Vintage Pioneer SX-838 receiver loses one channel after warmup

Push in the tape monitor switches when the problem occurs, tape monitor switches are a source of problems on a lot of old stereos they often need to be cleaned.

Seconded. Next suspect after that has to be bad joints, particularly on any devices screwed to the heatsink. You can look for bad joints (and to some extent, dicky connectors) by pressing and tapping on the PCB with some kind of insulated rod - a piece of wood dowelling for instance. After that, a can of freezer spray, and a hair drier are your best friends ...

Arfa

The OP just got through saying he had narrowed the problem to the main amp section...

Granted the controls and switches should be cleaned on general principles, but inn this case I'd strongly suspect tarnished or pitted speaker relay contacts. A knowledgeable tech can usually revive these safely; a DIY attempt may just ruin the relay.

The speaker select switch(s) should be cleaned as well.

Mark Z.

"Readily Visible" wrote in message news:LQpSl.30400$ snipped-for-privacy@nlpi064.nbdc.sbc.com...

The most common affliction older gear like yours gets is oxidized switches and pots in the signal path. Over time oxidation builds up on the contact surfaces of switches and pots (and relays), especially switches and pots which are rarely used (hence the tip about the tape monitor switch(es)). The oxidation is an insulative layer which resists the flow of current. When the contacts stop conducting, a voltage differential is set up across the contacts. By increasing the volume, you are increasing the voltage differential. When it gets big enough, the signal will arc and blast a tiny hole through the oxidation and you'll hear sound again... for awhile. If you'd like to clean it up yourself, it ain't rocket science. For tools all you'll need is a screwdriver to open up the receiver. You'll also need an aerosol can of contact cleaner such as Caig DeOxit and an aerosol can of cleaner/lubricant such as Faderlube (might be called "tuner spray"). You can buy these, or something like them, at any electronics store including Radio Shack. Unplug the unit and take off the cover. I haven't done a lot of work on Pioneers so I'm not sure how accessible the front-panel switches and knobs are, but you may need to take off the knobs and faceplate... hopefully not as sometimes these are a real pain to get apart/together. Use the DeOxit first. Use the thin straw that comes with the can to apply the cleaner INSIDE each switch and pot (knob)... you need to do ALL of them. Volume/balance/bass/treble/loudness/high-cut filter/low-cut filter/tape monitors/speaker selector/tone defeat/turnover frequency switches and any others you find. Each control should have one or more small holes through which you can spray the cleaner. Remember, soaking the outside will do no good, it's gotta get inside the control. After a good squirt, operate each control 50 times. Don't cheat. You're scrubbing away the oxidation. Allow the unit to dry out overnight. Now repeat the process with the Faderlube. Faderlube containes a solvent and a lubricant which coats the innards of the controls and keeps them from immediately re-oxidizing. After letting it dry out overnight again, give it a try. If the problem persists, you may try cleaning the contacts in your speaker protection relay, or just replacing the relay... they're cheap and available. Most relays of that era have a plastic removable cover... it may take some fiddling to figure out how it comes off. The relay will be located near the back of the unit and will likely be about 1" x 3/4" in size. The relay contains gossamer thin copper arms with a contact at the ends... they're very easy to bend or break so be very careful. You can use a fine emery board or a points file to clean up the contact surfaces. I find that if I put the emery board between the contacts, I can GENTLY push them closed while pulling the board out.

If none of this works, get a bright light and a magnifying glass and start looking at solder joints... it's tedious but must be done. Start around the speaker terminals and pay close attention to any joints which are subjected to any mechanical forces at all.

If you happen to have an oscilloscope and a schematic for the unit, you can probably save yourself some time as this appears to be a very reproducible problem. Apply a sine wave to the input and start probing at the output working backwards until you find the problem component(s).

Thanks for the careful reading, Mark.

I hadn't even considered the speaker relays. If that *is* the problem, I should be able to confirm with a simple continuity test when the condition prevails. Power down and test the left side for continuity.

But upon further thought, the fact that the condition persists through a power down and power up cycle makes me doubtful that it is the relay, which physically opens and closes during power down/up.

Also, the fact that the sound does not cut out immediately, but fades over a period of a second or two would definitely rule out the speaker relay. Plus the fact that when the volume is turned way up, the bad channel can be heard faintly.

What intrigues me about this problem is that the I can cause the channel to kick back in by turning up the volume with the speakers muted. That indicates that the problem can be overcome by merely increasing the voltages to the amp circuit by way of increasing the volume, without the flow of current through the circuit.

