• 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

SX-780 selector switch problem

ccheath

ccheath

Got an Idea!
Subscriber
Hi, a couple months ago I started working on the 2nd 780 I have. I discovered that the right channel was getting DC from the preamp. I thought it might be the bad capacitor but it wasn't the preamp. the problem is located before the preamp, before the tape 2 and tape 1 switch. I tested the jacks and was getting DC when the AUX was selected but nothing when AUX was not selected. That brings me to the selector switch itself. It was passing very small DC current just enough when I turn my volume knob, the meter moves quite a bit. If I fool with the volume knob long enough, it goes into protection. This receiver is from the bay area and it has some corrosion on the chassis and the circuit board appears to be covered with possibly thin layer of salt or whatever it was. I just washed it off with windex. I did deoxit all the controls and putting alot of effort on the selector switch. Even removing the selector switch solved the DC problem completely. Had anyone encountered this and was able to remedy the problem without taking the switch itself apart? Any suggestion please? Thanks.
 
Register to hide this ad
Could the salt UNDER the caps between the cap and the board be causing the leakage?

The thing is that there is a LOT of DC blocking between the aux /tape mon inputs and the power amp.
 
ccheath said:
This receiver is from the bay area and it has some corrosion on the chassis and the circuit board appears to be covered with possibly thin layer of salt or whatever it was. I just washed it off with windex. I did deoxit all the controls and putting alot of effort on the selector switch. Even removing the selector switch solved the DC problem completely. Had anyone encountered this and was able to remedy the problem without taking the switch itself apart? Any suggestion please? Thanks.
Click to expand...

Wow...did you pick this thing up from pier 39?(literally)...never heard of a salt layer on a unit- dust, yes, salt no...

Wish I had more answers, but at least you have it narrowed down, it sounds like. Good luck:thmbsp:
 
markthefixer said:
Could the salt UNDER the caps between the cap and the board be causing the leakage?

The thing is that there is a LOT of DC blocking between the aux /tape mon inputs and the power amp.
Click to expand...

It appears that the caps are doing their job. When I move the volume either up or down, the meter would jump up. While slowly moving the volume either up or down, the meter moves up and wiggles until I stop turning the volume. But both instances, the meter would go back to zero regardless of volume control's position. As we know, the cap will pass DC until it's charged. Basically at the input of the preamp, the capacitor is charging while the volume is turning up and discharging while the volume is turning down. When the selector is not set at AUX, the terminals would be zero volts both channel. When AUX is selected, the voltage reading on the left channel is still zero but the right channel is 13 volts at less than 10 microamp. The only thing I can see between the tape switches and the aux jack is the selector switch. I removed it and cleaned the area and it solved all the DC problems. After all the cleaning and putting the switch back(heavily deoxited), the dc is passing again on the right channel. So my question is, do you have some good suggestion what is the next move or what good stuff to use before I decide to take the switch itself apart and scrub it?

33&athird said:
Wow...did you pick this thing up from pier 39?(literally)...never heard of a salt layer on a unit- dust, yes, salt no...

Wish I had more answers, but at least you have it narrowed down, it sounds like. Good luck:thmbsp:
Click to expand...

Got it from the guy on ebay. It appears unmolested when I got it. It would have dust, fuzz and some corrosion on the transformer and chassis. But the thin layer of whatever that was looks like as if you sprayed cleaner all over it and leaves behind lubricant but I know it's not the cleaner because it's ALL over the board like mix of dust and thin film of 'very thin oil' which I assume possibly salt. I couldn't think of a better description. I'll post some pic. I gotta get it from my other computer. Stand by.
 
Here are some pics of the inside. On the 2nd pic, you'll see the layer I'm talking about. On the third pic, you may notice what the chassis looks like. The 5th pic would show a better clarity of the bracket holding the bezel. This 'stuff' is only on one side of the metal. Looks as if this receiver was stored on it's side. The 'stuff' is on the outside metal in the 5th pic and on the inside of the metal on the other side of the receiver shown on the 3rd pic. Transformer shows some corrosion. If it's not salt, then I don't know what it is. Remember, this receiver appears unmolested. Even the seller said he never opened it and never had it repaired. I think from the look of it, his words are true.
 

