hearing aids revisited

[audioguru]

You show a frequency response from about 800Hz (no adult male voices and no bass sounds in music) to about 10kHz (no sizzling highs in music) and no S, SH, T, TH, F and many more speech sounds. But your filter will play baby crying sounds very well.

There is a big problem here. I am guessing about what you need. Your hearing doctor has the results of your hearing losses, not me.
You have hearing losses and I have perfect hearing so we are talking in completely different languages.
Ask for a printout of the frequency response of your hearing losses then we can design a circuit to match them.

If we can fix your hearing the same as my eye doctor fixed the cataracts in my eyes then you will hear everything PERFECTLY.
I was going blind. Now I can clearly see things far away that I cannot discuss on this forum (you know what good stuff)! Colors are VIVID! HD TV is pretty good.

 

[MrDEB]

unorthdox circuit?

did some experimenting by adding a pot (see pot A)
leaving all the pots at 10% and just adjusted pot A from 10% to 90% = big changes.
the voltage regulator I was/am contemplating is a LM1117
https://www.mouser.com/ProductDetai...1J7HmVL2ZGIbFZ1FML%2bldZElRTmLoOci9kU20eU2qg=
perhaps this would be a STABLE 3.3v supply regardless if the battery goes low?
found some SMD multi turn pots to allow a precise adjustment.using current circuit just might be the solution to a user adjustable hearing device?
Then cataracts!! yea I had one removed and my eye went from 20/70 to 20/20

 

[audioguru]

Ordinary tone control circuits do not use the pot you added called "pot A".
Since all pots are set the same then the frequency response should be flat, not cutting the highs and lows like yours.

Why do you want a voltage regulator? The opamps do not need one. One 100uf capacitor across the battery filters it and two resistors and a filter capacitor will work fine to make the reference voltage. The reference voltage drops as the battery voltage goes low, it is expected and is normal. A voltage regulator is too noisy.

 

[MrDEB]

can the regulator

been trying different pot settings and get some really different graphs.
Trying to get the low around 300 and have it cut off around 14khz but not having alot of sucess.
Pot A really seemed to help
trying to figure out which pot controls which frequencies and perhaps change a cap or ? to get desired results.
Correct me if I am wrong but want a signal curve that starts bending around 300hz and goes in a smooth curve as high as possible then curve down around 14khz.

 

[MrDEB]

w/wo pot A

pot A really makes a difference but good or bad?
any thoughts on pot A

 

[audioguru]

The bandwidth for audio is defined as when the output level has dropped 3dB. Yours is 550Hz to 38kHz.

The 3-band tone control circuit shows that your pot A should not be installed,
pot B boosts and cuts treble sounds (maximum boost and cut at 20kHz),
pot C boosts and cuts mid frequency sounds around 1kHz and
pot D boosts and cuts bass sounds (maximum boost and cut at 10Hz).

Your circuit has something that is cutting most low frequencies and some mid frequencies.

 

[MrDEB]

looking better

here is the revised circuit as per suggestions. I assume I have the markers in the right location?
first op amp is the mic pre amp
want to figure out how to do the second pic you posted

 

[audioguru]

Your first opamp U1 is biased correctly and uses R1 and R2 to also power an electret microphone.
Your second opamp U3 has its (+) input biased at ground so it will be a rectifier with severe distortion that might be messing up your graph. The graph shows levels that are extremely low (a loss of -76dB which is a loss of 5000 times).

Opamp U2 has no input bias voltage (its (+) input is floating) so it will not work.

Today your tone controls schematic has 5 parts values wrong. I circled them. the uncircled parts vaslues are correct.

The second pic I posted shows the frequencies and range of the tone controls.
The bass has its maximum boost and cut of plus and minus 20dB at 10Hz.
The midrange has its maximum boost and cut of plus and minus 20dB at 1kHz.
The treble has its maximum boost and cut of plus and minus 20dB at 20kHz.
When the tone controls are centered then the frequency response is flat from less than 10hz to more than 100kHz with no gain and no loss.

 

[MrDEB]

something isn't right?

the circuit and values are this https://www.electro-tech-online.com/attachments/baxandall-tone-controls3-png.62650/ but the signal is all weird. no nice curve?
I tried w/ w/o the resistor cap to ground but didn't seem to make a difference.
reason for resistor / cap to ground??
I had R14 at 1k and the sig looked pretty good but now at 100k I get the pictured signal all squiggly.

