Yamaha CA-1000 repair. My first Yammy!

I wanted to make a fuller voltage map and check some components along the way, but this week is a really busy week for me at work. I will have some time for this on the weekend, but I did get a chance to quickly check voltages at the last few transistors (with output transistors installed (which I just tested again, after writing down these new voltages on the image below - haven't pulled TR610/11 for test yet but I did quickly check resistors from collect to base on TR610 and it is not short)). Here is the voltage map for now to give us something to think about...but I've always had a feeling about TR610 from the begining, because of it looking like the 'source' the -ve DC, as well the fact that it was the collector of TR612 (connected also to TR610 collector) which was not connected properly. I will pull TR610 first chance I get. Any good replacement for the 2SA566 in TO66 package? (I guess I have some TO220 that would do the job, like KSA940 or MJE15033G or some TO126 (KSA1220), in the stash at home, but if I need to replace I guess it would be nice to use the original TO66 package if possible - although these days I am starting to face the inevitability of these TO66 drivers all getting replaced by TO220...).

m4MLyuq.jpg
 
Last edited:
okay. I should be asleep now, but I just pulled TR610. Bad. TR609 tested okay. So I replaced the pair with KSA1220/2690 and we now have nice clean sound on the dead channel. It sounds about even with the other channel. However, we are not done yet. On the base of the drivers I noticed +9V (which is about 10V too high)...I quickly probed the bias transistor & pins are elevated to around +8-9V. Well... that's all I have time for today, so let's see what turns up next session... be we have some progress!
 
but this week is a really busy week for me at work.
Don't stress, a few days is no big deal, there is always someone from AK about

but I've always had a feeling about TR610 fr
Are you brave enough to take/switch the TR610 from the non-problematic side. I wouldn't take the risk.

Any good replacement for the 2SA566 in TO66 package?
Not that I know of, maybe others chime in.

I've seen the MJE15033 solution, not very pretty but supposedly works well.
I've not had to go down that path but that's likely to change. Just finished a CA-1000 total refurb, variac startup, voltages OK,
text book bias, 10 minutes in both outputs on right channel go, about 14Vdc on OUT pin.
Waiting for parts ...
 
Are you brave enough to take/switch the TR610 from the non-problematic side.
Clearly a bit of crossover distortion(?) with near simultaneous postings. Obviously no need to consider switch in view of latest findings.
 
Clearly a bit of crossover distortion(?) with near simultaneous postings. Obviously no need to consider switch in view of latest findings.

Yes i understood we had some messages crossing each other in the internet ether. I really felt for you with your CA-1000 issue... All was going so well then game over, we all know that feeling.

I am back to a normal work schedule now so should be able to get back to this with more energy. Last night when i got home, nosee if i could fiddle the VRs and get something more reasonable but no, I still do have some issue in addition to the dead driver I replaced. I will check the remaining transistors next and if they are okay i will hunt around for the source of the remaining issue(s). More soon.
 
Last edited:
One thing I should have been clearer about; when I said I have sound now on the previously dead channel, I meant via headphones... the outputs are still not playing dice. Previously before replacing the dead driver the headphone output was garbage too. No surprise.

I hadn't complained much about it earlier, but you might have noticed voltages on my diff pair bases/emitters and the source drain were about 0.2V higher than they should be. No matter how much I adjusted the VRs I couldn't get rid of it. Today I pulled the diff pair and while they both were working one of them had about half the gain of the other. So I replaced the two 2SA572 with KSA992 (which I spent some time on getting a nicely matched pair). TR601 tested okay. Now I can make the adjustment with VR601 and the voltages in that section look really good now! That is something I would have ended up coming back to clean up anyway, so I'm glad to have got that part sorted.

Then I tested TR604/605 (2SC1124). Both tested okay, but the legs of one of them broke off, so I replaced both with KSA2690A (not different pin-out).

I tested the TR606 yet again. I decided to replace it with a new KSC1845 even though it tests ok, since I've had bad experiences with the 2SC485 in my sansui units.

TR607/8 (2SC734/2SA561) tested okay and were reused. TR609/610 drivers are already new from the other days, and the output transistors are also new.

While matching up that diff pair helped out at the front of the circuit (and replacing TR610 the other day got rid of the weird -negative voltages all over the place), the issue I mentioned after replacing TR610 still exists. Take a look at the new voltage pic. For example you can see on TR606 I have several volts +ve more than I want on each pin. You will also see several of the bases downstream from are elevated by a few volts. I didn't note any voltages for the outputs or TP3/4 as they are all in the one or two mV range (or possibly open?).

