I built an PA board which follows the recommended schematic of STK4044V in the data sheet.
http://www.stk-audio.com/STK-Sanyo/STK4044V.pdf
This setup can obtain a gain of 40dB, which is too high for me. according to the data sheet (http://www.stk-audio.com/STK-Sanyo/STK4044V.pdf), I changed the value of R4 to reduce the gain (I only need 11 dB), but then the system starts to oscillate, and the amp becomes very hot....
Does anyone be familiar with this amplifier?
I'm just wondering What's the problem? and what can I do to get rid of the oscillation?
Thanks a lot!
http://www.stk-audio.com/STK-Sanyo/STK4044V.pdf
This setup can obtain a gain of 40dB, which is too high for me. according to the data sheet (http://www.stk-audio.com/STK-Sanyo/STK4044V.pdf), I changed the value of R4 to reduce the gain (I only need 11 dB), but then the system starts to oscillate, and the amp becomes very hot....
Does anyone be familiar with this amplifier?
I'm just wondering What's the problem? and what can I do to get rid of the oscillation?
Thanks a lot!
What did you do with R2, C7, C8, and C9? These relate directly to frequency and phase response.
Unfortunately, the datasheet provides no Nyquist diagram. I only glanced, but it does not seem to address how to change these components to provide adequate phase margin with lower gain.
You could try increasing the values of these capacitors one at a time. Alternatively you could find the lowest gain possible without ruining the phase margin required for stability.
You could also try adding a small capacitor between pins 1 and 2 (10-100 pf?).
Some devices just don't work well below a certain gain, even when compensated.
Easy way out - configure for higher gain (just use the datasheet circuit) and alter gain structure at the input. You might pay a small noise penalty for this.
Unfortunately, the datasheet provides no Nyquist diagram. I only glanced, but it does not seem to address how to change these components to provide adequate phase margin with lower gain.
You could try increasing the values of these capacitors one at a time. Alternatively you could find the lowest gain possible without ruining the phase margin required for stability.
You could also try adding a small capacitor between pins 1 and 2 (10-100 pf?).
Some devices just don't work well below a certain gain, even when compensated.
Easy way out - configure for higher gain (just use the datasheet circuit) and alter gain structure at the input. You might pay a small noise penalty for this.
According to the data sheet, I didn't change the values of R2 C7 C8 and C9.. I only changed R4(for gain adjustment) and C3(for the low frequency cut off)..
I didn't try the trick you mentioned that adding cap in pin 1 and pin 2. but I did try the lead compensation, which is adding cap in in+ and output, I added 100pF, but it doesn't work......
I didn't try the trick you mentioned that adding cap in pin 1 and pin 2. but I did try the lead compensation, which is adding cap in in+ and output, I added 100pF, but it doesn't work......
the input of this PA is a DAC's output, which has a 6V p-p output swing with 0dBFS digital input. what I want is with 0dBFS digital input, the PA will output 20V P-P swing...... if I continue using the 40dB configuration, the DAC's output should be about 200mV p-p. is there any way to make the DAC's output small but still mapping the 0dBFS scale ?
The feedback is the difference between the open loop gain and the closed loop gain.
As you change the closed loop gain you are actually changing the proportion of signal that is fed back. That changes stability margins.
I'ts why National specify a minimum gain of 10times (+20dB) for their lm3886. But even at this gain the stability margins, particularly the phase margin is quite compromised. I believe that is why you see many recommendations for the lm3886 to have the gain set to +27dB to +29dB. It gets the gain margins into a sweet spot for good reproduction of the signals.
You need to ensure your STK gain margins are not being compromised.
As you change the closed loop gain you are actually changing the proportion of signal that is fed back. That changes stability margins.
I'ts why National specify a minimum gain of 10times (+20dB) for their lm3886. But even at this gain the stability margins, particularly the phase margin is quite compromised. I believe that is why you see many recommendations for the lm3886 to have the gain set to +27dB to +29dB. It gets the gain margins into a sweet spot for good reproduction of the signals.
You need to ensure your STK gain margins are not being compromised.
. if I continue using the 40dB configuration, the DAC's output should be about 200mV p-p. is there any way to make the DAC's output small but still mapping the 0dBFS scale ?
That was my last suggestion - adjust gain at the input. It's the easy way out and maybe the most practical for this application.
Without more complete datasheet info it's hard to change the loop gain. You might be able to (probably can) reduce the gain of this amp, but not as low as 11 dB. Try it first, see how low you can go, then adjust gain at the input with a resistive voltage divider.
- Status
- Not open for further replies.