Yamaha CA600 advice

[greenstar]

Hi, I have a Natural Sound CA600 amplifier, that sadly has gone down on the right channel. I have opened it up and see no obvious damage, and power seems to be getting to each main board - about 70v.
I have very little idea where to look next.
I believe these are decent amps and worth saving. Be grateful for any confirmation of this and guidance in sorting it.
Tony

[julie_m]

Try elektrotanya.com for a circuit? Japanese stereo amplifiers usually have the same component layout in each channel, making it easy to compare like with like between the working and faulty channels.

If there is no sign of anything that has obviously suffered, I'd start with DC voltage measurements on each terminal of each transistor, and see what looked out-of-place. If I didn't have an oscilloscope, and there was no DC on the output, I'd wire up an old test speaker and try injecting a signal in various places.

[ajgriff]

Hi-Fi Engine has the manual for free download once you've registered.

Alan

[greenstar]

Found the manual on Electrotanya, thanks, that's a step forward.
Julie, there are voltages on the transistors, but I have no idea what would be out of place and the manual doesn't give them.
I do have the luxury of an oscilloscope (several) but don't in this instance know what to do with one.
I have a pair of 'phones plugged in at present so the signal injection method seems sensible now i have a layout and circuit. At present I hear nothing on the rhs unless I turn the volume right up, when there is faint sound. I have a signal generator - what level would be best and is there any danger of causing damage?
Tony

[julie_m]

There is a service manual at https://elektrotanya.com/showresult?...audio&kat2=all. I've had a look. You should find it very helpful.

If the fuses on the power amp board are intact, the chances are the problem is confined to the small-signal stages, or possibly even the switching.

EDIT: Use the voltages on the left-hand channel, which is working, as a guide to what to expect on the right-hand channel.

Play a music source through the amplifier. Use your oscilloscope (set the controls any way that gives you a picture!) to follow the signal through the left-hand channel, from input to output. Then repeat on the right-hand channel.

If using a signal generator, a maximum of 1V peak-to-peak via a 10nF capacitor should not do any lasting damage (except maybe to neighbourly relations, if you leave it playing through a speaker too long!)

[ms660]

The voltages are shown in the full schematic, for the PSU and one channel, the voltages for the other channel will be the same (manual page 19)

Lawrence.

[julie_m]

If you really don't know how to set the controls on your 'scope (and there's no shame in that; we have all been there) just post a picture, and somebody will be able to tell you what should be turned where and what should be pressed in.

[greenstar]

Many thanks. It will be a bonus to have a working mirror of the faulty amp for comparison!
May take a few days to set up and use this advice - all moves in slow motion here. A nonce like me has lots of scopes because they are so lovely and so cheap. I like them as much as radios, but they don't get much use. Here is a chance.
Tony

[greenstar]

I have just noticed Julie's point about the fuses, on the RH mainboard and checked. Both oc. Both ok on the LH board. What might this mean?

[julie_m]

Both fuses open circuit is bad news: it probably means the output transistors have failed short-circuit. Roughly four times out of five this will have been caused by a short on the speaker wires, in which case the output transistors will be the only dead components: one of them overheated and failed short, took its fuse out, this messed up the biasing and turned the other output transistor full on, then it died and took its fuse out. With any luck, most of the energy in the smoothing capacitors ended up in the short-circuit and none of the other transistors got damaged. (The less said about the fifth time, the better ..... Basically, if anything goes wrong inside the feedback loop, it can blow up all the transistors in a cascade as each one takes the next with it when it fails. In this case, your only hope is to replace all the transistors and any out-of-spec passive components.)

Unsolder and test TR613 and TR615. They probably will be short C-E. Before you replace them, unsolder and test TR609 and TR611 just in case, and replace them anyway if in doubt. The preset resistors will need adjusting, see page 10 of the manual linked above for procedure.

You could temporarily tack in BC337 for TR609 and TR615, BC327 for TR611 and TR613, power it up (with no speaker connected) and measure some voltages as a quick sanity check. The fact of having transistors of the right material and polarity will complete the feedback loop, even though they won't withstand any power. If all looks good, then order replacements; otherwise, look for a new amplifier .....

[greenstar]

OK, checked the power transistors. Luckily they plug in, no unsoldering needed.
2SB 531 C-E high resistance (does not read on low ohm scale) C-B low, about 5 Ohm.
On component tester (Chinese miracle) PNP B 98, Uf 647m
2SD371 C-E high C-B high, on tester NPN, B 66, Uf 611 mV.
Someone tell me what this means? (I know PNP/NPN).

[Ambientnoise]

B is probably the current gain (beta = collector divided by base currents). Uf is probably actually Vf and is likely to be base-emitter forward voltage. Readings look ok but do not necessarily indicate good health, for example there could be high voltage breakdowns which do not show on such a test box. Further checks are needed to try to find the root cause, starting with the other transistors.

[julie_m]

A working transistor should test on a multimeter as two diodes. NPN will have the anodes at the base and the cathodes at the collector and emitter; PNP will have the cathodes at the base and the anodes at the collector and emitter.

NPN  C ---|<|---B---|>|--- E
:
PNP:  C ---|>|---B---|<|--- E

C-E should be high resistance in either direction. If your meter does not have a dedicated diode test range, test a known good diode of the same material to see what is a reasonable reading in forward bias.

Note that an analogue meter is measuring current, which always flows into the red lead and out of the black lead, so the red probe will be the cathode when forward-biased so a current is flowing; whereas a digital meter is measuring voltage, and the red probe will be the anode, which will be at a higher potential than the cathode.

[Station X]

Illustrated guide to transistor testing with a DMM here:-

https://www.vintage-radio.net/forum/...ad.php?t=42194

[greenstar]

Well, both power transistors with the diode range on a dvm
C-B high one way 500 Ohm the other
B-E high one way 500 Ohm the other
Does this indicate ok?
Will look at Graham's link next.

[Station X]

Are you sure it's ohms and not millivolts?

[Diabolical Artificer]

With a DMM on continuity IE beep test a good trannie will beep once briefly then indicate a reading, on my DMM that's the forward voltage drop, a bad one will go beeeeeeeeep, IE it's shorted. This is with red lead to base on a NPN (negative, positive negative), black lead to base on a PNP (pos neg pos).

This is a quick test and doesn't mean the trannie is 100% ok but shows 99% of dead trannies. The other 1% will show some subtle breakdown behavior like leakage etc.

To try and fix amp take voltage readings of the dead channel and compare to the good un or schematic, best to power it up with a current limited PSU. Before that though do you have a SC to ground? Check resistance from the big smoothing cap/s/circuit side fuse holder to ground, if lower than around 100 ohms you have a short = find short before powering up.

Andy.

[greenstar]

Sorry Andy, in this instance what is I? I have E, B, C. Sounds a useful quick test, but need more elucidation.
Sorry, Graham, I have not a clue. I am using a DVM. As it's on an ohm range I struggle to see how it reads mV. Does it matter? Must read your linked thread.

[Station X]

In post #15 you said you were using a diode range. Now apparently you're using an ohms (resistance) range. I recommend using the diode test range (denoted by a diode symbol) assuming your meter has one.

I think Diabolical Artificer meant to type "ie" an abbreviation for the Latin "id est" meaning in other words, rather than "IE" meaning Internet Explorer.

[greenstar]

Graham, thank you, I am using a diode test range on a digital multimeter. As this was clustered with the ohms ranges I did assume it was measuring resistance. All is slightly clearer.