Analog to Digital Converters

[quote=Um bongo;52556]The mic has a given operating range. Believe it or not mics can withstand explosive decompression.

Below 25dB is irrelevant as the mic noise floor is about this, inputs over 110dB are also irrelevant to speech as the aid isn’t going to do anything with them. That gives you 85dB dynamic range, chop it up as you want.

There will be a calibrated level of sensitivity and a phase angle, why would you be introducing distortion under normal operating conditions. It’s not designed to add anything to the signal.[/quote]
Just to add to this, you are absolutely correct, however as per this forum topic, if you have microphones that distort with high spl, anything you do will be distorted.

For that reason you want to use high slope microphones which in general, roll off the low frequencies and will be less likely to distort during high spl environments.

As a side-note, REM’s should be done to add more low frequencies back to the amplification to match targets.

@Um_bongo said:

Try one on a music program.

Your high input levels are doing over the processing and kicking in the OLC (output limiting compression) or you’re overdriving the receiver and getting harmonic distortion which is usually about 10% at the fully driven state.

1 why do you need the sound to be this loud? (it’s damamging to your hearing and the aid)

2 get a quieter hearing aid music program with the processing turned off and the MPO set to the max.

3 don’t wear the aids in this environment - ER plugs would be better.

All of which is nothing to do with the mic.

AND I also have a Guitarist who is happy with the Unitron music program.

No disrespect, but I think your statements have inaccuracies.

The average SPL that a musician works with is 100dB. A singer will usually surpass that volume if they are really belting it out. Relatively speaking to a non-musician, that is quite loud, but to a musician, this isn’t loud at all.

When the hearing instrument on the input side is in distortion, all other signal going to the output stage is going to also be distorted. However it has nothing to do with the output when we are talking about high SPL inputs.

So, when you have a hearing aid that is set up to receive that level of sound, then the output should also be nice and clear with no distortion on the highs or lows.

I’m glad to know you know of the Unitron musician hearing aids. Not many people do.

Cheers!

@HearingAidHelper said:

No disrespect, but I think your statements have inaccuracies.

The average SPL that a musician works with is 100dB. A singer will usually surpass that volume if they are really belting it out. Relatively speaking to a non-musician, that is quite loud, but to a musician, this isn’t loud at all.

When the hearing instrument on the input side is in distortion, all other signal going to the output stage is going to also be distorted. However it has nothing to do with the output when we are talking about high SPL inputs.

So, when you have a hearing aid that is set up to receive that level of sound, then the output should also be nice and clear with no distortion on the highs or lows.

I’m glad to know you know of the Unitron musician hearing aids. Not many people do.

Cheers!

No disrespect, but if you operated any instrument at 100dB Leq Av. all the working day you would quickly deafen yourself. No music instrument runs at this level unless it’s a constant drone - peak and average values are hugely different due to the temporal nature of the music. It’s nothing to do with a musician or non-musician it’s the average weighting of the exposure. Even a drum beat at 130 dB doesn’t have a an AVERAGE SPL over 100dB unless you have a really high BPM.

You point about distortion is quite right: I couldn’t understand why the original question assumed that there would be significant input distortion. However at very high input levels, you’re either likely to get peak clipping occurring with the receiver or the aid electronically implementing the MPO, both of which will sound like distortion as they square-off the peak of the sinusoid.

None taken.

I don’t disagree that precautions should be taken when working in a noisy environment. However when I work with musicians, I treat them differently with respect to considering their livelihood.

I would never tell any musician to stop doing what they do. They will generally ignore the person who says that. Instead I do my best to understand their needs and give them what they want, so that they can continue doing what they do, safely.

In this case, HonkyTonkJohn is using an instrument which is preventing him from working. Plain and simple. I don’t believe he really cares about the technicality of the instrument. Based on his question, I read between the lines and I gave him a working, tested and result oriented answer. Hopefully he can find someone who knows how to work with a musician, and set his equipment up the way he needs.

Hopefully now, HonkyTonkJohn should be able to get back to work! LOL

Cheers!

There is a lot of wrong and misleading information in this thread.

Dynamic range is given in dB, not dB SPL.

Sampling resolution needs to be sufficient to represent the wide range of sound pressures and this range is much wider than the range of sound pressure levels (SPL).

Music has high peak-to-average ratio so high dynamic range is necessary for coding the waveform without distortion.

Music can have very high peak amplitudes at moderate level, it is not necessary to go to a rock concert to benefit from wider dynamic range in sampling.

Decreased hearing dynamic range does not imply that less sampling dynamic range is needed.

@Um_bongo said:

What do you want to know?

The mics are analogue MOSFET, A-D is done at the input stage of the amp circuit.

