Analog to Digital Converters

Ok I have been reading endlessly! I am excited. I hope someone on here can back me up. WHat has got me going is that I play music. So ok in a band we are louder than 90 db. I just learned that digital hearing aids are limited to a 96 db limit because of the A/D Convertors which are 16 bit. Now I have found out that there is a 24 bit a/d converter for hearing aids. Ezairo 5900. Does anyone know of or know when hearing aid company are going to start using the 24 bit A/D Converters. The cool thing about this is that it will increase the input headroom to 110 DB. Awesome!! I am very excited. Please if any one can confirm my thought please do. Or any other comments would be great!

Thank You!!


interesting question but I can’t for the life of me think of anyone with hearing aids who would want to further damage their hearing with an even louder input to their ears than is warranted for good hearing. :slight_smile:

I would have thought that what was needed was a higher frequency a/d converter to give a better dynamic range to the output signal - hi-fi quality hearing aids. I would also expect that the 24 bits would just make it more accurate in it’s replication of the input signal rather than give a louder output.

Want a louder out - get more powerful output transducers.


Thanks for the response tony!! This is the whole arguement!! its not higher frequency so much as it it is DB levels. Example: a 16 bit converter it limited to a 96 db input. After that it is distorted. On my hearing aid I am adding 25 db gain to the signal. If I play music and say the band is 100 db on stage. the input in my hearing aid is 4 db over what it can handle. So now I am dealing with a signal to the reciever that is distorted and also it stays at about 96 db so there for I have no dynamic range, I can’t tell how loud sounds are! Also a typical band can be up to 110 Db Loud bands are even higher. So from 96 db to 110 db its all the same volume to me, no Dynamics. No Louder or softer sounds!! its distorted and blurred! So its not so much a louder input to the ear but a clear signal after the converter to your ear. I think having more head room on the input is the best thing ever!! lol One last word for now, is what good is a hearing ( like the one I have) if it can put out 140 db if the signal is distorted before it gets amplified? Make sense?

that last sentence was suppose to be: What good is a hearing aid (like the one I have) if it can put out 140 db(to your ear) if the signal is distorted before it gets amplified? Make sense?

A 16-bit convertor is limited to a 96dB dynamic range NOT 96 dB.SPL absolute inputs.

You CAN hear super loud 130dB SPL sounds with a 16-bit convertor set up to handle such levels … but very quiet sounds will then be very quiet indeed!

Also, some hearing aids use an analogue compressing front end amp BEFORE the ADC stage so soft sounds have lots of dynamic range but very loud sounds are compressed into a narrow dynamic range.
(That’s OK because the human ear is not exactly working well at 130 dB.SPL inputs!)

Also, I’m not convinced that all hearing aid ADCs are limited to just 16-bits. Some are 18-bits or better if I remember correctly.

Overall I don’t think the aid’s are the problem when listening to loud sounds … the fitting software is the bottleneck. How often do you see on-screen adjustors for 130dB SPL inputs???

Essentially hearing aids and their fitting software are designed for the 99% of HOH users who do NOT attend rock concerts!

On top of this, the ears of many HOH have poor dynamic range … maybe 60dB or so … well within the capability of a 16-bit convertor.

Maybe you can explain Dynamic range to me better? Is dynamic range the soft to loud sounds? say the softer sound could be 70 db then if you have a 96db dynamic range in your opionon is the loudest sound would be 166 db? I thought that 96 db of dynamic would be limited to a signal no louder than 96 db input? Also I don’t play rock!!lol I play country.peak around 114 db spl. Also the dynamic range of all hoh are down from a normal hearing person this I know. The harder of hearing you are the less dynamic range you can have with a hearing aid.

There is no soft sound at 70 dB. Normal level of conversation is 45 dB and when I’m talking with 75 dB I’m practically yelling.

The “decibel” concept is what you don’t understand. An 80 dB sound is 10.000 times more powerful than a 40 dB sound.

“The unit dB (SPL) is often abbreviated to just “dB”, which can give the erroneous impression that a dB is an absolute unit by itself.”

Lancaster, I agree with you! But englishs definintion left it open. I wasn’t understanding him or her! I will say though with music real world numbers are in that 110 to 115 db spl levels. So i am back to the input again, can it handle those numbers? or can it only handle a 96 db input? If everyone just looks ta the out put, and doesn’t consider what is coming into the aid, what is the purpose of making a hearing aid. I do understand the first concern for a hearing aid is being able talk to people. Not music, but with technology where it is today why not help HOH musicians, also I think it will just help everyone in hearing. You walk into a noisy dance hall. i know the Spl levels are over 90 db which seems to be the industry standard on measuring the input of an aid. So where am I confused on a 24 bit system that can handle 110 db? 96db to 110 db is like wow in my book!! clarity would be improved greatly. I don’t think hearing aids should be any louder. If you can’t hear with something putting 140 db in your ear. thats tuff! I understand! you can’t pump 200 dbs in you ear you would be instantly deaf totally no hearing no more ouch that would suck. No offence to the deaf. I know the deaf have more mind over matter than me, I am weak.

