• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

Yamaha CA-810 Relay Randomly Engages

mbz said:
Yes, I have to agree, @AngelAT has done a fantastic job, most people would have given up, mind you, the CA-810 is worth saving.

Time for some more brainstorming,
TR605 Vb increases since TR604 has shut down so low TR605 Ib and volt drop at R613.
Same with TR606b, ie, TR604 shuts down so low current through TR605Ic, D603, R604.
TR605 and TR604 provide a current path to GND that needs to be open for TR606 to correctly bias.
When a real dc offset occurs additional Ic through TR601 or TR603 causes higher volt drop through R605
setting TR604b to 180mV?

Possible minor error in schematic, would expect TR601e and TR602e to be connected to GND, ie, missing "dot"
Click to expand...

I have some questions after reading your last post, another very plausible explanation.

Why would a shut off Tr604 raise the voltage of Tr605b? When Tr604 shuts off, its collector correct drops to near zero, that current is also the emitter current of Tr605 and the collector current of Tr605, all dropping to near zero, i.e. I-tr604c ≈ I-Tr605e ≈ I-tr605c. Base currents of both Tr604 and Tr605 are very small and negligible (μA range). The collector current of Tr605 is provided directly from the +36V rail through R614, when it drops to zero, the current through R614 decreases and Tr606b voltage increases. This is reflected in the collected data.

Tr605b voltage is determined by a simple voltage divider comprised of R612 and R613, setup on that rectified unsmoothed supply line from the bridge diode. The output of the bridge diode prior to smoothing is around 29.5V (easy to calculate). Calculating the base voltage of Tr605, we get ≈ 3.7V. V-tr605b = (29.5 x R612) / (R612 + R613) ≈ 3.7V
Why would this voltage change at all? C604 is, I think for some degree of smoothing?
Edit: I think I’ve answered my own question, the base impedance of Tr605 is of course different in On and Off states, changing the eq. resistance seen at the base by the current coming through R613 and hence different voltages measured.

None of this explains where the whole thing gets triggered though, I’m looking at a couple of other things too, we’ll see …

@AngelAT , as if you hadn’t done enough, could you monitor the main supplies (+25V, +39V, etc.) during the relay fault? Do they stay constant? Thanks and apologies for so many questions.
C7390A6C-C3FD-46DA-8967-49036D2A5A06.jpeg
 
Last edited:
At C363 and C364 we have 45.2 VDC and -45.2 VDC
Still waiting for the relay to start to click on and off...
Managed to check the +25 and the -25V - all was ok - no fluctuation.
Furthermore, I started a bit of a shotgun approach - replaced all small transistors at the power supply (4x A872 and 4x C1775) and .... again the same problem.
 
Last edited:
@AngelAT thanks for providing all this data. To me at least, it shows the circuit is acting as it should when there’s a fault but doesn’t indicate what or where that fault is. Is it possible for you to get hold of a DM with MAX/MIN voltage function so we can revisit the DC offset voltages of the power amp? Those voltages are the main triggers for the protection circuit and it would be nice to be able to see how widely they vary.

I have populated the schematic with the latest data below and also noticed a possible error in the schematic (correct me if I’m wrong). It seems it was shown that the -45V out of the bridge rectifier was directly shorted to ground, which I took out (circled in red), and added a connection with the -ve side of C363 in black.
01618FF6-A03B-40C8-BAB5-E23227A1A5A4.jpeg
387C4605-ED85-4A58-AF0A-18ACFB26820C.jpeg
 
Last edited:
  • Like
Reactions: mbz
@AngelAT , Perhaps you’ve done so already but please make sure ALL solder joints look like the one circled in green below and not like the one circled in red. And of course the one in blue circle and others like it need mending, I think it was highlighted before … just covering all the easy bases until a definite clue pops up …
B082287A-76F8-4DA3-9CB1-363C171FBF99.jpeg
 
Thanks @Sparkplug!
I think there are a lot of mistakes at the schematic.I just checked the voltages at the power supply and they do not even come close to the ones at the schematic. I was also reading other K threads and it comes out that we cannot reply on the schematic...
I already resoldered the joints and removed any remains from the desoldering (as in the blue circle). I will doublecheck, but what I can say is that the light sometimes give strange shadows.

I will also order a better MM with peak values memory.
 
mbz said:
monitoring voltages at TR603b and TR601e (or TR602e)
Click to expand...
try to capture when relay is cycling.

However also thinking more likely there is a spike comming from one channel of the power amp. This causes the mute cap to partly discharge which then gives you 1-2 seconds to capture these measurements. Everything forward of TR601(+ve) and TR603(-ve offset) looks ok to me.

EDIT: When the relay is cycling, do you get any deflection on the VU meters?
 
mbz said:
try to capture when relay is cycling.

However also thinking more likely there is a spike comming from one channel of the power amp. This causes the mute cap to partly discharge which then gives you 1-2 seconds to capture these measurements. Everything forward of TR601(+ve) and TR603(-ve offset) looks ok to me.

EDIT: When the relay is cycling, do you get any deflection on the VU meters?
Click to expand...
Thanks @mbz!
I just replaced D805 and D806 (kinda desperate measures)

Connected the MM to the emitters of TR601 and TR603. For now I get solid -25V and 0V. Will wait and check when the relay starts to cycle if the voltage changes and if the meters move.
Unfortunately still do not have a Min/Max MM.
 
Whenever you feel like you need a break then maybe an alternative approach...lift on leg of R603 and leave amp on in background and note if the relay still cycles. R603 provides the power amp input to the protect circuit, if relay still cycles with R603 lifted then it's a protection circuit issue.
Can't form definitive conclusion if relay does not cycle. Don't have speakers connected since protect circuit is disabled.
 
Just got the relay ciclying again and meters move, but voltages stay stable, according to my MMs
Will lift one leg of R603 and will report
 
AngelAT said:
and.....surprise!
Clicking again!
Click to expand...
:rflmao:
We all need a vacation … also I know of an exorcist who occasionally works on possessed amps … LOL
Don’t give up, it’ll be figured out …
@mbz ’s suggestion was an excellent one, I really thought this would be the end of it … we’ve definitely moved into the supernatural territory …
 
AngelAT said:
Just got the relay ciclying again and meters move, but voltages stay stable, according to my MMs
Click to expand...
Schematic error, missing GND connection at TR601e, TR602e so test was not real helpful. Fact that TR603b remains stable at -25V indicates the release order is not coming from TR602. So the candidates are TR601c-e or TR603c-e or D602 or D601 breaking down. As I understand, these parts have all been replaced. Need some thinking music...
EDIT possibly an error in the schematic and we are not seeing the true connections?
 
@AngelAT ,

Have you tried setting the speaker switch to off position and see if the relay trips or not? You might want to remove the speaker cables too for this.

What are the front panel settings? Can you set tone control, all filters, Rec out to Off position if they’re not currently?
 
What happens if we pull TR601 and/or TR603? Both should be off under normal operation.
Suggest take your pick, remove one or both then leave amp on in background and wait for relay to cycle. No spkrs connected.
 
Last edited:
It’s certainly getting to that stage where you take out (disable) parts one by one and see where the tripping stops. Here’s a new schematic with the missing ground connection in.
B3DF248F-34DC-41D1-8E4B-C2286B7C6685.jpeg
 
Last edited:
You must log in or register to reply here.
Back
Top Bottom