What is interesting is what is happening as I type!

The volume is fading out more slowly now, over a period of 3 or 4 seconds and then fading back in. It did it twice since beginning to type about what intrigues me. Now it just faded back in over a period of about 15 to 20 seconds!

Real time troubleshooting... now it's playing okay.

One of the fade-outs was so gradual that I thought it was part of the music on the jazz station, but then it faded back in just as gradually and the same song was playing. It did that long fade out/in twice.

Still playing okay, but those fade-outs and fade-ins were unmistakably long and drawn out. Now it occurs to me that it might have been the station fading in and out for some reason. I have the receiver set to mono and the balance set all the way to the left channel so I can monitor it while hearing both channels of music. I don't know what was happening with the right channel during those drawn out fades. But, as I now recall, I have noticed this same behavior once before with the CD player driving the unit.

Now the left channel has done a quick fade out. I will wait to see if if it revives on it's own.

Ten minutes and the channel is still out. Turning the volume up a bit past half way kicked it back in. It then kicked out more three times in well within a minute and kicked back in when the volume was turned up to the same point, just a bit past halfway.

It seems to be cutting out more frequently after about 30 to 45 minutes of playing. The last time the cut out was abrupt, as if a switch were opened.

Time to put it on the desk and bust the case open.

...

unit

left

of

in

not*

his

ue

d.

ough

ly

just

curs, tape

to

fter

..

amp

es,

Bad FET's were a problem with some of the older receivers too. Sometimes a can of cold spray used in small areas at a time can reveal much. Then reheating the suspect component with a soldering iron can help nail down the culprit. But a word of caurtion here. Don't overheat. Lenny

I think you be on to something, Lenny.

After reading about FET's on Wikipedia, I came across this:

"The FET controls the flow of electrons (or electron holes) from the source to drain by affecting the size and shape of a "conductive channel" created and influenced by voltage (or lack of voltage) applied across the gate and source terminals."

This jives with the behavior that I mentioned in the post you replied to about how the channel will kick back in by applying a higher voltage to the amp circuit by turning up the volume, even with the speakers muted so that no actual current is flowing through the amp.

Now, how to identify the FET's as opposed to generic transistors... a quick google image search for 'fet' returns several schematics. The symbol for FET's look different than the symbol for what I refer to as 'generic transistors'. I don't see any FET symbols in the amp section of the schematics for the SX-838, however. Could the schematics from this period use the round transistor symbol for FET's rather than the squarish looking symbol?

I can think of a number of possible causes, but it should be possible to run down the problem without any sophisticated test equipment. Yes, a signal generator and scope would make it a trivial problem. The schematic / service manual seems to be available free here:

I'd suggest setting the volume at a moderate level and measuring the AC signal level at various points in the signal path, comparing the levels at right and left channels. Use a mono source. You might get exceptionally lucky and discover the signal pops back in when you probe a particular point.

PlainBill

I think the type of transistor that I am referring to as a 'generic transistor' might more accurately be referred to as a 'BJT':

This uses the symbol that I am used to seeing for a transistor.

The FET's seem to be represented by this type of symbol:

Could the FET's, if any, in the Pioneer SX-838 be represented in the original schematics by the symbol commonly used for BJT's?

FET part numbers will start with 2SK (N-channel) or 2SJ (P-channel) as opposed to BJT's which are more likely start with 2SA, 2SB, 2SC or 2SD. Note that the "2S" is often omitted on the transistor due to space constraints.

I understand. But doesn't the fact that the left channel continues to be the one that drops out when I switch the right channel preamp output to the left channel main amp input and same for the other two ins and outs at the back of the unit... doesn't this isolate the problem to the main amp section? The left channel is out at the moment and operating the tape monitor switches does nothing to bring it back. But turning up the volume to just past halfway *always* brings it back.

I've done that in spades in the past, except for the Faderlube, with no improvement. And again, I hate to belabor the point, but it seems certain that the volume pot is not the problem since the problem channel does not switch from left to right when I switch the left and right preamp outputs and main amp inputs in the back of the unit so that the right preamp circuit drives the left main amp circuit and the left preamp circuit drives the right main amp circuit. If a faulty volume pot were the problem, the faulty channel would change from left to right when I do this.

But it doesn't.

Not a problem. I've done it already. I have also hot soldered all the cold soldered joints just to rule that out. Must have been 150-200 of them, if not more. Got them all.

