op-amp config

[audioguru]

Your schematic has a power supply that is "HI and LO" but shows no voltages. Since the input voltage reference of the first opamp at its pin 3 is 0V through R15 then the opamp needs a positive supply and a negative supply.
If its pin 3 was biased at half the supply voltage then a single-polarity power supply can be used.
Without R14 its voltage gain will be 1+ (270K/10k)= 28.

R1 should be removed since it reduces the input level.

 

[zesla]

Ok audioguru,

I modified the circuit as you can see at the below pic,

Please are you able to help me to select the values of R12, R6, C1, R5 and R11 so that I could get the best performance at 20kHz of bandwidth by a gain of say 500 to 800 please?
As you can see I am using TL072 for the task.
By thr way I am not sure the values of C2 and C4 are good enough or not? And yes Of course I use a dual power supply for this circuit.


P.S Finally I could convert the circuit so that It produces an output which is 180 degrees out of the phase to the other output by a scheme I took from this site:
Shavano Music Online - Introduction to Op-Amps - Part 2
I added it so that you take a look at it to and make an evaluation regarding the whole circuit (any trim?)

Thanks a bunch

 

[audioguru]

Your circuit does not show a voltage for the negative supply.
Why do you add an offset voltage to the outputs with R14 and R8?
Why do you need the inverted output?

You have already selected 1uf as the input capacitor value for C2 so simply calculate the value of the input resistor R12 so that you have good bass response.
Since the values of R14 and R8 are so low then the bass respose created by them and C4 causes a -3db frequency of 100Hz and deep bass sounds will not be produced.

Don't you know the extremely simple formula for the gain of each opamp?

 

[zesla]

Your circuit does not show a voltage for the negative supply.

It just uses +- 7.5V, So the "LO" comes from -7.5V, and the "HI" comes from +7.5V.

Why do you add an offset voltage to the outputs with R14 and R8?
Why do you need the inverted output?

That's because I need to have them for the next stage.

You have already selected 1uf as the input capacitor value for C2 so simply calculate the value of the input resistor R12 so that you have good bass response.
Since the values of R14 and R8 are so low then the bass respose created by them and C4 causes a -3db frequency of 100Hz and deep bass sounds will not be produced.

Sorry As you know my knowledge in electronics is limited. How to solve the problem? Can I use two larger resistors for R8 & R14 yet have the 1/4 ratio? or...?

Don't you know the extremely simple formula for the gain of each opamp?

I now the formulas but I don't loke to do it becaise I am just a beginner in the OP-amp fileds, I fear to forget something like unitiy gain problem or other problmes regarding to the op-amp Characteristics or the delay I faced before due to the C1 and R5, So I kindly want you to tell me the best values for them because you are audio guru.

Thanks like always.

 

[audioguru]

That's because I need to have (the offset voltages) for the next stage.

Then the opamps will clip the signal when it is high level. Why?

As you know my knowledge in electronics is limited. How to solve the problem? Can I use two larger resistors for R8 & R14 yet have the 1/4 ratio?

Of course. The datasheet for the TL072 shows that its input current is almost zero so resistor values very high can be used (20k and 80k) then smaller, better and cheaper capacitors can be used for coupling the signal. But nobody makes an 8k or 80k resistor. Maybe you can use 82k.

The simple formula for the resistor load on a coupling capacitor is 1 divided by 2 pi C f. The "f" is the -3db cutoff frequency.

I know the formulas but I don't like to do it because I am just a beginner in the OP-amp fields

The formula for the gain of an opamp is the extremely simple ratio of two resistors so use it.

I fear to forget something like unitiy gain problem

What problem?

or other problems regarding to the op-amp Characteristics or the delay I faced before due to the C1 and R5

What other problems?
Your delay problem was when you used a single supply voltage so the input coupling capacitor needed to charge up to half the supply voltage which took a long time. With the dual polarity supply then the input capacitor does not charge.

 

[zesla]

OK here's the values I selected for the circuit.

Please check it up and let me know if the values are good selected??!

Thanks a bunch

 

[audioguru]

Since you need 3 opamps then a TL074 quad opamp can be used instead of one TL072 dual plus one TL071 single.

The gain is 505 times.
The input capacitor C2 and input resistor R12 cause the gain to be -3dB at 10.3Hz.
Capacitor C4, R14 and R8 cause the gain to be an additional -3dB at 4.5Hz and takes a long time to charge.
Capacitor C1 and R5 cause the gain to be an additional -3dB at 10.3Hz.

Capacitors C2 and C1 should be non-polarized which are huge and expensive if they are film types. If the resistor values were 10 times higher (1M and 47k) then ordinary small and inexpensive 330nF (0.33uF) film capacitors can be used.

R14 and R8 cause an output offset voltage of 3.0V. So the outputs will have a max voltage swing of only 3.3V and will clip the signal if the input is only 4.6mV RMS or more.

