That's all very good and shows that all TO220 drivers seem to be O.K.
Now go back to WH.1 WH.5. and turn the 500R pot to get between 2.0 and 2.4 Volt between these points.
Then, attached to these two points are R132 and R133, both 20K.
Connect the outer sides of these resistors with a wire.
Connect both +/- VCC voltages and switch on the amp, but eventually you could take extra precaution by inserting two bulbs in the +/-VCC lines.
No current should be flowing through R20 and R44 with the VCC's now connected.
If so, switch off the amp and remove the short between R132 and R133.
Switch on the amp back again and measure the voltage over R20 and R44.
Hans
Now go back to WH.1 WH.5. and turn the 500R pot to get between 2.0 and 2.4 Volt between these points.
Then, attached to these two points are R132 and R133, both 20K.
Connect the outer sides of these resistors with a wire.
Connect both +/- VCC voltages and switch on the amp, but eventually you could take extra precaution by inserting two bulbs in the +/-VCC lines.
No current should be flowing through R20 and R44 with the VCC's now connected.
If so, switch off the amp and remove the short between R132 and R133.
Switch on the amp back again and measure the voltage over R20 and R44.
Hans
That is very strange, it may be that there is the source of your problem.
Let me think what to do as a next step.
Hans
Let me think what to do as a next step.
Hans
We will look at the pot at a later stage, but for now connect both ends of R132 and R133 with a short, making the voltage between WH.1 and WH.5 zero volt.
Connect both +/- VCC voltages and switch on the amp, eventually you could take extra precaution by inserting two bulbs in the +/-VCC lines.
No current should be flowing through R20 and R44 with the VCC's now connected.
If so, this will prove that the end stage is O.K.
Hans
Connect both +/- VCC voltages and switch on the amp, eventually you could take extra precaution by inserting two bulbs in the +/-VCC lines.
No current should be flowing through R20 and R44 with the VCC's now connected.
If so, this will prove that the end stage is O.K.
Hans
Mystery solved. I checked resistance between WH1 and WH5 and measured an open circuit. I did some circuit tracing and discovered no continuity between R21 and R124. There is a trace that runs underneath the board between R21 and Q50. Somehow during disassembly of the board for a recap that trace got severed or damaged. I ran a jumper wire and now it works. I will do final bias and test tomorrow.
My sincerest thanks Hans for helping me understand the circuit and developing strategy for troubleshooting. Have a happy New Year. Also thanks to others who chimed in.
My sincerest thanks Hans for helping me understand the circuit and developing strategy for troubleshooting. Have a happy New Year. Also thanks to others who chimed in.
Well, I´m afraid you haven´t solved the whole problem.
With just the trace to R124 being interrupted, ca 1.2 Volt has to be expected between WH.1 and WH.5 and certainly not 3.6 Volt.
Hans
With just the trace to R124 being interrupted, ca 1.2 Volt has to be expected between WH.1 and WH.5 and certainly not 3.6 Volt.
Hans
Everything looks great Hans. I measured WH1 to WH5 on both channels and they are the same at 3.5V. Bias is stable and within .2mV for R20 and R44 both channels (I set it for 26.5mV). DC offset 4mV on R, 9mV L. A 1K sine wave looks good on both channels and amplitude matched to .02V. Amp was running for over two hours.
Hello Hans,
I hope I can still reach you regarding this ML23 amp.
I was so fed up with the very soldered AC-3 PCB which have been broken multiple times that I decided to take another AC-3 PCB in to use.
Currently I am testing the PCB without the the big heat sinks and only with one pair of output transistors. The PCB seems to work fine except the high DC at the LS output which is 21mV. I am sure that when turning on the amp this amount of DC will have a popping sound from loudspeaker which is annoying and is not good for the loudspeaker.
Previously in this topic (#237) you mentioned that adding a resistor parallel to the R121 it is possible to get the voltage down to 0V.
Can you help me to find the right value for this? Now it is a good moment because I can freely modify the PCB because the big heatsinks are not interfering the soldering.
Br,
Handel
I hope I can still reach you regarding this ML23 amp.
I was so fed up with the very soldered AC-3 PCB which have been broken multiple times that I decided to take another AC-3 PCB in to use.
Currently I am testing the PCB without the the big heat sinks and only with one pair of output transistors. The PCB seems to work fine except the high DC at the LS output which is 21mV. I am sure that when turning on the amp this amount of DC will have a popping sound from loudspeaker which is annoying and is not good for the loudspeaker.
Previously in this topic (#237) you mentioned that adding a resistor parallel to the R121 it is possible to get the voltage down to 0V.
Can you help me to find the right value for this? Now it is a good moment because I can freely modify the PCB because the big heatsinks are not interfering the soldering.
Br,
Handel
In #240 I explained that 2Meg//100K on the minus input is the same value as 95k4 on the plus input.
So start with 2Meg or take a pot of at least 5Meg, tune the output to zero, remove the pot, measure it's resistance and then place a fixed resistor in parallel top R121.
Hans
So start with 2Meg or take a pot of at least 5Meg, tune the output to zero, remove the pot, measure it's resistance and then place a fixed resistor in parallel top R121.
Hans
Thanks, I wonder why I did not realize that. Now the DC level is more acceptable level, 10mV. Thank you!
Now you can replace R76, the 95k3 resistor by the 100K you just removed.
This will bring the offset even further down.
Hans
This will bring the offset even further down.
Hans
Perfect, now the DC level is only 1mV. I think the board is now ready for full assembly with the heatsinks.
Is there other transistors than output that needs to be matched? 2sa1006 or 2sc3336 in the predriver?
I have been wondering if thermal paste is really needed. It is pretty messy job. What do you think?