 

[MrDEB]

signal w 1k resistor

at R14 above 0 and nice curve but the highs are to high?
ALL pots at 50%

 

[audioguru]

Good. Your new tone controls schematic has corrected parts values.

Opamps U2 and U3 have no DC input bias voltage so they will not work. Instead of R13, R15, C3 and C6 you just need two 100k resistors in series from +3.3V to ground making a bias voltage of +1.65V plus a 10uF filter capacitor to ground. Connect it to pin 3 of both opamps.

If R14 is 1k then opamp U3 will have a gain of 100 which when combined with the 10.8 gain of U1 makes the total gain 1080 which is way too much.
Where is a volume control? Your own voice might be much too loud.

Your frequency response curve is correct with no boost and no cut since the controls are at 50%. If you turn down the controls then they cut and if you turn them up then they boost their frequency bands.

 

[MrDEB]

volume control and interesting vari gain

The volume control I plan to put on the TDA2822 input
Was researching op amp gain calculations etc and came across this
https://www.radio-electronics.com/info/circuits/opamp_variable_gain_amp/op_amp_variable_gain_amp.php

Tried to simulate briefly but not real happy but kinda resembles the pot configuration I had?
Curious how you came up with 10.8 gain?
gain is R2/R1 (R1 is 1k and R2 is 100k) or am I missing something again?

 

[alec_t]

am I missing something again?

Well the attachment in post #32 is certainly missing something.......content

 

[audioguru]

Curious how you came up with 10.8 gain?

U1 has a voltage gain of 10.8 times.

gain is R2/R1 (R1 is 1k and R2 is 100k) or am I missing something again?

The gain of your U3 is 1 or 100.

I don't know how deaf you are so I don't know how much gain you need. Also the tone controls add some gain if they have a boosted frequency band. Maybe you need gain to be somewhere from 5 to 50. A gain of 1080 is way too much.

 

[MrDEB]

lower gain total 64.8

Don't know what happen to the vol control pic?
Here is the lower gain = 64.8 total BUT should it drop below the 0 level?

 

[audioguru]

You have TWO voltage dividers to bias the opamps but only one is needed because the input current of the opamps is extremely low.
You have TWO 10k resistors at each input of the TDA2822 power amp (one is the volume control) but only the volume control is needed.
You have wrong value capacitors in series with the 4.7 ohm resistors at the outputs of the TDA2822. They are not 0.01uF.

The first opamp has a gain of 10.8. The second opamp has a gain of 6. The TDA2822 has a gain of 95 but its volume controls are audio-taper so if they are halfway then the gain is 9.5. Then the total gain is 616 which is a lot. The total gain should be less.

For your graph, I don't know the input level and I don't know what output level is 0dB.

Where in the universe will you find weird resistor values of 4k, 5k, 6k and 49k?

 

[MrDEB]

bias is based on a post suggestion

you just need two 100k resistors in series from +3.3V to ground making a bias voltage of +1.65V plus a 10uF filter capacitor to ground. Connect it to pin 3 of both opamps.

these are R 16 and R17 =100K voltage dividers.
I had wrong caps based on data sheet should be .1 (I had .01, thanks for correction)
the two 10K resistors to ground R8 and R4 are as per data sheet. The tda2822 has two separate amps in one chip.

 

[audioguru]

Only ONE 100k voltage divider is needed since it can bias many opamps. Here it biases only two opamps. "Connect it to pin 3 of both opamps." But you have TWO voltage dividers, R16-R13 and R17-R15. Get rid of one voltage divider.

The TDA2822 needs a 10k resistor OR a 10k volume control pot to ground on each input. You have both which is wrong.

 

[MrDEB]

I see where your referring to the voltage dividers. Basically connect the inverting inputs together on U2 and U3 w/voltage etc.
On the TDA2822 I am kinda lost. The schematic has pin 7 (input # 1 with 10k pot to grd) and pin 6 (input #2 with 10k pot to grd)
maybe the screen shot wasn't clear or am I missing something again? wouldn't be the first time.
Going to Mars to locate the weird resistors---LOL

 

[audioguru]

Pin 3 of the opamps is the NON-inverting input that has an extremely low current. It is used for a DC reference voltage that is half the supply voltage so that the output can swing symmetrically up and down from half the supply voltage.
Yes connect them together and connected to a single voltage divider and filter capacitor.

The inputs of the TDA2822 need a DC bias voltage of 0V. It is provided by a resistor to 0V. You have a 10k volume control on each input that provides a DC input voltage of 0V so your 10k resistors R8 and R4 are not needed.