 

[zesla]

Since you need 3 opamps then a TL074 quad opamp can be used instead of one TL072 dual plus one TL071 single.

Then What about Crosstalk distortion? Do't I need to consider it if just use a Tl074?

The gain is 505 times.

But I guess the gain is 474?!

The input capacitor C2 and input resistor R12 cause the gain to be -3dB at 10.3Hz.

So those values should be good to be used for audio?

Capacitor C4, R14 and R8 cause the gain to be an additional -3dB at 4.5Hz and takes a long time to charge.
Capacitor C1 and R5 cause the gain to be an additional -3dB at 10.3Hz.

Are you telling that I will have almost -9dB at the lower thereshot (i.e 20Hz)? If so, it would not be so good, right?!

What will happen if it takes long time to be charged?

Capacitors C2 and C1 should be non-polarized which are huge and expensive if they are film types. If the resistor values were 10 times higher (1M and 47k) then ordinary small and inexpensive 330nF (0.33uF) film capacitors can be used.

Why C1 should be non-polarized too?

Please take a look at your post here:
https://www.electro-tech-online.com/threads/op-amp-config.111707/#post919538

There you just suggested not to use high values feedback resistor, but now you are telling that I CAN use 1M resistor for the feedback??!! If I do so then There would not be any problem with anything??!! Is that due the BJT input transistors for AD826?

 

[audioguru]

Then What about Crosstalk distortion? Do't I need to consider it if just use a Tl074?

crosstalk is not distortion. It is a signal on one channel that leaks into another channel.
Simply look at its datasheet. Its crosstalk is at -120dB which is about 1/1000th as much as you can hear if only one channel has a signal.

But I guess the gain is 474?!

The first opamp has an inverting gain of 100k/4.7k= 21.3 times.
The second opamp has a non-inverting gain of (100k/4.7k) + 1= almost 22.3 times.
The third opamp has a gain of 100k/100k= 1.
The total gain is 474. I don't know why I said 505.

The difference in gain between 474 and 505 is so small that you won't hear any difference.

So those values should be good to be used for audio?

The cutoff frequencies of the three capacitor add together which reduces the bass response.

Are you telling that I will have almost -9dB at the lower thereshot (i.e 20Hz)? If so, it would not be so good, right?!

No deep bass.

What will happen if it takes long time to be charged?

Then the amplifier will not work for 5 seconds after it is turned on and will not work properly until about 23 seconds after it is turned on.

Why C1 should be non-polarized too?

I had the circuit mixed up since you have an offset voltage. Only C2 should be non-polarized.

There you just suggested not to use high values feedback resistor, but now you are telling that I CAN use 1M resistor for the feedback??!! If I do so then There would not be any problem with anything??!! Is that due the BJT input transistors for AD826?

Yes.
The AD826 has a fairly high input bias current for its good very high frequency response. Then its input impedance is fairly low and reduces the gain if a high value feedback resistance is used.
But the TL072 has Jfet inputs that have almost no input current so its input impedance is extremely high.

 

[zesla]

crosstalk is not distortion. It is a signal on one channel that leaks into another channel.
Simply look at its datasheet. Its crosstalk is at -120dB which is about 1/1000th as much as you can hear if only one channel has a signal.

-120dB is very good, but I think it is 1/1000,000th.

The cutoff frequencies of the three capacitor add together which reduces the bass response.

No deep bass.

So don't you suggest to incrase the values of the caps so that I could get the lower roll off freq?


I selected 330k and 10k resistors for both op-amps so the gain is 1122. MAybe I reduce it yet by incraseing the 10k resistors. For now I hear some of his and maybe some of noise at the output. That said his is much reduced if I use an individual 47pF cap in parallel with those 330 feedback resistors on 2 cascaded op-amps.
I do'nt know why it reduces the his?? As far as I know this job will cause each op-amp to act as an integrator right? Which values are good to be used for those said caps?

By the way whats the max of freq which I should expect on this gain?

 

[audioguru]

-120dB is very good, but I think it is 1/1000,000th.

Correct.
You might hear something very faintly at -60dB (1/1000th) and -120dB is 1/1000th of it.

So don't you suggest to increase the values of the caps so that I could get the lower roll off freq?

Yes.
If each has a -3dB rolloff frequency of 6.5Hz then the total of all 3 will have a -3dB cutoff at 19.5Hz.

I selected 330k and 10k resistors for both op-amps so the gain is 1122. MAybe I reduce it yet by incraseing the 10k resistors. For now I hear some hiss and maybe some of noise at the output. That said hiss is much reduced if I use an individual 47pF cap in parallel with those 330 feedback resistors on 2 cascaded op-amps.

An OPA2134 dual opamp has much less hiss than a TL072 but it costs more.