I cannot stress enough two things:

1. Switching the right and left preamp outputs with the right and left main amp inputs does *not* change the problem channel. It remains the left channel that drops out.
2. The dropout is not always instantaneous. Sometimes, it fades out and fades in. Sometimes in goes out instantaneously and does not come back. It occurs to me as I type this that this could very well be two different problems.

I will pop the top and check the relay. If it proves difficult to open up the speaker relay, I could easily shunt around it for the left channel just to rule it in or out before busting in to it.

Excellent link, Bill, thank you!

I have the schematics but not the service manual.

I am preparing to bust open the case again.

Ah-yup.

There it is, then. Absolutely confirmation :-)

I have the unit on it's side now with the top and bottom removed. I popped the top off the speaker relay and carefully cleaned the contacts with a strip of 600 grit wet/dry sandpaper and the unit is playing through some test speakers in mono mode with the balance all the way to the left. Usually the left channel will cut out within 15 to 30 minutes.

Needless to say, it would be *great* if that turns out to be the problem... or at least *one* of the problems: the cutting out and staying out until the volume is turn up just past halfway. The fading in and out over a period of ten 10 to 15 seconds is obviously something else altogether.

"Readily Visible" wrote in message news:%YBSl.23076$ snipped-for-privacy@nlpi069.nbdc.sbc.com...

You've shown a good thought process in diagnosing the problem area. I agree with you that the most likely suspect is the power amp section. Now to get into the circuit. If you have the schematic and a couple of good multimeters, it's time to use them. Open the unit to the point that you can easily probe the power amp circuit board and the output transistors. And yes, they are plain old BJTs (not FETs). Play the unit until the left channel dies. While it's still powered up, check the voltages around all the transistors on the power amp board (as well as the power transistors for the left channel. Use the voltages around the working right channel as a reference when the voltages aren't present on the schematic. Since this is a DC coupled circuit, it might be a little tricky to identify the bad guy immediately, but with a bit of logic and transistor theory, it can be done. You will probably see a significant discrepancy somewhere in the signal chain. If you have trouble sorting out the measurements and what they mean, write them down and post them back here. Some very good techs here can help. I've seen transistors become intermittent. That is, they work for a while and then for no apparent reason, they die. This could be due to one of the bonding wires inside the transistor being broken away from the silicon die or the transistor lead. Sometimes, a physical shock such as a tap with a dowel or screwdriver handle will make it act up. Other times, when the transistor heats up from use, the bonding wire will separate. Since the audio dies slowly (as opposed to suddenly), the problem has the bookmarks of a heat-related problem. As has been suggested, a very *quick* blast of circuit cooler can be very helpful. Be aware of the *quick* blast here. The most common mistake in using circuit cooler spray is the tendency to freeze the components to the point that ice forms on them. A one-second (or less) blast is all that's needed to cool a small component such as a small transistor, diode, resistor, etc. Focus the spray on a single component at a time, so you can easily identify it if the channel immediately starts working again.

Hth,

I downloaded the service manual and, according to the info you supplied, there are no FET's in the power amp section. I have the unit on it's end with the top and bottom removed. I buffed up the speaker relay contacts with some 600 grit sandpaper and so far, after 15 minutes of play, the left channel has no cut out. I'll leave it like this for at least 6 hours before concluding that the speaker relay was likely the problem.

Mark, I have the unit on its end on my desk with the top and bottom removed. I opened the relay and buffed the contacts with 600 grit sandpaper. After 20 minutes, so far, so good. I will let it play until I get back from a barbecue and if the left channel is still playing I will call it good and reassemble.

Dave, I am thoroughly impressed with all the help I am receiving here. Thank you all very much! Mark Z suggested checking the speaker relay for bad contacts. This made sense at first as it is on the downstream side of the preamp-amp connectors on the back and I have established that the problem is downstream from this point and the speaker relays are about as downstream os you can get.

But then I realized that I remember the left channel quickly fading out in the past I jumped to the conclusion that a bad speaker contact would tend to cut out instantaneously. But of course, I can't say why exactly I arrived at that conclusion. At any rate, after cleaning the speaker relay contacts, the left channel has been playing for about 25 minutes without cutting out.

If it *does* cut out again, I will follow your advice as far as I can as soon as I can get a can of freeze at the local electronics supply store.

I will keep the group posted.

Jack

You are absolutely correct, I did not read your post clearly enough. If the problem stays with the amp when the channels are switched, it's definitely in the amp. Usually when transistors start to fail you get static, not the fading in/out you describe. I'll keep my fingers crossed for you that cleaning the relay works... it can be a real pain to track down this type of fault. I have spent literally days working on a piece of equipment and it ends up being something I never even thought of.