elektret mike amplification

It's usually PVDF, which is basically fluorinated Saran Wrap. You pole it by heating it up, applying a voltage, and stretching it. Two-dimensional stretching is apparently better than 1-D. Lab tools for doing that are called, unsurprisingly, "film stretchers", and cost $100k.

Of course you could probably build a 1-D one out of an old laser printer.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC 
Optics, Electro-optics, Photonics, Analog Electronics 

160 North State Road #203 
Briarcliff Manor NY 10510 

hobbs at electrooptical dot net 
http://electrooptical.net

Plastic ones are both. They're also pyroelectric.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs 
Principal Consultant 
ElectroOptical Innovations LLC 
Optics, Electro-optics, Photonics, Analog Electronics 

160 North State Road #203 
Briarcliff Manor NY 10510 

hobbs at electrooptical dot net 
http://electrooptical.net

It looks like the top surface of the electret is not metalized.

Metal base, electret film, air gap, metal diaphragm. Sound modulates the width of the air gap.

--

John Larkin         Highland Technology, Inc 

lunatic fringe electronics

So are all the microphones-on-a-chip electrets, or are they using other technologies.

Or do you know?

And on a related note, is the US capability to make analog chips going to grind to a halt when you retire?

--

Tim Wescott 
Wescott Design Services 
http://www.wescottdesign.com

I should be possible to terminate the capsule with a very low impedance instead of a very high one, using an amplifier specially designed for that purpose.

The resulting response will have a 6dB/octave slope (higher at high frequencies) which could be corrected by an integrator. If the integrator took the form of a capacitive feedback circuit around the amplifier, this could give a very elegant and relatively simple circuit.

I don't have enough mathematical ability to calculate how this would perform relative to the conventional circuit, but it is something I have been meaning to try one day..

Are you aware that very low noise capacitor mics sometimes use radio frequency energisation? The small variable part of the capacitance appears as a lower impedance (avoiding all sorts of problems associated with very high impedance circuits) and the unavoidable fixed capacitance can be balanced or tuned out in some way.

As usual, if someone thinks they have come up with a good idea for audio, either Blumlein, Baxandall or Walker will have thought of it first. In this case it was Baxandall:

--
~ Adrian Tuddenham ~ 
(Remove the ".invalid"s and add ".co.uk" to reply) 
www.poppyrecords.co.uk

Couldn't the diaphragm be a thin metal foil or metalized plastic? I'd expect that to make more signal and provide shielding.

--

John Larkin         Highland Technology, Inc 
picosecond timing   precision measurement  

jlarkin att highlandtechnology dott com 
http://www.highlandtechnology.com

That's exactly what I propose. They call this a charge amplifier or transimpedance amplifier, I think you could make that more specific a transcapacitance amplifier.

Here is an example I found after posting my OP:

They miniaturised it, but I think you could try the same circuit with discrete components and see how it works out.

I should have put that here: "...you could try the same circuit with discrete components and see how it works out."

Yes, people posted the Sennheieiser MKH 105 here:

Thanks for the articles, I don't see them covering the charge amplifier, however.

joe

}snip{

No. :)

joe

Yep.

Some use gold-sputtered Mylar.

There are nickel diaphragm mics.

--
Les Cargill

Well, there are MEMS microphones, as well.

Old analog engineers don't fade away. They die. ...or something like that.

Most MEMS microphones seem to be condenser types, not electrets.

--

John Larkin         Highland Technology, Inc 

lunatic fringe electronics

** Let me reword, a charge amplifier will not give as good a s/n ratio as a FET buffer stage when used with a condenser mic capsule.

The idea was tried and quickly dropped by the fellow who designed mics for Rode.

But you can try it anytime.

** Gets inside the capsule and causes leakage or just sits on the diaphragm and reduces the output. It's more of an issue with externally polarised ty pes, some of which you dare not breath on let alone take outside on a rainy day like you can with most dynamics and RF types.

The contaminated moisture from human breath has long term effects too as it promotes mould growth and surface corrosion.

... Phil

These charges appear on packing tape, between the layer of stick and Mylar. I made a speaker with front and rear plates, 70 volt transformer, with tape in-between. Not very efficient. If you remove the sticky, the charge vanishes.

Greg

Perhaps one of them tried it and it turned out not to be a good idea :-)

If you get around to building a discrete-component prototype, let us know how it performs, particularly in regard to S/N ratio. There will always be a Brownian Noise floor which you cannot get below, the smaller the diaphragm area, the greater the Brownian noise. I increased the effective area without worsening the horizontal pattern at high frequencies by stacking conventional capsules vertically.

The circuit incorporates a similar idea to yours - except that it is the whole capsule array which is terminated with a low transimpedance, not the individual diaphragms.

--
~ Adrian Tuddenham ~ 
(Remove the ".invalid"s and add ".co.uk" to reply) 
www.poppyrecords.co.uk

Do you have a reason to assume this is the case?

I did not propose a super-low noise amplifier, only one that is 3x better in sensitivity _and_ in S/N-ratio than the one with the original circuit. I still haven't seen any argument convincing me that I'm wrong. That can have two causes of course, either I don't want to be convinced, or the proper argument hasn't been presented yet. :)

I like the biassing, but in my plan the transimpedance is a capacitor, not a resistor. I think that also lowers the noise.

joe

I do capacitive-feedback TIAs fairly often, for just that reason. My other usual approach down below 10 nA is to make a bootstrapped BF862 source foll ower, and then amplify after that. Adding a shunt-feedback PNP wrapped roun d the FET reduces the output impedance and helps linearity.

The key is to not let the bias scheme dominate the noise.

Cheers

Phil Hobbs

That's not how it works. If you have an idea about how something can be done better, it's up to *you* to come up with a convincing demonstration of its superiority.

We're waiting.

Jeroen Belleman

r usual approach down below 10 nA is to make a bootstrapped BF862 source fo llower, and then amplify after that. Adding a shunt-feedback PNP wrapped ro und the FET reduces the output impedance and helps linearity.

I thought I heard that the noise in electrets (and maybe all mics?) is dominated by stuff other than electrical noise.

Jump to the end of this...

it says the noise is mostly acoustic flow impedance... which surprises me.

George H.

Not really, but it wouldn't surprise me to find that it was a possibility they had thought of.

I think you are right about the increased sensitivity because the stray capacity to ground does not take any charge from the variable capacity because there is no overall change in voltage across them.

The question is what sort of amplifier you would need in order to take advantage of the low impedance presented at the input point by the feedback system? I was wondering about a very low impedance one of the type used for moving-coil gramophone cartridges, optimised for an input impedance of around 15 ohms. Alternatively, the NE5534 gives its best noise performance in the region of 600 ohms - but I don't know how to calculate the apparent impedance of the circuit at that point..

It was a way of avoiding any loading of the capsules by the power supply and also removing the L.F. noise coming from the voltage stabiliser, which was greater than the noise contribution of the mic capsules.

It prevents the rise in noise with frequency - but the conventional circuit also does this, because the multi-megohm input resistor is shunted by the capsule capacitance over most of the audio frequency range. The hiss from a capacitor mic sounds much 'softer' in nature than the hiss from a purely resistive noise source.

--
~ Adrian Tuddenham ~ 
(Remove the ".invalid"s and add ".co.uk" to reply) 
www.poppyrecords.co.uk