Coming back on this very interesting subject that’s new to me, the reference for a microphone seems to be 94dBA SPL or 1 Pascal and I assume this sound level has to be generated and recorded by the microphone (at what distance, 1 meter?) ( Single tone, white noise or what?) and its output from 94 dBA compared to its own self noise, which in your case is 74dBA lower, true ?
Self noise from a microphone above 25dBA can only be used in (very) loud surroundings to mask the noise.
To record voices with little noise from a distance, self noise should be at 20dBA or below, meaning 94-20 = 74 dBA S/N.
So your mike is almost suited for demanding studio applications.
The other microphone that you mentioned with 34 dBA self noise could only be used for low level applications like cheap intercoms at the doorbell.
Did I somehow get the very basics ?
What still puzzles me is how this translates from sound into into voltages and noise requirements for the amplifier because this largely depends on the sensitivity of the microphone and the required termination.
Are there any ballpark figures for condensor microphones for these parameters: sensitivity, output impedance and required termination ?
That would enable to get some view on what noise the amp is allowed to produce to possibly stay below mike's self noise.
Hans
You have got it.
As to the amplifier first stage, there is an FET built into the capsule as an impedance converter. So the next stage needs to be reasonably quiet. The real limit would be the noise of the equivalent resistance, about 1000 ohms.
One advantage is the frequency tapering to optimize speech intelligibility lowers the 1/F noise.
Some of the skill in design is not just optimizing the signal levels, but also the frequency response for good speech ineligibility. One issue frequently missed is that of the subsonic boom from lightning. We only hear reasonably close lightning, but if you analyze the spectrum, there is an enormous amount of energy we don’t perceive, but an improperly designed limiter can not properly deal with it.
As the microphones often feed the 12 bit or less resolution of the audio input to a TV camera compression or limiting is required.
The other practical limit is the microphone system must fit in a standard electrical box. I use two circuit cards of 4 layers to fit everything in with enough spacing to avoid self interference.
I did build a small RF test chamber to check for interference signal ingress. I was pleased to see mine worked noticeably better than the other choices. Although my little RF sweep oscillator only goes up to 450 MHz.
As to the amplifier first stage, there is an FET built into the capsule as an impedance converter. So the next stage needs to be reasonably quiet. The real limit would be the noise of the equivalent resistance, about 1000 ohms.
One advantage is the frequency tapering to optimize speech intelligibility lowers the 1/F noise.
Some of the skill in design is not just optimizing the signal levels, but also the frequency response for good speech ineligibility. One issue frequently missed is that of the subsonic boom from lightning. We only hear reasonably close lightning, but if you analyze the spectrum, there is an enormous amount of energy we don’t perceive, but an improperly designed limiter can not properly deal with it.
As the microphones often feed the 12 bit or less resolution of the audio input to a TV camera compression or limiting is required.
The other practical limit is the microphone system must fit in a standard electrical box. I use two circuit cards of 4 layers to fit everything in with enough spacing to avoid self interference.
I did build a small RF test chamber to check for interference signal ingress. I was pleased to see mine worked noticeably better than the other choices. Although my little RF sweep oscillator only goes up to 450 MHz.
Last edited:
Disappointed with your microphone video Ed. I've watched the programs, the correct way to tell an SM58 from a copy is to throw them both off the roof. The one that still works is the genuine one! Oh and you must have spent ages in post processing getting the white balance to change that much between takes
Hope you new mic range is profitable and can't be disabled with a grape like in the bond movies!
Have an array of perforations in the top plate, only one of which is the actual sound inlet. Make a patentable capsule holder so that during assembly, each unit gets constructed with the mic in a different random spot. The attacker will run out of grapes...
Or, arrange them in a way that draws the eye to a particular spot, which is the wrong one. Or (harder) the perfs are fake and the sound intake port is actually somewhere else.
The video didn’t do all that well. As usual it seems everyone knows more than me!
I just got the notice my next 5,000 capsules are shipping to me this week. I actually have lost count on how many microphones have already shipped! Last order to ship was about 900 units. Current orders are 235 from a backorder and 1,100 from a new tender. Also expecting one for 5,000 units.
But all is not roses, was putting together 100 or so today and realized just before boxing them up I forgot the first step! Have to take them apart and insert a “wind screen”. (It actually serves a different very important function but that gets into secret sauce.)
I'm sure grape refers to any POS someone could cram into the perfs. Of course, if these are customarily installed well out of reach... I was imagining a light switch, AC outlet, keyswitch, area volume control, IR motion sensor, etc to be installed in the adjacent box. Guess you'd have dual width steel covers for those if that was the case!
Microphone self noise is an issue but the "normal application if the ECM mikes is close range (like an handset) and the peak is more of an issue at 110-120 dB. The 94 dB reference is something that that industry decided on as a reference but its a reasonable one. And its the SPL of the field around the mike capsule that the reference and sensitivity is references to. Here is a typical spec for probably the best vendor: https://primomic.com/pdf/EM273.pdf Note that the SNR is a weighted and referenced to 94 dB SPL as is the sensitivity. The screen Ed mentioned has several functions. There is a whole industry around woven acoustic screens. Some are even waterproof.
A Primo mike takes less then 0.5 cm3, 500 less then a quarter of a liter, so including packing material easily fitting in a bottle of milk 😂
The specs are disclosing much of what I was looking for, very impressive.
Unfortunately they are still much too large to be used in a hearing aid, having a miserable performance.
Hans
The program I saw had a Shure employee chucking the SM58 off the roof of one of the shure offices. i assume that has done the rounds? The rest of the video was fascinating. I've seen the abuse SM's take on stage and keep working.The video didn’t do all that well. As usual it seems everyone knows more than me!
Pretty sure it didn't do anything for Roger Moore either but stuck with me as a kid.
Now I’m starting to understand why sound quality of hearing aids is so disappointing.
With the FR of a jojo this isn’t exactly studio quality.
Hans
Attachments
The balanced armature looks to be tuned to typical hearing loss. The fixture is not necessarily going to give a flat response so be careful about making judgements. I suspect most hearing aids today use MEMs mikes. They are much more uniform and consistent than ECM mikes. https://www.knowles.com/docs/defaul...eet-rev-g.pdf?Status=Master&sfvrsn=76bd77b1_4 You can see the resonance at supersonic frequencies. They have help info for the external acoustics to control that.
One of the more successful roadies who hangs around here considers the SM58 as an almost perfect microphone. Most likely it is because it does hold up well and he is very experienced with the actual performance hints and kinks.I always thought the SM series sold as much on beatings proofness as anything else? Like when ford did the 'build ford tough'... at least as long as the saddle tanks didn't get hit in an accident
He is one of those who don’t seem to be showing signs of heating damage. FOR SOME OF YOU I SAID HEARING DAMAGE!