MnWyzNw.jpg

So maybe I will check my traces around the output transistors tomorrow, but I still don't really see how that should cause those voltages at TR606 (?). Looking at for example the base of TR606, I actually have a reasonable value for 10 seconds, let's say somethinkg like a few hundred mV, and then it starts creeping up slowly to ~6V. (EDIT: also that 53V on the rail starts at 50V and creeps up over the same time scale) And when I say slowly, I mean it takes a minute to reach steady state. I feel like that time scale rules out a few things already. Thoughts??? If no suggestions I will focus on taking a few more voltage points around TR606 and try to find why the voltages there are so high.

Also note that these symptoms are the same from after I replaced TR610, to after I replaced the several other transistors.

(NOTE: TR604 runs hot, but that is the same on the good channel. Another note: R619 is still 60Ohm instead of 68Ohm due to previous tech work so that still needs to be fixed).
 
Last edited:
A slightly updated voltage map, with the switch lines included now. It's really maddening switching on and seeing many of the voltages are okay and then seeing them drift over a few minutes. I rechecked R613-R617 this morning and their values are all okay, which makes sense as the voltages on them look great at start up. okay, back to it....
4mY1IRM.jpg
 
Ok. I found a mess of the traces around TR610. Fixed that up, and then VR603 brings things into line. Output transistors are now singing via speakers and voltages around the TR606 are normal and stable.

But not able to measure a voltage over the output emitter resistors to just idle. So a little more fiddling, but we are getting there! More soon.
 
had a quick probe around the driver/output stage and here is what we have. Obviously I can twiddle around VR603 and shift the values around but the thing is that while TP4 responds to VR603 adjustment, TP3 is not changing with VR603 adjustment. This is pretty understandable if you look to the base of TR611 & see the voltage is pretty low there at 0.16V instead of 0.55V so it's not 'switching' on - so that is what I'm focusing on fixing now. (I'm not worried about the difference between voltage on the two drivers for example... I will worry about that if it's still there after finding why base of TR611 is so low).

Oh and I'm obviously missing some point: if you look at the schematic it seems to indicated that the emitter of TR609 should be 0.55V, but on the other side of D607 + R627 at the base of TR611 we also have 0.55V. Can that be? I guess I'm just missing something from staring at this schematic tooooo long...

I checked R626 and R625 - okay. I replaced R627/8 as one was a bit cooked and had gone a bit low in value.

GncuURY.jpg
 
Last edited:
Hi Avionic. thanks for chiming in.

Resistors don't go down in value.They go up.If measured in circuit you will get lower values.

The parallel resistances are sufficiently large (MOhms in this case) to measure R627/8 incircuit, but in any case I also measured out of circuit when replacing. One of them had a band stripe on it from being cooked - at some point the bias had been full on I guess. That one measured 3Ohms in and out or circuit. The other without signs of heat stress measured 5ohms (correct value) in/out of circuit.

What resistance do you measure across the emitter resistors(R633 and 634) ..out of circuit...?
Same goes for the emitter resistors - can be approximately measured in circuit in this particular case. Anyway, in or out of circuit they are about 0.5 Ohms as they should be.

Cheers
 
Last edited:
Oh and I'm obviously missing some point: if you look at the schematic it seems to indicated that the emitter of TR609 should be 0.55V, but on the other side of D607 + R627 at the base of TR611 we also have 0.55V. Can that be? I guess I'm just missing something from staring at this schematic tooooo long...

GncuURY.jpg

The schematic is wrong here; there is no way the voltage can be both 0.55 at the emitter of the driver and the base of the output transistor with the diode in between. You need to have about 1.1V at the emitter of TR609 to have around 0.55V on the base of TR611. So I've cranked the bias up via VR603 to pump this voltage up and now everything looks great with the slightly low -0.4V from emitter of driver TR610 shifting up to around -0.6V, and the new emitter voltage of ~1.1V at the emitter of TR609 translating to about 0.55V at TR611 base. Now TR611 has a measurable idle current and I'm able to bias up in Normal mode. (following from this I also call into question some of the voltages at the bias transistor too). I'm reassured of this conclusion by jumping over to the good channel board (which I have already adjusted/biased) and checking voltages around the drivers and bias transistors and they are identical to my 'bad' channel in Normal mode within 5mV, here or there.

So I've saved myself for a complete rebuild (although I did replace a few of the transistors). So I will now order a couple of parts before putting this back together (that 68 ohm resistors needs to be replaced and I need some new 4.7Ohm base resistors (I only have 6Ohm laying around in sufficiently small wattage, as well as a replenishing the transistors a bit). Oh, I also have replaced D607 on the 'push' driver emitter; it tested fine, but the legs were corroded and it wasn't going to last another 30 years for sure.

