Yes it's -18V
Printable View
Yes it's -18V
I thought I woudl take a different approach to solving this. Since we can't figure out why C12 isn't filtering DC properly, I thought may we should figure out why it getting so much voltage in the first place, so i took some measurements (see pic). The first value is the problem left channel, second value is the now working right channel. If there is only one value, it is the same on both channels. You can see there is roughly 3 volts floating around on the left channel where there is not on on the right. Q1 was in a socket so I pulled and tested - tested good no leaks. Q2 was soldered in, but I pulled and tested too, again, tested good no leaks. R56 and R19 slightly out of tolerance so I pulled and replaced. R12 measures fine so left in place. C12/R20 of course were replaced previously. The only other item I suspected was C10 that 27pf cap. I pulled but not sure my meter reads correctly that low. Seeing about 70pf, but then again, my meter shows some 30+pf with no leads connected to anything, so I'm not trusting any readings there. i don't have a replacement to test with. Does it make sense that if C10 was bad that could be the cause of the 3 volts there? Or possibly C15 that 8.8nf cap? Or Q2 really is bad even though tested good? (Since Q1 was socketed I swapped channels - no change) I don't see any other avenues for that "extra" voltage to get created.
Attachment 20058
it's also coincidental that the right channel was 3v too low, raising it up 3 volts and now 3 volts show up on the left where unexpected....
It shouldn't really matter what the voltage is on the left hand side of C12 (as we look at the diagram) because it's the voltage on the right hand side that is being sent to the line stage, and that voltage should be zero. Measurements of that voltage might be useful.
I suspect a DC path in parallel with C12 is causing the problem, as if a large value resistor was in parallel with it. You could remove the cap and try to measure resistance across the PCB pads. Is the PCB fiberglass or paper? Old compressed paper boards are more likely to become conductive.
BTW, there are two components in the phono circuit which look a bit odd to me - R21 and R23. Why is R21 a preset? It looks like it's for fine-tuning the gain and should have any effect at DC, but I'm not sure about that. Also, why is R23 "selected"? And what is that arrow pointing to? I can't even work out what Q5 is doing. Jez might have a better idea than me.
It's certainly a very weird design! It also shows use of basically a Hawksford cascode long before Hawksford "discovered" it! Q5 appears to be further filtering of the -18V rail in conjunction with c17. This is probably shared between the channels hence the arrow from there to the other channel. It has separate AC and DC feedback loops and it appears to me that r23 is selected to set up the correct DC conditions but will also affect RIAA accuracy and gain, hence the preset to adjust out any inaccuracy due to this.
I've only skimmed through this thread but if the problem is only when phono is selected I still suggest R10 as a likely culprit with R11, Q3 and C11 worth looking at.
Having investigated further it appears that this pre-amp has adjustable gain for the phonostage, as can be seen in this picture
http://www.soundsgoodtomehouston.com..._rightrear.jpg
I presume the adjustment pots seen in the picture are R21 on the diagram.
BTW, I wish I know how to embed images into a thread :scratch:
I tried that and failed :doh:
I'll try again...
http://www.soundsgoodtomehouston.com..._rightrear.jpg
Flippin'eck - it worked :eek: