I've got some Ideas of adding a trimmer to this circuit. The DC offset is off -7 mv on one channel and 10 mv on the other. When I check the outputs, one side is barely biased, while the other is over 10 mv. Bias is only 5mv, so this is effecting low level listening and causing notch distortion. I temporarily have it biased to 9mv so each side has some bias.
I was wondering where the best place would be to fine tune this circuit.
Perhaps putting a trimmer where R519 is? or in front of the input so the feedback circuit doesn't fight against it?
Thanks for any input. This is a great sounding and very dynamic amp, I would like to get it tuned in.
DC rail on the High side is over 200 VDC!!
Here's the schematic:
I was wondering where the best place would be to fine tune this circuit.
Perhaps putting a trimmer where R519 is? or in front of the input so the feedback circuit doesn't fight against it?
Thanks for any input. This is a great sounding and very dynamic amp, I would like to get it tuned in.
DC rail on the High side is over 200 VDC!!
Here's the schematic:
Attachments
The DC offset (-7 and +10mv) is inconsequential tbh. I wouldn't start altering or tweaking to correct that imbalance. The matching of the dual FET and the current mirror transistors play a big part.
Output stage bias should be set as per the manual to whatever current is recommended. Be aware that drift is normal and often the procedure will be to set the bias after the amp has run at say quarter power or whatever for 30 minutes... so read the manual.
In normal use you should never see the high voltage rails activate. In fact its worth checking that the lower rails are in fact OK and are in fact present on the outputs rather than the higher voltage.
Output stage bias should be set as per the manual to whatever current is recommended. Be aware that drift is normal and often the procedure will be to set the bias after the amp has run at say quarter power or whatever for 30 minutes... so read the manual.
In normal use you should never see the high voltage rails activate. In fact its worth checking that the lower rails are in fact OK and are in fact present on the outputs rather than the higher voltage.
Thank you for the reply. I have since gained the knowledge that the proper way to do this would be a trimmer installed between Q501 and Q511 (each side of input pair) like the Sony TA-F700ES does. The Sony uses this between Q401 and Q408 respectively. Curiously, Sony omitted this trimmer from the TA-F707ES. The 707 is very similar but has mosfets instead of bjt's in output stage. To me, the 707 sounds slightly more detailed and natural while having every other quality the same.
Attachments
You're welcome. It 'twas quite a while ago 🙂
Adding a preset to allow emitter (or source) current to be balanced is done on some amps. Usually if that use matched FET's for the input stage such as here then its not really needed. Any offset is low to begin with.
Adding a preset to allow emitter (or source) current to be balanced is done on some amps. Usually if that use matched FET's for the input stage such as here then its not really needed. Any offset is low to begin with.
It's hard fixing one though by swapping and swapping until it is low though.. solder traces can only handle so much resoldering, and beta matching is not close enough.. when they are in circuit for the first time they change from new beta measurements..
Also, It's not just finding a matched pair either, sometimes a little offset is required for the rest of the circuit to clear the protection circuit.. or to not be way off, like 50 mv..
Adding a potentiometer can be tricky too though.. But now we have the answers!
Also, It's not just finding a matched pair either, sometimes a little offset is required for the rest of the circuit to clear the protection circuit.. or to not be way off, like 50 mv..
Adding a potentiometer can be tricky too though.. But now we have the answers!
- Status
- Not open for further replies.