Hi, everyone,
I tried to understand and recreate the composite function.
Unfortunately, I always have strange effects.
Here is a small excerpt.
I currently have a China LM3886 on the table and a NE5532 on a breadboard
I've tried several things and weird things happen every time.
I think I have a fundamental problem that I can't find.
Here is a rough overview:
In the front where it says ok I have a nice sine wave 1kHz with 500mV p.p. , so ok !
If I take out the composite parts, everything goes fine!
I don't have a DC in the signal path.
As soon as I add the composite components, there is suddenly a DC of 500mV at the + input of the LM3886, on which the sine runs with 1k..
Can anybody help me , please .
SO if i Understand it right the composite should be a overall Feedback from Out Speakter to - of the OPV.
But why there comes a DC of 500mV in the system....?
I tried to understand and recreate the composite function.
Unfortunately, I always have strange effects.
Here is a small excerpt.
I currently have a China LM3886 on the table and a NE5532 on a breadboard
I've tried several things and weird things happen every time.
I think I have a fundamental problem that I can't find.
Here is a rough overview:
In the front where it says ok I have a nice sine wave 1kHz with 500mV p.p. , so ok !
If I take out the composite parts, everything goes fine!
I don't have a DC in the signal path.
As soon as I add the composite components, there is suddenly a DC of 500mV at the + input of the LM3886, on which the sine runs with 1k..
Can anybody help me , please .
SO if i Understand it right the composite should be a overall Feedback from Out Speakter to - of the OPV.
But why there comes a DC of 500mV in the system....?
Fundamentally there's nothing wrong with your circuit. The reason for the DC shift could be that the amp is unstable. That's very common for composite amplifiers. Have you simulated the loop gain of the circuit to see if you have sufficient phase margin and gain margin?
Tom
Tom
Hi Tom, fine that you wrote back, so I know that the prinzip is right.
Many Thanks.
Maybe it is because of the mix of chinaboard and my breadboard.... long wires and the china board layout is also not perfect.
many thanks for your reply.
Can you advise me how to set up the best values for the C and the R in red ?
I try to simulate in Lt- Spice but I do not get the right OPV and the LM3886.
I will be honest:
For me as young girl its to complicate to insert the right parts in LT - Spice.
And for P-Spice my PC is to old and to slow, also i do not check it :-(
I simulate with the OPVs which are in the standard Librabry...
No good start condition, I know.
So also , only theoretical i see...
Do you have a Tip for me, how can I set up the two values in
sensible way ?
Good solid values as start condition ?
kind regards
Michaela
The red resistor needs to be calculated from the desired closed loop gain of the circuit.
One way to compensate the circuit is to just increase the red C until the circuit has sufficient phase margin. That'll result in dominant pole compensation. It'll work but you can do better. I would need to simulate the circuit to be able to make a recommendation, but I'd think the red C would land in the 100-470 pF range.
Honestly, the best way is to simulate the loop gain. You can see some of my tricks here: https://neurochrome.com/pages/stability
Tom
One way to compensate the circuit is to just increase the red C until the circuit has sufficient phase margin. That'll result in dominant pole compensation. It'll work but you can do better. I would need to simulate the circuit to be able to make a recommendation, but I'd think the red C would land in the 100-470 pF range.
Honestly, the best way is to simulate the loop gain. You can see some of my tricks here: https://neurochrome.com/pages/stability
Tom
Dear Micha2007,
In case you're not confident about being able to learn stability analysis, here's a circuit you can try. However, note that you may have to play around with some of the capacitor values to get it right (see square wave inset).
The TL041 is similar to NE5532, so do not worry. For best results, try to get rid of the breadboard and any long wire connections before anything.
All the best.
In case you're not confident about being able to learn stability analysis, here's a circuit you can try. However, note that you may have to play around with some of the capacitor values to get it right (see square wave inset).
The TL041 is similar to NE5532, so do not worry. For best results, try to get rid of the breadboard and any long wire connections before anything.
All the best.
Attachments
Hello Michaela,
referring to your name i assume you understand the german language, so have a look at my homepage where i describe my way of building a composite amp.
https://bebo65.webador.de
Bernd
referring to your name i assume you understand the german language, so have a look at my homepage where i describe my way of building a composite amp.
https://bebo65.webador.de
Bernd