Replace STKs with rcs16's Design - Technics SA-300

[merlynski]

The schematic I have shows the cathode of that zener should be +14.0 vdc and yours measures 14.6 vdc. I would try replacing it with a 14.0 v zener if you have some to select from. Leaky old electrolytic caps and dying transistors are also suspects.

 

[merlynski]

Poking around on the forums here suggests replacement transistors:
2SA798A => 2 matched gain and Vbe KSA992 transistors that are thermally connected
2SC828 => KSC1845FTA
2SC1885 => KSC2690AYS

 

[rjd3000]

The schematic I have shows the cathode of that zener should be +14.0 vdc and yours measures 14.6 vdc. I would try replacing it with a 14.0 v zener if you have some to select from. Leaky old electrolytic caps and dying transistors are also suspects.

Ah. My assumption was that 14.6vdc was within 5% tolerance and therefore was acceptable. In your opinion do you think it's worth replacing the zener with something with a 2% tolerance like this? https://www.mouser.com/ProductDetai...=sGAEpiMZZMtQ8nqTKtFS/ISyHb0gtGyWq8NiIgMhDkA=

My long term plan was to do some recapping anyway so I'll probably do that sooner rather than later with some of the adjacent caps. Thanks for the advice and thanks also for the transistor replacement suggestions. Do you think that's more likely to be causing the offset than leaky caps?

 

[merlynski]

Ah. My assumption was that 14.6vdc was within 5% tolerance and therefore was acceptable. In your opinion do you think it's worth replacing the zener with something with a 2% tolerance like this? https://www.mouser.com/ProductDetail/Nexperia/NZX14B133?qs=sGAEpiMZZMtQ8nqTKtFS/ISyHb0gtGyWq8NiIgMhDkA=

My long term plan was to do some recapping anyway so I'll probably do that sooner rather than later with some of the adjacent caps. Thanks for the advice and thanks also for the transistor replacement suggestions. Do you think that's more likely to be causing the offset than leaky caps?

I have some reservations about whether changing the zener will fix the problem, but it is giving a higher positive voltage, just as the offsets are positive. The current mirror circuit controls the current through the diff amp which sets the voltage drop across the 8.2kΩ emitter resistor. If the positive side of that resistor is greater + voltage than it should be then the negative side will be also. The common emitter should be +0.6vdc according to the schematic, which should put the bases closer to 0 vdc than they are. Until the left side bases are close to 0vdc the rest of the voltages will be offset as well.
It is a tossup whether a bad transistor or leaky cap 'can' be causing it. The options are replace, or remove and test, each component. Once it is removed you have already done half the work to replace anyway.
As for the zener, what I would do is buy 100 of them at ≈7¢ each, make a test jig with a power supply and a resistor and select one that is as close as I can to 14 volts. Maybe more work than you want to put into it, and may not solve the problem (My OCD is showing).
You will probably need a bunch of the KSA992s to get a good match so a little bench testing will probably be in order anyway.
If you are thinking about recapping anyway putting all the parts in one order will save on multiple order shipping costs.
Have you tried to see if the bias is adjustable on the module? That would be good to know before going to crazy with trouble-shooting something else . . .

 

[merlynski]

If you have some cheap speakers you could try running it the way it is, if the bias is adjustable. Since the offset is not adjustable the manufacturer was not all that worried about it, and a few hundred millivolts is not a dangerous amount of offset. Some say that much affects the sound quality. Some of us perfectionists will spend way more time trying to get it perfect than it is worth . YMMV

 

[elnaldo]

400mV offset, to me , is a problem to be solved. Not OK . Up to 100mV, I'd try the amp anyway, but not with 400mV...

The 2SA798 always go bad, so that would be the 1st thing to do, to replace with 2 KSA992. If you buy them from Mouser, they come in a cardboard strip, so they are from the same batch and closely matched, you don't need too many to get a matched pair.

If you have 400mV right at the transistors bases, try to see where is that voltage coming from. It could be even a leaky capacitor letting DC pass from a previous stage. Try to disconnect or remove the coupling capacitor coming from the preamp section. (C601 C602)

 

[rjd3000]

If you have some cheap speakers you could try running it the way it is, if the bias is adjustable. Since the offset is not adjustable the manufacturer was not all that worried about it, and a few hundred millivolts is not a dangerous amount of offset. Some say that much affects the sound quality. Some of us perfectionists will spend way more time trying to get it perfect than it is worth . YMMV

The bias is adjustable since this is rcs16's design that includes the variable resistors at R7. I have some cheap-ish speakers and it doesn't sound half bad even with the bias. I'll probably end up messing with it to get it under 50mv.