A 24-bit ADC can handle 110 dB dynamic range, but this range could be from 35 to 145 dB SPL. Not from 90 to 200. The softest sound that a hearing aid can process will be practically blended in noise. Since the normal conversation level is 45-50 dB SPL, the designer can not afford to lose the sounds of those levels blending them in noise. Moreover, no hearing aid receiver (the equivalent of the loudspeaker) can handle 160 dB SPL and this is already 100 times more powerful sound than a 140 dB SPL which a power receiver can handle normally. 200 dB SPL is a 1 million times more powerful sound than 140 dB SPL. You will need a hell of a speaker to deal with.

Nowadays, using ADC’s with 96 dB dynamic range, the manufacturers can design HA’s for mild to moderately severe hearing loss handling sounds from 20 to 115 dB SPL or HA’s for severe and profound hearing loss, handling sounds from 50 to 145 dB SPL. They can’t design a hearing aid able to deal with sounds from 20 to 145 dB SPL, since a 125 dB dynamic range will be needed for this.

later edit: the above explanation is oversimplified, since the hearing aid gain is not taken into consideration.

They can’t design a hearing aid able to deal with sounds from 20 to 145 dB SPL, since a 125 dB dynamic range will be needed for this.

A compression analogue front end feeding a 16-bit ADC would allow 20 - 145 dB SPLs … BUT the signal precision for the loud inputs would be ‘squished’ … whilst the speech levels could be at almost normal resolution.

However how many steps of resolution do you need to distinguish between different gunshots or explosions ???

(Siemens have a patent on such a design for a hearing aid front end … but this patent is clearly invalid as this pre-ADC compression trick has been around in other fields for decades so prior art can be invoked)

Ok Now I think we are getting some where!! I think!! :slight_smile: YOu all are getting good now!
Ok First on lancasters comment. The whole point to my question is what the hearing aid can handle before gain is added. I do believe hearing aids are loud enough on the output.
Now for the both of you or others!!Correct me if I am wrong. The basic setup on a hearing aid is:

Microphone, A/D Converter, Dsp, D/A Converter, Speaker(receiver)

Now do you think a 16 bit hearing aid can handle the intake of a sound that is 110 db spl? No gain on the output. I am asking clean sound, the least distorion before it becomes major. Also my understanding of the Digital world is once it is clipped it is clipped. Where the Analog world alittle clipping didn’t hurt. If it can take the 110 db spl level into itself. Then what is the gain “in technology” in dynamic range of a 24 bit system? Its not 0-96 db vs 0-110 db?

Also I believe hearing aid cannot do not need to be louder than 140 db on the output. Unless you would have liked to heard alittle bit than to never hear anything!!

Now to English. How many steps of resolution do you need to distingish a Bass guitar, rythmn Guitar, lead guitar, Fiddle, Snare drum for the kick drum?? Also The pre compression would maybe help, I won’t disagree to that. i heard a suggestion of putting scotch tape over the microphone to dampen the sound.

this IS getting interesting.

Analogy of the hi-fi amplifier…When I was a kid it was considered great to have a 20 Watt amplifier…now you can buy 700 watt things just to power your “In-Home Entertainment System” - the rationale is that the “power” gives the better dynamic range. A higher dynamic range requires higher frequencies, which require more energy to make happen. A 24 bit ADC will require more power to operate than a 16 bit one, but not to increase the volume. Volume still requires better, more powerful output transistors.


TonyA Can you explain what you mean on " A higher dynamic range requires higher Frequencies"? I thought Dynamic just meant the difference between softest sounds and loudest sounds. Say your softest sound is: 50 DB and your loudest sound is: 80db you have a dynamic range of 30 db. Is this right? Also with sound equipment, I have always found that having too much power tends to be cleaner. Also I thought that it helped the bass response. It just fills in the sound. Saying the low tones eat up power. The higher tones don’t need as much power.

Also to talk about your more power more volume. Its not the out put I have a problem with. My hearing aid is supoose to be able to produce a max out of 140db.

ok so on a sound that is 90db my hearing aid adds 25 db of gain to the signal making 115db output. Now my band plays up to say 110db my hearing aid adds 25 db of gain to it and you get 135db. I should be in the range of my hearing aid right? if the max output is 140db! Now this leads me to my question, and also what must be my misunderstanding: I read that a 16 bit A/D converter can handle a input of 96 db max. After that it would be distorted. So I get a signal in at 96 db, my hearing aid adds 25 db of gain, you get 121db output correct? that should be 19 db under the max output of my hearing aid, giving plenty of headroom right? now lets put a signal of 106 db into the hearing aid what happens? it gets flatened out because the input converter can only allow 96db to pass right? but since it is over the 96 it is bound to bealittle louder, but its all distorted. now I have say, a total distorted signal right around 96 db so then you add 25 db to it you have 121db of total garbage. And still it is under the max output of the aid, meaning no problem with the output. Also when the band is playing from say 80 to 115 db everything that is over the 96 is flattened out and distorted. My understanding of dbs is that every 10 db it is twice as loud is that not right? So my dynamic range example with normal ears should be from 80db to 115 db make a dynamic range of 35 db. (over 8 times as loud right?)that is good. With my hearing aid it would be 80 db plus 25 db to 96 db plus 25 db making it 105db to 121db making a dynamic range of 16 db(it would be like 3 times as loud right? Also sacrificing the other 5 times meaning less awareness, of what is really happening.), I could probably get by on the 16 db but everysound that is inputed above the 96 db is distorted so it becomes a roar. It makes me so I cannot tell what is going on in the band, it is a blurr. No definement between sounds. so having a 24 bit A/D converter that is good up to 110db should give me 20 db more of dynamic range allowing another 4 times as loud making the dynamic range 7 times louder overall , and still being under my max output of the hearing aid. Its not 8 times like normal hearing but it would be a whole lot closer!! I am praying that I am right, but I am probably wrong!! :slight_smile:

Now to all of you is this right??

You’re right. And when a loud sound comes right after a soft sound, you will need an amplifier able to deal with really high frequencies, to avoid distorsion of the loud sound. This is why you need a 50 kHz bandwith for a quality amplifier, although you will never be able to hear such frequencies. In the end, the “slew rate” parameter of an amplifier is all about frequency.

Power gives indeed better dynamic range, but it doesn’t explain all. The sound of a 20W vacuum tubes amplifier is perceived as being much louder than a 50W transistor amplifier.

The bit rate is a major factor in the accuracy of the conversion.

The sampling rate must be twice the bandwidth of the input to avoid aliasing errors and is a factor in the distortion generated in the conversion.

The dynamic range of an ADC or DAC is a function of the noise level of the converter…Increasing the bit rate lowers the error (noise). But,there are 1 bit Sigma-Delta converters that have excellent dynamic range. That means that it is not just the bit rate that determines the dynamic range.

As a practical matter, HI dynamic range is a function of many factors, including the mike(s),the pre-amp, and very important is the range of the receiver. Ed

As a practical matter, HI dynamic range is a function of many factors, including the mike(s),the pre-amp, and very important is the range of the receiver

… and don’t forget the 60dB dynamic range of the outer hair cells (amplifiers) and the 40dB dynamic of the inner hair cells (microphones) and the 15dB dynamic range of the ear reflex muscles (loud noise protection) and the 40dB dynamic range of the individual auditory nerves … in a NORMAL hearing person.

Zap a chunk of the OHCs then you find that your 16-bits of dynamic range is quite sufficient!

In most HOH cases (possibly not all) it is probably unwise to use terms like ‘hi fi’, ‘CD quality’ and so on.

It’s unfair to raise the expectations of HOH clients who usually have bandwidth limited, dynamic range limited ear function.

As a dispenser my primary role is to allow my clients to hear SPEECH so they can socialise effectively. Hearing aids are also designed with the same goal.

Sure, a handful of hearing aids offer wide bandwidths and might help musicians who can still hear high frequencies … but this is a niche market.

Well I went Sky Diving!! 13000 feet freefall of 8000 then the parachute!! It was fun!! Anyways I want to thank you all for the great forum! I think it answered some of my question. Still if there is one. I want to try a 24 bit hearing aid!!

Thanks again!

Woo Hoo!!! I guess your music HAs are gonna have to have good wind block as well!! =D

I don’t know if this would be considered “off topic”, but I found this Table of various Sound pressure levels on the web:

It has a handy appendix describing sound pressure as Atmosphere. Pressure force and as Power flow.


0 Beginning of human hearing, a mosquito 10 feet away
13 Ordinary lightbulb hum
15 pin drop from a height of one centimeter at a distance of one meter
40 A whisper
114-107 A very large portable radio
120-130 Front row at a rock concert
127 Human tinnitus begins
137-140 Human ear all frequencies are painful
194.1 one pound of TNT at 6 feet
240 one kiloton of TNT, 233 wide crater 29 feet deep

Well, I’ll say this, you seem to understand much of this, but may be getting a little tangled up in Dynamic Range vs Sound pressure levels, they’re different things. If your hearing had a 96dB dynamic range, you probably don’t need a hearing aid as you can hear from 25dB sound pressure level to 120dB sound pressure level (well wherever the pain threshold is). If you had a perfectly flat 80dB hearing loss, you can only hear between 80dB and 120dB (40dB of dynamic range). The 96dB of a 16 bit A/D has 65536 possible outputs and this can be used to map the input sound pressure level into the 40dB dynamic range of remaining this patient.