47pF in parallel with 330k is a simple lowpass filter (integrator) with a high frequency cutoff at 10.3kHz so most high audio frequencies are gone. 10.3kHz is at -6dB if both opamps have the filter.

By the way whats the max of freq which I should expect on this gain?

High fidelity is flat to 20kHz so 50kHz is at -3dB. I don't know why your gain is so extremely high.

 

[zesla]

Well here's the frequency response of the circuit with 3 high pass capacitors. The out put of the final op-amp will go to the LM3875 configured as the below pic. As you can see the LM3875 has a high pass filter too. the 3dB roll-off of the LM3875 due too Ri and Ci is 7Hz. so it adds a 3dB to the freq resonse graph too. I have to incrase the values of Capacitors so that I improve the freq response at low frequencies, but the main problem is the biggest ceramic cap which I have is 1uF (I dont like to paralell them so that incrase their capacitance). What to do so? I fear to incrase the resistors connected to the caps and hense the feedback resistors too (I fear I face the DC offset problem audioguru told about or something else). So What to do please?

Thanks

 

[audioguru]

Ceramic capacitors are horrible for audio coupling. They are fine for power supply filtering to reduce high frequencies.
Ceramic capacitors change their value when the voltage changes which causes distortion at low frequencies and they are microphonic.

Metalized plastic film capacitors are excellent for audio coupling but a 1uF little pcb mount capacitor is about $1.00 each.
Increase the values of the resistors so that capacitors with lower values (and lower price) can be used.

Your first opamp is inverting with an input resistance of only 10k (so the input coupling capacitor must be 1uF or more). If the opamp is non-inverting then it can have a much higher input resistance then the value of the input coupling capacitor can be much less.

The TL072 opamp has Jfet input transistors so it has an extremely low input current so very high resistor values can be used. But you also list a very old 301 opamp (LM301A?) which has a fairly high input current so it amplifies its input offset voltage if high value resistors are used.
The old LM301A opamp will not work in your circuit because it needs an external frequency compensation capacitor to avoid oscillations and the capacitor kills high frequency response.

 

[zesla]

Ceramic capacitors are horrible for audio coupling. They are fine for power supply filtering to reduce high frequencies.
Ceramic capacitors change their value when the voltage changes which causes distortion at low frequencies and they are microphonic.

Well I used to use the polyester and mica caps for coupling. Are they good for the job?
Are they on the Metalized plastic film capacitors family?

What microphonic means here?

Metalized plastic film capacitors are excellent for audio coupling but a 1uF little pcb mount capacitor is about $1.00 each.
Increase the values of the resistors so that capacitors with lower values (and lower price) can be used.

How much of increaseing is natural here?

The TL072 opamp has Jfet input transistors so it has an extremely low input current so very high resistor values can be used. But you also list a very old 301 opamp (LM301A?) which has a fairly high input current so it amplifies its input offset voltage if high value resistors are used.
The old LM301A opamp will not work in your circuit because it needs an external frequency compensation capacitor to avoid oscillations and the capacitor kills high frequency response.

I just use a TL074.

 

[audioguru]

Well I used to use the polyester and mica caps for coupling. Are they good for the job?
Are they on the Metalized plastic film capacitors family?

A polyester capacitor is metalized plastic film. A mica capacitor is very old but is also a good capacitor for audio.

What microphonic means here?

A microphonic ceramic capacitor is a microphone and picks up sounds and vibrations. It might cause ringing or feedback howling if the gain is high.

How much of increaseing is natural here?

Instead of using an expensive 1uF film capacitor you can increase the value of the resistors 5 times then use a less expensive 220nF (0.22uf) film coupling capacitor. You could increase the resistors 5 times then use a 330nF (0.33uF) film capacitor for better bass response.

 

[zesla]

A microphonic ceramic capacitor is a microphone and picks up sounds and vibrations. It might cause ringing or feedback howling if the gain is high.

Are you talking about common microphones? like electret one's??

Instead of using an expensive 1uF film capacitor you can increase the value of the resistors 5 times then use a less expensive 220nF (0.22uf) film coupling capacitor. You could increase the resistors 5 times then use a 330nF (0.33uF) film capacitor for better bass response.

Well I did so but the output is not so good! I do not know the reason though.
Please tell me What's the MAX of capacitance that I can use if the series resistor be just 10k?
At the circuit discased here there were just 3 high pass capacitive filters each have a 3dB lose at frequencies below 20Hz, But for other designs if I come across to more than 5 or 6 high pass filters then can I just incrase the capacitors value (if the resistor considered to be constant) to any value I want without facing any problem?

Is there any polyester cap with the value of more than 1uF in the markets?

The final question is that the I hear a hum sound at the output (i.e at the speaker), when I put my finger near the ground of the circuit!!!!!?
Wha'ts the reason of this problem? The ground is the reference place of the circuit (zero point) so I did not expect a such problem?!

Thanks like always.