Technics SE-A5 with random fault (supposedly bias)

[david winter]

Hi,

I'm stuck on this amp. The input board has been recaped. I also changed two small electrolytics on the power stage board.
Right channel works fine, and bias adjusts without trouble.
Left channel will intermittently work well with stable bias once adjusted. However, when the problem happens again, I get 0mV accross both test points (i.e same positive voltage around 14mV) and the trimmer won't adjust anything. DC protection circuit clicks in both cases.
As a result of the problem, the test signal gets clipped on the negative side, which seems to indicate on what section of the output stage the problem happens.
I don't think it's the left section (intput board with double transistors). If I disconnect it from the output stage, the audio output becomes ok whereas the problem messes it up almost all along the audio path.
I tested all diodes (in diode mode) and supposing I did everything correctly, I didn't find any problem even in the synchro bias blocks.
The first question I wonder is whether there is any known 'crap' transistor type in this amplifier.
Output stage uses the following types:
- 2SC1845 and 2SA1015 (I doubt they're candidates)
- 2SD661
- 2SB745
- 2SC2623
- 2SA1124

I uploaded the service manual there (temp location, may be removed in the future).
Below is a partial schematic as well as the signal once in trouble...:

Any hint appreciated. Thanks & Merry XMas !


David

 

[Skywatcher]

Hi,

I'm stuck on this amp. The input board has been recaped. I also changed two small electrolytics on the power stage board.
Right channel works fine, and bias adjusts without trouble.
Left channel will intermittently work well with stable bias once adjusted. However, when the problem happens again, I get 0mV accross both test points (i.e same positive voltage around 14mV) and the trimmer won't adjust anything. DC protection circuit clicks in both cases.
As a result of the problem, the test signal gets clipped on the negative side, which seems to indicate on what section of the output stage the problem happens.
I don't think it's the left section (intput board with double transistors). If I disconnect it from the output stage, the audio output becomes ok whereas the problem messes it up almost all along the audio path.
I tested all diodes (in diode mode) and supposing I did everything correctly, I didn't find any problem even in the synchro bias blocks.
The first question I wonder is whether there is any known 'crap' transistor type in this amplifier.
Output stage uses the following types:
- 2SC1845 and 2SA1015 (I doubt they're candidates)
- 2SD661
- 2SB745
- 2SC2623
- 2SA1124

I uploaded the service manual there (temp location, may be removed in the future).
Below is a partial schematic as well as the signal once in trouble...:

Any hint appreciated. Thanks & Merry XMas !


David

I had a similar situation with an Onkyo receiver a while back. The BIAS would intermittently drop to zero. It ended up being a bad connection on the Base of one of the output transistors. It didn't "look" bad, but after desolering, and re-soldering, it took care of it.

 

[david winter]

I had a similar situation with an Onkyo receiver a while back. The BIAS would intermittently drop to zero. It ended up being a bad connection on the Base of one of the output transistors. It didn't "look" bad, but after desolering, and re-soldering, it took care of it.

I thought about this before posting as the solders didn't look old. Continuity checks didn't fail at all, even while pushing the legs one after the other.
You would be right if this happened to Q421/Q423. I'll double check that and redo the solders soon.
It could also be a random connection inside a component, probably a semiconductor, but as to find the culprit, it may be a bit hard to find until the problem happens again, as the problem disappeared again.
Replacing all semiconductors (except the outputs) could solve the problem, although I'd need good cross references for most of them except the 2SC1845 and 2SA1015.

 

[S.corp.]

Just guessing, but: rotate the two double diodes (D305,D307) between the channels, and see if fault is still there...

 

[david winter]

I also thought about this but unless I missed something, I found that the problem persists if, when it occurs, I disconnect the link between both sections.
I'll also check this, but supposing one of those MA27A2 diodes is bad, would it be replaceable by a pair of 1N4148 (or even 1N4448) in series ?

 

[Skywatcher]

I also thought about this but unless I missed something, I found that the problem persists if, when it occurs, I disconnect the link between both sections.
I'll also check this, but supposing one of those MA27A2 diodes is bad, would it be replaceable by a pair of 1N4148 (or even 1N4448) in series ?

Seems to me somebody tried that, and they wouldn't work properly. There was some talk about an NTE601 working, as well as possibly an MV1, but I do not know for sure. Some of those threads on DIY just end.

 

[Hamish119]

...The first question I wonder is whether there is any known 'crap' transistor type in this amplifier.

NO! That's an 'Internet fact'. Don't believe everything you read on the Internet (including this...).

However, semiconductors do not like to be operated outside of their 'safe operating area'. Matsushita are well aware of this FACT and addressed the issue on the last page of the service manual that you provided.


Apply the mod to both channels (if not already applied).
Check continuity between Q315 Collector and Q615 Base.

Good Luck! Bonne chance!

Addendum: Nice oscilloscope picture!

 

[david winter]

Thanks for your hints. I applied the 100 ohm resistor update before XMas but this didn't change anything so far as I recall.
However there's something strange in the service manual: those D305...D308 diode are listed as MA27A2 in the components list (Page 20) but appear as SVDMA26-2 (aka MA26-2) in the schematic (Page 21, lower left).
Is this only a typo, or is there a possible factory upgrade made during development stages ?

As suggested I removed and soldered the output transistor legs, and made continuity tests between terminals while their legs were disconnected. No problem so far.
Continuity between the collector Q315 and base of Q615 is good.
I also compared diode tests between both channels. Nothing suspicious.
So far the amp works but I absolutely can't know whether the problem is solved or of it will appear again at any time.
Note that I did a continuity test at the output transistors while the problem happened, and did not see anything wrong.