Attachments

  • 100_1195.jpg
    100_1195.jpg
    118.2 KB · Views: 38
  • 100_1197.jpg
    100_1197.jpg
    99.4 KB · Views: 39
  • 100_1198.jpg
    100_1198.jpg
    115.7 KB · Views: 33
  • 100_1206.jpg
    100_1206.jpg
    99.8 KB · Views: 34
  • 100_1299.jpg
    100_1299.jpg
    70.9 KB · Views: 34
Classic salt corrosion.

And I meant that there are SEVERAL DC blocking caps in the chain after the UNCAPPED volume control. It should NOT be getting through.

Does the low filter have any effect upon the meter / volume control interaction.

You have leakage, pure and simple, and my inclination is that it is hiding under components that are protecting it from being cleaned away. I mentioned electrolytic caps because they are usually tightly mounted to the board.

Also see Fred Longworth's salt corrosion posts about scraping the deposits away from the undersides of transistors.
 
That's what I figure to be salt. I will get to checking the transistors and capacitors for any salt contamination. Right now I been trying to figure out to get that selector switch cleaned. I pulled it back out of the receiver to give it a second cleaning. I been using alcohol, circuit board cleaner and DeOxit. That darn thing keep leaking. The ohmmeter will not register, not even at 20 megohm setting. So I fed 13 volt from my power supply and use the meter to check the same spot that has been leaking and there it is still leaking. I dismantled the switch, and scrubbed it to death using alcohol and soap and water. It still leaks. I started scraping between the two switches to stop the leak. No success.:rant: Finally, I came up with the idea to where if I can't get to what is hidden inside possibly the board itself that is supporting the switch parts, I use the soldering gun. I put the soldering gun on one of the lead that is adjacent to the next switch and burn it but not too much burning. That finally cleared the leak. I had been working on that switch for a couple hours. Now, I'm off to bed. In the morning, I'll be reassembling the switch after I check again for any more leak. I hope it don't come back.:scratch2: See pic.
 

Attachments

  • 100_1301.JPG
    100_1301.JPG
    73.2 KB · Views: 36
ccheath said:
That's what I figure to be salt. I will get to checking the transistors and capacitors for any salt contamination. Right now I been trying to figure out to get that selector switch cleaned. I pulled it back out of the receiver to give it a second cleaning. I been using alcohol, circuit board cleaner and DeOxit. That darn thing keep leaking. The ohmmeter will not register, not even at 20 megohm setting. So I fed 13 volt from my power supply and use the meter to check the same spot that has been leaking and there it is still leaking. I dismantled the switch, and scrubbed it to death using alcohol and soap and water. It still leaks. I started scraping between the two switches to stop the leak. No success.:rant: Finally, I came up with the idea to where if I can't get to what is hidden inside possibly the board itself that is supporting the switch parts, I use the soldering gun. I put the soldering gun on one of the lead that is adjacent to the next switch and burn it but not too much burning. That finally cleared the leak. I had been working on that switch for a couple hours. Now, I'm off to bed. In the morning, I'll be reassembling the switch after I check again for any more leak. I hope it don't come back.:scratch2: See pic.
Click to expand...

Did you actually disassemble the switch itself, exposing the contacts? If you did you will see plenty of crannies that salt fog can condense in to cause problems.

Be very careful with the sliders, and REMEMBER exactly where they were and which way they were pointing, then pull them, and use a soft eraser to clean the stationary contacts to bright metal. I use a toothbrush, water, wd40 and finally a zero residue cleaner to clean out the small spots, then reassemble the contact sliders, spray it soaking wet with deoxit, reassemble it and then work the daylights out of it to let the deoxit clean the sliders.

That has become a routine part of a MTF whole hawg fix er up.... and I ALWAYS sweat getting them out where the factory bent the D$#@ pins over. Takes a LOT of finesse not to break them....

