters here: https://sengpielaudio.com/calculator-transferfactor.htm

[UCLAPost]

You are indeed awesome! Thank you.

[ef37a]

By 'eck! Dynes! That takes me back to the early days of Hi Fi News, Studio Sound, Wireless World and such leading lights as Rex Baldock and Hugh Ford.

One thing to note about older microphone specifications is that they usually gave the sensitivity at the equivalent of 70dB SPL and so will appear less sensitive than their modern brethren. 70dB is in fact much closer to normal speech whereas 94dB is pretty loud! Makes the figures look better for the adpuff dept' though!

Dave.

[Hugh Robjohns]

The sound pressure reference level was 74dB SPL (not 70) and was, as you say, a good reference for normal speech.

It came about because back when dinosaurs roamed, and Dave was a wee lad, we used imperial measurements and the CGS system (centimetre, gram, second). And the standard physics pressure reference at that time was 1 Barye, but more commonly seen in documents as 1 dyne/cm^2, or the equivalent of 1 microbar (now 0.1Pa). A dyne is defined as "1 gram accelerated by 1cm per square second".

When those pesky Frenchies ( ) foisted Système International d'Unités upon us, the Pascal became the new pressure reference. The Pascal is defined as one newton per square metre and is the equivalent of 10 microbars or 10 dynes/cm^2.

And that x10 factor equates to 20dB in sound pressure levels, hence the new reference being 94dB SPL.

[ef37a]

The sound pressure reference level was 74dB SPL (not 70) and was, as you say, a good reference for normal speech.

It came about because back when dinosaurs roamed, and Dave was a wee lad, we used imperial measurements and the CGS system (centimetre, gram, second). And the standard physics pressure reference at that time was 1 Barye, but more commonly seen in documents as 1 dyne/cm^2, or the equivalent of 1 microbar (now 0.1Pa). A dyne is defined as "1 gram accelerated by 1cm per square second".

When those pesky Frenchies ( ) foisted Système International d'Unités upon us, the Pascal became the new pressure reference. The Pascal is defined as one newton per square metre and is the equivalent of 10 microbars or 10 dynes/cm^2.

And that x10 factor equates to 20dB in sound pressure levels, hence the new reference being 94dB SPL.

Ah, I did think it might be 75dB, long time ago Hugh.

Dave.

[Wonks]

That triceratops who gave you the advice knew nothing!

[cashhewn]

Resurrecting this one:

Using an AKG C414EB as an example (according to my research a C414EB sensitivity is 6mV/Pa or -44dBv re 1V/Pa), is it correct to say that in order to get an output of +4dBu from the microphone one would need to give the signal 45.79 dB of mic pre amplification?

Many thanks to the real knowers of maths and engineering...

[Hugh Robjohns]

The output signal voltage from a microphone depends on both the microphone's sensitivity and the sound pressure level (SPL) reaching the mic's diaphragm.

The usual reference for sound pressure level these days is the Pascal, which equates to 94dB SPL. Other references have been used historically, such as the Dyne and micro-bar.

Mic sensitivity can be stated in several ways, too, but is most commonly denoted today in milliVolts per Pascal, or mV/Pa — 6mv/Pa in the case of the C414EB.

An alternative form, often favoured by American manufacturers is to state decibels relative to 1Volt for 1Pascal, or dBV/Pa — slthough they usually neglect to include the pressure reference which is unhelpful... so they often just state a dBV value and you have to assume that's for 1Pa or 94dB SPL... The C414EB is specified as -44.4dBV/Pa.

So, if 1 Pascal of air pressure acts on the mic diaphragm, the C414EB will generate 6mV of signal level, or -44.4dBV.

But you want the output in terms of dBu (so relative to a reference voltage of 0.775V instead of 1V). The dBu value is around 2.2dB higher than the dBV value, so the mic's sensitivity is -42.2dBu... or your +4dBu target is +1.8dBV.

It doesn't matter which reference you use to calculate the gain as long as you are consistent and use the same format throughout the maths.

If you want that 1Pa of sound pressure to deliver +4dBu then you would need (42.2+4=) 46.2dB of gain from your preamp.

Or, if you want to stay in dBV, 44.4+1.8= 46.2dB of gain.

I'm not sure how you arrived at your 45..79dB figure... but perhaps you confused dBV and dBu values?

However, this nominal +4dBu output will change if the SPL reaching the mic changes.

For example, if you're recording speech at, say, 74dB SPL (0.1Pa), you'd need 66.1dB of preamp gain to achieve +4dBu at the output. And if your source was generating 100dB SPL you'd only need 40.2dB of gain.

You can work this out with standard maths easily enough, but various online calculators and tables are quicker for solving this kind of problem. My personal preference is for the Neumann Recording Tools app for smartphones.

https://www.soundonsound.com/reviews/ne ... ding-tools

[cashhewn]

You’ve gone above and beyond in this thorough response, as you usually do Hugh, thank you.

And thanks so much for the Neumann calculator tip!

[cashhewn]

Sadly it seems that Neumann app is no longer available…at least I can’t find it.