troubleshooting the circuit

[bugeyes]

Hello,
   was wondering if anyone had the procedures for checking the circuit at the end of the build.
My amp worked great for a couple of years but now the right channel is totally unresponsive
to the volume control. I can turn the volume down but the right channel stays at "10"
  Initially I thought it was the volume control itself, and replaced it with no effect. I'm
thinking maybe something is shorting out, but lost the build documentation long ago.
  Any help would be greatly appreciated,
        thanks....  

[Grainger49]

Doc posted a procedure here:

http://www.bottlehead.com/smf/index.php/topic,1667.0.html

From what you say the input is shorted to the wiper of the volume pot.  That is, the wire coming from the selector switch is shorted to the wire that goes to the right channel tube's grid (tube pin 4 through a 220 ohm resistor).

[bugeyes]

Grainger49,
    thanks for the input. I took my meter and verified both channels from the RCA inputs, to the attenuator, to the resistor 220, to pin 4 on the tube socket. Both channels have the same readings. I tried switching
tubes just in case, same result. Did a visual on the connections coming off the tube socket they all look OK.
It seems like it should be a problem at the front where the signal inputs, but I just don't see it...

[Doc B.]

It sounds like maybe the wire coming from the switch, which should be connected at one of the end volume pot terminals, is shorted to the center terminal of the volume pot. on each channel of the pot measure from each end terminal to the center terminal with the volume setting turned down all the way and see what you get.

[bugeyes]

Got a new unexpected development. Trying different things to try to isolate the problem I noticed that when the RCA plug of the channel that responds to the volume control is plugged in without the other channel, there is a 1/2 second oscillation screech outputted. When the volume unresponsive channel RCA is only inputted there is no oscillation..

[bugeyes]

Doc B,
   followed your advise and indeed found the channel in question to have infinite resistance from input to ground with the attenuator CCW. The other normal channel has 100k the value of the attenuator. The power switch goes from the switch to the power cord terminal then to the transformer and then I kind of lose it, though by implication the short should be in the bad channel. Would you have any advise as to where to look? Or did I misunderstand your statement as to meaning the power switch?

[Doc B.]

OK that means that the grid of the input tube does not have the 100K load it wants to see, hence the screeching.  The volume never goes down because the signal is not getting shunted to ground thru the voltage divider created inside the pot. So it kinda sounds like the terminal lug of the pot that should be grounded is not really grounded.

[Grainger49]

How about posting a picture of the pot and the associated wiring?  To get a picture in you have to first preview, rather than post.  Then at the bottom left there is a pull down menu for additional options.  You can upload a picture from that menu.

[bugeyes]

While I was turning the attenuator around doing resistance checks, apparently I stressed the solder joint
on one channel and it finally broke. This turned out to be the source of the oscillation as when I soldered
it back on the input and the  oscillations went away. The pot is a stepped attenuator from china, an Ebay special, with SMT resistors for the steps. I guess I'm the last person on the planet without a digital camera so I won't be able to post pictures.
   To reiterate the circuit worked just fine for 2 years +, so it would seem to be a wire or component
has shifted in position and is shorting out. But always beware of the assumption I guess....
   So I am back to just one problem the unresponsive channel, it still sounds phenomenal even with out
being able to turn it down ; >

[Doc B.]

Did that resistance reading change when you resoldered the joint?

[JC]

To go back for a moment to your original post, bugeyes, I think the check-out step I would be most inclined to try would be as follows:

This is a resistance check, so the amp can be off and unplugged from the wall, with nothing attached to the inputs.  Measure the resistance from pin 4 of each tube socket to a good circuit ground.  Arrange matters with clip leads on your meter so that you can rotate the Volume control from one end to the other as you watch the meter.

At one end of the rotation, you should read approximately 220 Ohms; at the other end, approximately 100,220 Ohms.  Do this for both A and B tube sockets.  You really don't even need to have tubes in the sockets at the time.

My reference to approximate values is presuming that your replacement Volume Control is of the same value as the original.  If it is not, you may get different readings, but take the readings and see what you get.  Don't be overly concerned about any reference to a "selector switch"; the standard SEX doesn't have one.

[Grainger49]

Just one little thing to add here to Jim's post:

Most, but not all, pots have three lugs per channel.  The center of the three is almost always the wiper of the pot, the wiper is what moves and changes resistance to the other two (sometimes three) lugs.  The other two will have a constant resistance to each other.

And one thing to add... It has also occurred to me that the channel that has full volume might not have a solid ground on the "bottom" of the pot.  That would run the volume wide open regardless of the pot position.

[bugeyes]

I followed JC's suggestion with the clip leads on my meter, point A at the wiper, point B at pin 4
tube socket. The good channel tracks perfectly 100k to 220. The bad channel goes from infinite resistance
to 220 at the very end of travel, the very last step of the attenuator.
  I didn't mention this in previous posts but the volume does shut down on the last step of rotation
on the bad channel. I guess the meter confirms this. I changed the range on the meter to the up-most
scale, which on this meter is 20meg. From the lowest resistance or last step to the next step where it
goes infinite, there is a delay with the resistance ramping up to infinite, like a capacitor is charging up or something. About a 1 second delay. This may be a function of the meter or not I'm not sure. In any of the lower resistance ranges of the meter the change is instantaneous 220ohms 1st step, 2nd step infinite..

[Grainger49]

I followed JC's suggestion with the clip leads on my meter, point A at the wiper, point B at pin 4 tube socket. The good channel tracks perfectly 100k to 220. The bad channel goes from infinite resistance to 220 at the very end of travel, the very last step of the attenuator.

I didn't mention this in previous posts but the volume does shut down on the last step of rotation
on the bad channel.  .  .  .  

Using Jim's method:

Measure the resistance from pin 4 of each tube socket to a good circuit ground.  

What you describe would happen if the "top" of the pot (input from the RCA Jacks) is shorted to the wiper.  Now that would short your source, which is not necessarily a good thing.

Another thing that would cause this is if the wiper only made contact with the ground, fully counter clockwise, lug and not to the resistive part of the pot.

But you have had the identical problem with two pots.  That makes it unlikely.

[bugeyes]

A variation on the previous experiment... meter clipped on the bad channel RCA input, other side clipped to the attenuator out, to the circuit... resistance tracks perfectly. I think this would take the attenuator out
of the possible causes.
   I am considering the nuclear option now, taking the pot out of the circuit and just doing some basic
measurements at the front end of the circuit. The one shinning light is I have one normal channel to compare to