P.s. They usually come out quite blackened, especially at the ends where air gets in easier. This is an after pix iiac.


attachment.php
 

Attachments

  • switch_internal.jpg
    switch_internal.jpg
    39.2 KB · Views: 111
Last edited:
Yes, I did disassemble the switch itself. I didn't see much of any crannies. Probably because I was working it to death before I decided to take it apart. Thank you for explaining your procedure. that is something I need to know. I had been scrubbing that switch like nuts. It's a nightmare. It still leaks. The part I mentioned "burn" helps but until I wiggle, I did notice the leak is about coming back when the switch blades was pressed a little. If I can't get the surface that looks so glossy clean any cleaner, it seems to be embedded in the wafer. One thing I'm lacking is the WD40 and DeOxit is about out. I know I have it but I can seem to find it. As soon as I find it and get more DeOxit, I'm going to do your procedure all over again except the eraser, I did that and it looks shiny.

Have you ever encountered that problem with any switches you been cleaning and successfully eliminate the leaks?
 
Are you sure that the crosstalk isn't somewhere on the board itself?

IIAC there may be ground traces acting like shields in between the active traces that corrosion may have isolated - making them more like either antennas or capacitors.

Can you try operating the unit without the switch to see if it isn't there then?

Plus have you cleaned the plastic that the switch contacts ride in?

This IS a new one on me - crosstalk has been rare in my experience...

IF it IS in the phenolic / whatever that holds the pins - since it is made to withstand soldering heat, I would try boiling it in clean water a bit to see if cracks and crannies that may hold salt would have their contents diluted or removed.

I would make SURE that it wasn't on the bottom of the pot, which could get hotter than the water.

That "heating" by the iron would have dried out any salt crystals in the cracks - but salt IS hygroscopic and would suck moisture in again as it cooled - thus the return of your symptoms.
 
markthefixer said:
Are you sure that the crosstalk isn't somewhere on the board itself?
Click to expand...

There was no crosstalk. Couldn't really test for that because there was serious popping, causing the protection circuit to engage. There was 13 volts bleeding from switch "C" to switch "E", getting into the input of the right channel. I think I recall a couple months ago, it was getting to BOTH channels (from switch "C" to "A" and "E" before I first cleaned the switch.

IIAC there may be ground traces acting like shields in between the active traces that corrosion may have isolated - making them more like either antennas or capacitors.
Click to expand...

The main board don't look corroded anywhere. Just thin layer of "goo" all over the top of the board which I already cleaned out. But what was corroded or at least green were the wirewraps. I soldered them.

Can you try operating the unit without the switch to see if it isn't there then?
Click to expand...

Yes, I ran it with my CD player and it played perfectly. DC offset is very stable and no popping. Just music. Currently the switch is still out but before I took it out, I desoldered the 13v feed to the switch "C" and that solved the leak problem. It played music perfectly even with the switch in. If the 13v is reconnected, it'll bleed again.

Regardless if the switch is in or not and if either Tape 1 or Tape 2 is engaged, the leak does not occur. I can play music perfectly.

Plus have you cleaned the plastic that the switch contacts ride in?


This IS a new one on me - crosstalk has been rare in my experience...

IF it IS in the phenolic / whatever that holds the pins - since it is made to withstand soldering heat, I would try boiling it in clean water a bit to see if cracks and crannies that may hold salt would have their contents diluted or removed.

I would make SURE that it wasn't on the bottom of the pot, which could get hotter than the water.

That "heating" by the iron would have dried out any salt crystals in the cracks - but salt IS hygroscopic and would suck moisture in again as it cooled - thus the return of your symptoms.
Click to expand...

Haven't gotten that far yet. I haven't reassembled the switch yet because I am still detecting the leak. (see pic below) Does it look like Phenolic? I sure hope boiling it would help but I want to make sure from you if this material can hold up. Once I get your confirmation, I'm gonna try it.
 

Attachments

  • 100_1318.jpg
    100_1318.jpg
    54.6 KB · Views: 20
  • 100_1320.jpg
    100_1320.jpg
    66 KB · Views: 20
  • 100_1321.jpg
    100_1321.jpg
    120.6 KB · Views: 20
You Sir, have proved the case and narrowed down the culprit most succinctly.

A tip of the hat.

The piece should withstand boiling water, I would hang the piece with a length of wire through an air gap, when inserting it into the boiling water. This is to avoid the possibly hotter bottom of the pot.

Let it cook for a while, it took years for the salt to infiltrate the insulator.