Modern aids use a pre-amp comparator to observe degradation/feedback in the circuit. If the A-D or the processing destroys the signal too much or you start getting temporal artefacts, the AI will revert the signal to an unprocessed/less processed version.

Quantisation issues aren’t really a problem unless you’re looking for Hi-Fi reproduction and you have near perfect hearing - which you won’t if you are wearing hearing aids.

Speech clarity first, Violin Concertos further down the list. Get a Widex to hear both.

Out of all hearing aids I have used, Widex (music program) is closest in quality to analog hearing aids. But not close.

@arni said:

There is a lot of wrong and misleading information in this thread.

Dynamic range is given in dB, not dB SPL.

Sampling resolution needs to be sufficient to represent the wide range of sound pressures and this range is much wider than the range of sound pressure levels (SPL).

Music has high peak-to-average ratio so high dynamic range is necessary for coding the waveform without distortion.

Music can have very high peak amplitudes at moderate level, it is not necessary to go to a rock concert to benefit from wider dynamic range in sampling.

Decreased hearing dynamic range does not imply that less sampling dynamic range is needed.

Sorry, at least one of those was my bad. The inference of the reduced dynamic range question was based on a recollection of relative and absolute sequential samples - as you say though, it doesn’t have any bearing as you still need to be able to code all parts of a HF signal irrespective of the amplitude.

Point taken about sampling rates: they still have to reflect Nyquist: however if your mic doesn’t do anything above 10Khz you won’t need to sample at much more than 20KHz

@HonkyTonkJohn said:

Wow this is cool! I must say that Hearing Aid Helper understands what I’m talking about! Thank you. Also I must thank um bongo too. I have tried bongos thoughts already, and getting rid of the compression and setting the mpo has put me in the best spot i can be. It still doesn’t solve the problem of the mic being distorted. It is amazing how loud a crowd is when your playing a a honky tonk dance!! So without even playing they already distort the microphone. The only average of spl I have is that my spldb meter (which is a cheap radio shack digital) when I have it on it stays right around 100. it doesn’t go up or down a whole lot when we’re playing. I will see if I can get into a unitron with musician microphones. Also hearing aid helper would you email me direct? honkytonkjohn@live.com

Thank you all for the interesting debate! To clearer sound may we all be happy! One last thing. I use a starkey Davinci PXP. With a silver oxide battery. its the best combo I have found to date. If you have some thoughts on that it would be interesting to hear them. Thanks again!

As far as a know, microphone noise is only a problem at very low input levels. It is not a problem in music. Sampling depth limitation introduces noise at high levels and are therefore much more audible.

Arni i am not a technician so forgive my ignorance! Ok so are you saying that a microphone in a hearing ai doesn’t distort in loud music scenes, but it has a sampling issue? I do know that in quiter inputs, that the floor noise is louder. is this what your talking about? I guess what it boils down to is I’m not sure what your saying. A more lamens response may help my ignorance. thank you!!

Hi

Microphone noise is low and independent of the incoming sound, so it can only be heard when the incoming sound is low. For loud input level, it is simply masked and cannot be heard.

The sampling issue is a problem at very high levels, when the sound pressure is above the sampling scale. This kind of distortion can only be heard at very high input levels.

Most people don’t need dampening or high SPL microphones. Therefore, the manufacturers don’t build them as such. These high SPL microphones are placed in hearing aids by special request.

The software controlled dampening you speak of, is after the microphone input stage. It is controlled by the expansion, threshold kneepoints, compression ratios and output limiting controls. All of which would be ineffective to use if your microphone diaphragm is being warped by the impact of sound pressure.

As a relatively simple but questionably effective concept, putting tape over top of the microphones is actually the right thing to do. This is similar to using an ER musician’s filter in a normal ear to help the eardrum not go into physical distortion thereby making the sound clearer to the listener.

Cheers.

Correct me if I am wrong, but I do believe that everybody would benefit from a higher spl microphone!

No, not exactly. Most people don’t need to have this type of microphone.

Generally, the high SPL microphones are a little ‘tinny’ sounding compared to regular microphones, thus you have to correct the sound within the software by adding more low frequencies.

As mentioned before, most people don’t spend much time in 100+ dB sound environments. Most of them would get as far away from it as possible.

@HearingAidHelper said:

Most people don’t need dampening or high SPL microphones. Therefore, the manufacturers don’t build them as such. These high SPL microphones are placed in hearing aids by special request.
<snip>

I think you may have misread my post. I was suggesting a workaround for converting a “standard” hearing aid into one that would work for Honkytonk’s purposes. If you take a 110 dB input and physically dampen it by 20 dB before it actually reaches the microphone then you’re effectively giving the hearing aid a 90 dB input, which it should be able to handle just fine.

Again this is far from trivial (you essentially need to create your own gain matrix through testing and properly map it to one that already exists in the hearing aid), but should work in principle.