Exactly I use the same
The switches on these are not very reliable best to replace it some switches are more serviceable but this one is best replaced
It’s nice to but not really necessary I do it on my restorations if you are buying transistors on a strip they are often pretty close but still good to check if you want to be really fussy
bolero
Super Member
Yes! On my AU-999 rebuild, I was getting noise thru my phono section with ksc1845's. Changing them out to ksc1815's solved the issue
I wasted a lot of time looking for the cause, because the ksc1845's in the line stage were fine. Finally I tried the swap & it worked.
Last edited:
I'll start looking for a good replacement. Shouldn't be that hard to find something that'll work. Luckily mine is electrically fine so it'll work for a while I think.
I'll have to see if it works long-term but I put a drop of mineral oil on the power switch shaft and put a box under the front of the chassis so the oil would hopefully flow down into the switch a bit. Let it sit that way overnight then pressed it on and off a couple dozen times in the morning. It seems to have it working pretty well now. I haven't done any more work to it since the weekend, just been listening to it a little and really happy with how it sounds now. Should have those KSC1815s in a couple days so when they come in, I'll get to re-capping and changing the 458's in the tone boards and head amp board.
I got started on the tone boards this evening and I'm glad I'm doing it. I got the electrolytics replaced on the left channel board so far and found a few bad ones. One of them was completely open. I wasn't testing them as I pulled them out. Just did them all after I turned the iron off. The open one was a 10uF 25V, so either C715 or C725. I do remember the treble pots didn't seem to be having much effect. I guess that would be why on the left channel. Might have something similar on the right channel.
Tomorrow evening I'll keep plugging away at it. Should have it done by the end of the weekend I hope.
Tomorrow evening I'll keep plugging away at it. Should have it done by the end of the weekend I hope.
Well, I figured this was going a little too smoothly. I got all the electrolytic caps and transistors replaced in the tone boards and head amp. I figured this was a good time to check my work and see if everything's good so far. The good news is the right channel works great. When I first tried it out I was using the Aux 2 input and the left channel was dead. I did a little visual inspection and found a wire that came loose from the tape monitor switch so reconnected that and tried the tape record input and both channels work. The tape play input has a weak right channel but the left works. Taking the front panel apart on a 555 means a ton of wires flexing so it's probably inevitable that they won't all stay attached.
Anyway, figured out that only the inputs that go through the head amp are missing the left channel so that helped narrow it down. Started checking voltages on that board and having a good right channel helps. The voltages at the left channel collectors were low so I pulled the transistors and the voltages were back up at the pads. Replaced those two transistors with new ones just in case and the voltages went right back to where they were. So I'm guessing it's one of the new caps on the other side of the transistors? Maybe I'll pull them tomorrow but I didn't feel like wrestling that board again tonight.
Here's a board layout with the voltages I'm seeing if anyone has any ideas. The right channel has 32V more or less at the collectors. On the left channel, it's 2v and 14V at the two collectors.
I checked to make sure the transistors are in the right way about 20 times and checked the polarity on the caps too. Everything seems right so I guess it's either a bad new part or I broke an old one.
Probably over half the old electrolytics were more than 20% off their values so it was time for this. I replaced the two tantalums in the head amp too. Their ESR seemed high - 40 ohms, even though they were still really close to the rated capacitance. Maybe that's just how old tantalums are but no need to live with that nowadays. The oil snubber cap on the power switch was long gone. It had turned into a 9 Meg resistor. So I got that replaced with a Panasonic X2 film safety cap.
Anyway, figured out that only the inputs that go through the head amp are missing the left channel so that helped narrow it down. Started checking voltages on that board and having a good right channel helps. The voltages at the left channel collectors were low so I pulled the transistors and the voltages were back up at the pads. Replaced those two transistors with new ones just in case and the voltages went right back to where they were. So I'm guessing it's one of the new caps on the other side of the transistors? Maybe I'll pull them tomorrow but I didn't feel like wrestling that board again tonight.
Here's a board layout with the voltages I'm seeing if anyone has any ideas. The right channel has 32V more or less at the collectors. On the left channel, it's 2v and 14V at the two collectors.
I checked to make sure the transistors are in the right way about 20 times and checked the polarity on the caps too. Everything seems right so I guess it's either a bad new part or I broke an old one.
Probably over half the old electrolytics were more than 20% off their values so it was time for this. I replaced the two tantalums in the head amp too. Their ESR seemed high - 40 ohms, even though they were still really close to the rated capacitance. Maybe that's just how old tantalums are but no need to live with that nowadays. The oil snubber cap on the power switch was long gone. It had turned into a 9 Meg resistor. So I got that replaced with a Panasonic X2 film safety cap.
Well, I'm a little stumped. I took the new caps out and they test ok on the LCR meter. On the right channel there's no DC on the transistor emitters but on the left channel there is. So I took the transistors on the left channel out again (already replaced once with no change in symptoms) and now there's no DC on the traces the emitters go to so the DC must be coming through the transistors right? With those transistors removed, the voltage at 6C goes up to 40V so the left channel seems like it's pulling the voltage down (to 32V) when the transistors are in place. I don't know what else it could be other than bad transistors but they test fine on the Peak Atlas and what are the odds of 2 pairs doing this with the same symptoms? Any thoughts? I'd really appreciate some expert opinion here.