I don't think have any spares of that switch pattern lying about, but that (replacement) might be the next option.


edit2: classic stereo parts on ebay does have a 780 function switch listed - OUCH - for just short of 27 bucks shipped....

I need to look at the elements inside of the tone switches for the 3900, they look sneakingly similar. While the tone switch is a 2 position mechanically limited mechanism, the contacts look quite similar.
edit: sorry - it only has 4 sections, not 6, so I didn't open it to check the contact configuration.
 
Last edited:
I'd recommend something with less potential for destruction...a good soaking in Simple Green followed by a thorough rinsing. An ultrasonic cleaner with the Simple Green would be even better. Been there, done that with contaminated boards from Adcom amps.
 
I agree. $27 is a bit high for a switch. You can almost buy a whole other SX-780 fixer for that price. I have thought about looking for one but would hate to waste a perfect looking switch that just need cleaning and decontaminating.

Simple Green (my favorite stuff) and Ultrasonic Cleaner? Why didn't I think of Simple Green? That's my favorite stuff. I gotta find that stuff stashed somewhere.:D

If we can do without boiling the part, let's try Simple Green first. I don't have the ultrasonic cleaner so I'll have to do without. While a good soak, would warming it up some do? After reading this http://en.wikipedia.org/wiki/Ultrasonic_cleaning , I'm considering getting one in the near future.

Glenn, that Adcom amps. Did you encounter leaks in switches that may not go away with conventional cleaning but cleans successfully with the procedure you just posted?
 
Some Adcom's are prone to leaking caps, which soak the PC board (made of similar material to your switch) and cause all kinds of issues with highly unstable offset. One was so bad that a standard cleaning with the Simple Green three~four times didn't cure it, so I got fed up and bought an ultrasonic cleaner and ran the board through about 5 or 6 cycles with a 1:3 ratio of Simple Green and water. That did the trick.

Warming the cleaner could only help. Once you think it might be clean, I'd put the part in an oven at a very low temp (125°F) for a few hours.
 
Looks like the ultrasonic cleaner is a way to go. That is my near future plan to get.

I tried using simple green to soak the board ALL DAY. Then thoroughly rinsed it. Still leaks but a teeny teeny bit better. Then I used an XActo knife to scribe between sections on top and bottom of the board. That brought the leak way down to millivolts. The bad part is when I pressed the bar, it spikes to several volts briefly then back to millivolts. So, I boiled the part for probably about 15 minutes, dry it, then put it under the dimmer controlled heat lamp. It seems to help a little further to where it's a few volts spikes when the bar is pressed rather than several volts spikes. But it still went back to approx 50 millivolts. Still not satisfied but I realized that my volume pot is 100k and that might bring the leak further down to near zero since the leak is so small that even my meter itself has the resistance of 10 megohm to drain the voltage down from the leak from 13v to millivolts. So, I took the chance putting the switch back together with deoxit to lube the sliding parts then put it back in the receiver. The good news is the meter does not fluctuate while rotating the knob even quickly. The receiver seems to play very good. I even rotate the switch to see if the meter will bump and sound would pop. Nothing happened. Not even the protection kicks in while rotating the volume knob so quickly. I would assume that I may have solved the big part of that problem. It still bugs me to think about the leak still being there. So I went ahead and bought the switch from Classic Stereo Parts in case this switch goes south. At least I'll have a spare.

Thank you both Glenn and Mark for helping me figure this out. I tested it last night in place of my first SX-780. It actually sounds better. I would assume because my first one has those NTE outputs while my 2nd one still has the original outputs.
 
Last edited:
I thank you both Glenn and Mark for helping me with the suggestions on how to get this switch clean. It appears to be successful. Not a hint of DC problem going to the main amp. I been listening to it in place of my first SX-780. It sounds very good compared to my first one. This 2nd SX-780 has original modules and my first one has both replacement with crappy NTE devices. Please see here why: http://www.audiokarma.org/forums/showthread.php?t=267587 Now since BDent is out of stock of those good modules, I plan to figure out the combination of transistors that will succesfully replace the module in my 1st SX-780 since it had been recapped with high quality capacitors. 2nd one is all original.
 
You must log in or register to reply here.
Back
Top Bottom