400mV offset, to me , is a problem to be solved. Not OK . Up to 100mV, I'd try the amp anyway, but not with 400mV...

The 2SA798 always go bad, so that would be the 1st thing to do, to replace with 2 KSA992. If you buy them from Mouser, they come in a cardboard strip, so they are from the same batch and closely matched, you don't need too many to get a matched pair.

If you have 400mV right at the transistors bases, try to see where is that voltage coming from. It could be even a leaky capacitor letting DC pass from a previous stage. Try to disconnect or remove the coupling capacitor coming from the preamp section. (C601 C602)

I'll try messing with C601/C602 and see what happens. Thanks!

Never tried to match a pair of transistors before but it doesn't seem too hard. I'll go that route and give the rest of the caps more scrutiny if the coupling caps aren't leaky.

 

[merlynski]

The bias is adjustable since this is rcs16's design that includes the variable resistors at R7. I have some cheap-ish speakers and it doesn't sound half bad even with the bias. I'll probably end up messing with it to get it under 50mv.

Exactly what I meant. Now that it is "functional" its time to get it fixed correctly. There was no point in trying to just correct the offset if it had a bigger problem .

 

[rjd3000]

So it's been a while but I have an update. I replaced most of the electrolytic caps in the amp section, matched a couple of pairs of transistors for Q601 and 602 as best I could, replaced the zener D603 with one that measured exactly 14v in my test rig, and also managed to shoehorn rcs16's STK replacements into this little guy. Ended up having to mount them vertically on the heatsink because they would not fit horizontally. Sounding pretty great so far. I still have a little bit of DC offset at the speaker terminals, maybe 60-80mv.

The zener that's supposed to be doing 14v is more like 13.7v when measured in-circuit with the amp running, and I might not have done the most thorough job of matching the transistors. But for now I'm quite happy with the way it sounds.

If you look closely below you can see the heat shrunk matched pair of transistors to replace Q601 and Q602. I had to add heat shrink tubing to the collector pins to make sure they didn't touch any of the others.

 

[jhjove]

I recently acquired a Sherwood AD-2210CP.... found this thread searching for DC offset and STK modules.

Work I've done:
-performed a e-cap replacement
-swapped out 2SA798 with KSA992 pairs, thermally connected

Overall this amplifier sounds nice.

This unit uses STK1030 modules. So far the best the DC offset measure is 44mV / 34mV for the left / right channels.

I was wondering if zener replacements might be warranted or if I'm just splitting hairs to get the DC offset lower.
- WZ197 rated 20V measures 21V
- WZ150 rated 15V measures 15.9V
- WZ240 rated 24V measures 23.5V
All are within the +/- 5% spec

I probably should leave well enough alone.... but any thoughts may be good.
Thx,
Joe

 

[merlynski]

I recently acquired a Sherwood AD-2210CP.... found this thread searching for DC offset and STK modules.

Work I've done:
-performed a e-cap replacement
-swapped out 2SA798 with KSA992 pairs, thermally connected

Overall this amplifier sounds nice.

This unit uses STK1030 modules. So far the best the DC offset measure is 44mV / 34mV for the left / right channels.

I was wondering if zener replacements might be warranted or if I'm just splitting hairs to get the DC offset lower.
- WZ197 rated 20V measures 21V
- WZ150 rated 15V measures 15.9V
- WZ240 rated 24V measures 23.5V
All are within the +/- 5% spec

I probably should leave well enough alone.... but any thoughts may be good.
Thx,
Joe

Joe,
I don't see a problem with your voltages, dc offset less than ±50mVdc is fine.
Since this thread is about a different unit you will probably get a better response if you start your own thread and continue from there.
And welcome to AK!

 

[jhjove]

Thanks for your reply.
Yea, I really don't want to hijack a current thread. The common theme for me was trying to adjust DC offset when a power amp is a STK module.

So yes, I think I will start a new thread.

regards,
Joe

Joe,
I don't see a problem with your voltages, dc offset less than ±50mVdc is fine.
Since this thread is about a different unit you will probably get a better response if you start your own thread and continue from there.
And welcome to AK!