So summary of repairs done: 1) one bad driver transistor, 2) poorly matched diff. pair replaced (one had half the hFE of the other), 3) one cooked resistor on bias leg of output, 4) collect socket pin on output not soldered properly, 5) a couple of bad traces repaired, 5) reshape the KSC1845 bias transistor with some thermal expoxy to mate properly with the heat sink, 6) realizing schematic was wrong...

Also, I should now move onto Class A....
 
Last edited:
Great to see that you're on top of it.

So I've saved myself for a complete rebuild
The CA-1000 rebuild is very painful, suggest removing/unsoldering the tone board all together. Apart from the
problematic transistors and electrolytics, I changed out the polar tantalums, trimmers on power amp, operations
microswitch, speaker relay. Need to get back to it...
 
Thanks mate. I'm just waiting on a few parts to finish of the repair properly (I have a couple of temporary components). I may replace the VRs at some point as well, but will just enjoy it as is for now, before trying a rebuild.

BTW, do you have a bit of surface oxidation on your chassis? I've got some and I've seen it in a lot of pics. For now I've just soaked the chassis in oil, but if I had it all apart for a rebuild I might be tempted to do deeper treatment of that.
 
BTW, do you have a bit of surface oxidation on your chassis?
Yep, everywhere. These amps (CA-1000, 1010...) appear to be particularly bad. Faceplate/knobs/RCA... I'm
tempted to buy an ultrasonic bath for the knobs. Maybe a fine paint cutting compound for the back panel. Others
must have had this problem...
 
Yep, everywhere. These amps (CA-1000, 1010...) appear to be particularly bad. Faceplate/knobs/RCA... I'm
tempted to buy an ultrasonic bath for the knobs. Maybe a fine paint cutting compound for the back panel. Others
must have had this problem...

Oh, maybe mine isn't too bad. My faceplate and knobs are fine. The back panel is mostly good too. Where i see it is mostly on the inside of the chassis, some of the PCB holders and covers, and even then it's mostly on one side (as if it was treated properly on one side and not the other). I do have an ultrasonic cleaner and find it useful, but I usually end up putting in some elbow grease with some metal polish to finish it off anyway so I wouldn't consider the ultrasonic to be essential unless you have some other stuff you want to use it on.
 
The resistors i needed came, and I've installed them. Here is a pic of some of the old resistors - the obviously cooked one is the one that measured low. I replaced a couple of others too, including that bad substitute by the previous repairer. All resistors on the board are now as they should be!
CO7seSH.jpg


So I installed the board onto the heatsink and then into the amp and I was having some intermittent ability to set bias. One of the TO3 sockets was need to be unsoldered and resoldered because the spacing with the heatsink meant that the transistor pins where making an intermittent connection. That was a blunder on my part; when I corrected a missing/damaged solder on the TO3 socket I must have shifted the position on the socket towards the PCB too much.

Okay, so here we have the nice and stable 'Normal' Class bias setup. One thing I would say is that the manual says to wait for a 3-4 minutes or something like that, but the class A/B bias takes longer than that to warm up those big heat sinks to steady state.
nmzSRWy.jpg


And Class A:
MDmfNja.jpg


With an ambient temperature of about 19C Class A/B pushed the heatsinks only to around 32C, where as Class A hit about 58C. One A/B the mains line draw was about 0.4A and the power supply was dead silent and barely warm; for Class A the idle mains line draw was about 1.2A and the power supply had a feint hum to it if you put you ear within a foot, and temp went to about 25C (relatively cool!).

So we are onto the Function Board adjustment and I want to ask your advice. The PCB layout is slightly different to that on hifiengines CA-1000 manual, and I have seen this mentioned in other threads. And the transistor complement is a bit different to the options in the schematic I have, which makes it slightly tricky at a glance. Here are some pics of my board, which I'm going to reflow on lacquer on once I have any possible issues sorted.
Rg5no5s.jpg

NUF6UHB.jpg

ezDdWyD.jpg

The red guys are the are the input FETs. The test points in the manual can not be identified on the rear of my board, so I've looked for the pairs of 47Ohm resistors between the transistors at the output of the board which the test points are meant to be between, but I have 50V on each side there. So either I have a problem, or I haven't found the spot. So if anyone has already identified the correct point please let us know, otherwise i will leave it as a job for next week to sketch the layout of this board and be sure I have the right points, before I rip into this board.
 
So we are onto the Function Board adjustment and I want to ask your advice.
I read in the CA-1010 Sm not to touch the adjusters on the function board unless you have a distortion analyser and ,,, So that scared me
off touching the CA-1000 adjusters. I will watch this space to see if ok to touch these adjusters...
 
Those two slide switches on the function board will need to be disassembled and properly cleaned. The fusible resistors are usually drifted way out of tolerance.
 
Back
Top Bottom