I did check the resistors in circuit and don't see any problems there. Everything seems the same between the left and right channels.
I did check the resistors in circuit and don't see any problems there. Everything seems the same between the left and right channels.
I'm wondering now if I have the channels mixed up on the head amp board. Maybe the voltage drop is the way it's supposed to work. After staring at the schematic for way too long, I think maybe the side I was working on was actually the right channel and it works. Did I mention I'm new at this? I guess making dumb mistakes is how you get smarter. I put all the parts back in and everything's still the same. Did some more voltage probing and the base of TR602 has 0.8V which seems like that's how a transistor should work. The base and emitter of TR601 and the emitter TR603 of have nothing on them. So my new theory is TR603 isn't working right so I'll probably pop that one out next.
Success! TR603 was the problem. I guess there's a chance my soldering wasn't perfect but I pulled the old one out, popped a new one in and it works again. The one that was in there tested and still tests good on a Peak Atlas but I guess 32 volts can hit different. I've still got all the mylar and mica caps in it but I think I might just enjoy it and make sure I've got all the little issues like the weak right channel on the tape play input fixed before I take it apart again.
I meant to note one thing on the KSC1815s. I got the only version Mouser had in stock, the KSC1815YTA. Hfe on them was right around 175 (tested with a Peak Atlas) which seemed to be a good match to almost all the 2SC458s I pulled out. They tested around 160-170. But the ones closest to the top of the tone boards had Hfe's around 250. Not sure it'll make too big a difference and probably not worth worrying about but for anyone doing this later, if you can get a higher Hfe version of the KSC1815 you might want to get a couple of those too.
I've only had a little time to listen to the 555 since I got it working but it sounds pretty good. Even though the electrolytics needed replacing, the new ones in the tone and head amp boards haven't made as big a difference in the sound as new ones in the main amp board did. I'm guessing now it's somewhere close to where it was when it was new so I'll listen to it for a while and then probably do the mods and a couple other little things I was waiting on parts for (like changing the one 1/2 watt resistor I had to leave in the power supply temporarily). The last thing will probably be replacing all the mylar and mica caps with polypropylene WIMAs.
There's still a little hiss at full volume but it is reduced from where it was. I think my old 2SC458's were still ok but they're troublemakers apparently so good to get them out.
There's still a little hiss at full volume but it is reduced from where it was. I think my old 2SC458's were still ok but they're troublemakers apparently so good to get them out.
Decided to check the alignment on it after replacing those parts and went to check bias. Left channel was good. Right channel showed nothing. Re-checked left channel and nothing. Realized I popped the fuse in the meter. I guess this is why more modern amps use a resistor so you can just check voltage instead of milliamps across the fuse holder. Checked the bias on the amps setting and they're the same at least so I don't think I'll do any damage. Will recheck when new meter fuse comes in. Rechecked the output adjustment and it was still pretty close. Power output measures 25.7W on the left channel and 22.6W on the right with no clipping into dummy loads that measure 8.4 ohms. That is probably slightly better power output than I had before. Last time I just used an RMS voltmeter and turned it up all the way. I got similar measurements. This time I used the scope and checked peak power with no clipping so not quite turned all the way up. I'm not sure how much those two ways of measuring power output would differ. Maybe not all that much. But I put the dummy loads away before I thought to check it both ways. I'll try that next time.
I've been listening to it this morning and it sounds great. I'm really happy with it and it gives a good feeling of accomplishment to make progress on a project like this. Running into little problems and figuring out how to fix them has been a great learning experience.
I've been listening to it this morning and it sounds great. I'm really happy with it and it gives a good feeling of accomplishment to make progress on a project like this. Running into little problems and figuring out how to fix them has been a great learning experience.
I was just poking around with a voltmeter to see if I could find a good explanation for the 3W difference between channels. It's not a huge difference but more than 10% so figured I'd check into it. I started following the voltage around the board and found what I think is just tolerance stacking. The right channel loses about an extra volt at every resistor until it gets to the output caps where it's 5V lower than the right channel. It's still meeting the spec output power so I guess nothing to worry about but I may check all the resistors with an ohm meter just to make sure there aren't any way off.
Last edited:
I checked several of the resistors in the main amp, especially the ones where the voltage drop was greatest and they're all lower in resistance than their twins on the left channel. One at least is barely within the 10% tolerance. It's a 3.9K and measures 3.55K. So I guess there's some extra current flowing in the right channel and that's pulling the voltage down on that side. I guess it's not worth worrying about since they are in spec, if barely, and the output power is in spec too. Anyone think it's worth replacing those resistors?
Are you measuring the resistance in-circuit or out-of-circuit?
In-circuit but also comparing left and right channels. The resistors on the left channel measure very close to rated value but the ones on the right channel are consistently lower. Still within their 10% tolerance but the tolerance stacking from a half-dozen or so of them is making the right channel voltage 5V lower than the left channel at the output caps.
I would lift the leg of one to confirm. Hard to imagine resistors in one channel are all lower compared to the other channel. Maybe it got lucky.
Lifted the legs of a couple resistors - ones that seemed like they were the farthest off and they tested the same out of circuit. So I guess it did get lucky. Worth replacing any of them? Or if it's making the spec output power and sounds ok, good to leave it alone?