MCI Tube Compressor(now with some gear porn)

[MagnetoSound]

ON the non working channel, If I connect just plate on the second triode, I got around 105VDC. The moment I connect in the plate on the first triode stage, the voltage to each plate on that tube is even but way low. More over both plates are fairly even in voltage on the non working channel.

If I'm reading this right, doesn't that suggest some kind of short in the first triode?

What are the respective cathode voltages when both triodes are connected, compare to just the second triode?

 

[joe-electro]

When in doubt, RIP IT OUT!  (Replace it)

Joe

working it down from big to small. I have arrived at this. If I connected the 6386 on the working channel, voltage between the 2 plates of the 6386 is almost identical. ON the non working channel, If I connect just plate on the second triode, I got around 105VDC. The moment I connect in  the plate  on the first triode stage, the voltage to each plate on that tube is even but way low. More over both plates are fairly even in voltage on the non working channel. Then this takes me back to one working channel and one non working channel with a 30VDC difference in plate voltages between the 2 X 6386 tubes at one per channel.

I hate tested all the grid resistors, all the surround caps etc and cannot seem to find any issue either in wiring or component failure.  I am beginning to suspect that although unlikely, the tube socket for the 63686 tube is shorting out and or has become conductive.  it's the only thing else I can think of. i have been through all of it triple checking every thing I can't find anything else wrong.

hmmmmm?

 

[pucho812]

ON the non working channel, If I connect just plate on the second triode, I got around 105VDC. The moment I connect in the plate on the first triode stage, the voltage to each plate on that tube is even but way low. More over both plates are fairly even in voltage on the non working channel.

If I'm reading this right, doesn't that suggest some kind of short in the first triode?

What are the respective cathode voltages when both triodes are connected, compare to just the second triode?

I would have thought so except this happens with or without a tube installed and  stays with the socket when tubes are swapped between channels.


Joe I for one do not wish to ripe out anything willy nilly but a hunch is a hunch.

 

[joe-electro]

Not willy-nilly, but you've tried everything else.  I'll bet your instinct is serving you well this time.  It's a philosophical debate, whether it's worth the time to troubleshoot without ripping out a potentially bad component, and how much difference it will make to the sound to keep it in the new component in there if the old component turns out to be okay. A tube socket is more effort than most components, but they can be as troublesome as any.  Also, I seem to have more trouble with tube sockets on units that are wired point-to-point, so that might be another reason to consider replacing the socket.  One other suggestion, I've had quite a bit of trouble with the white, ceramic, Chinese sockets that seem to be so popular these days, especially the gold plated ones, so I would recommend a good quality bakelite or plastic socket.  I know that goes against popular wisdom, but that's been my experience.

This is one of my favorite units of all time and I can't wait to get a chance to build a version of it. I guess that's why I'm taking an active interest in it and hoping you can get it tip-top ASAP.

Joe

ON the non working channel, If I connect just plate on the second triode, I got around 105VDC. The moment I connect in the plate on the first triode stage, the voltage to each plate on that tube is even but way low. More over both plates are fairly even in voltage on the non working channel.

If I'm reading this right, doesn't that suggest some kind of short in the first triode?

What are the respective cathode voltages when both triodes are connected, compare to just the second triode?

I would have thought so except this happens with or without a tube installed and  stays with the socket when tubes are swapped between channels.


Joe I for one do not wish to ripe out anything willy nilly but a hunch is a hunch.

 

[MagnetoSound]

I would have thought so except this happens with or without a tube installed and  stays with the socket when tubes are swapped between channels.

OK, that wasn't clear before, could indeed be the socket. Maybe take some resistance measurements across the pins? Sounds like it would be in the kilohm range if there was some contamination or other short happening there.

Does it look like arcing may have happened at some time? Carbon is quite easy to remove with isopropyl and an old toothbrush. Whiskers of wire can find their way in. All just conjecture of course, yours is as good as mine ... 

 

[lassoharp]

9 pin sockets have very small clearance between pins.  Add solder glob and even less so.  As, Magneto mentioned, wire whiskers or other semi conductive wisp.  Have spent many a hunched over hour with a magnifying glass on 9 pins.

 

[lassoharp]

I would have thought so except this happens with or without a tube installed and  stays with the socket when tubes are swapped between channels.

Is there anything else tied in to the plate circuit?  With tube out, you would normally see the entire supply voltage sitting at the plate pin (assuming good coupling caps).

You could disconnect the cathode ground from ground on bad channel and see if that brings plate voltage reading back up.

 

[pucho812]

I would have thought so except this happens with or without a tube installed and  stays with the socket when tubes are swapped between channels.

Is there anything else tied in to the plate circuit?  With tube out, you would normally see the entire supply voltage sitting at the plate pin (assuming good coupling caps).

You could disconnect the cathode ground from ground on bad channel and see if that brings plate voltage reading back up.

on the 6386 pins? no. Coupling caps are good. didn't get a chance to disconnect the cathode ground yesterday. Will try it today and see what's what. Will be replaceing the tube socket today and see what's what.

 

[maxwall]

Nice MCI unit , I'm a fan.

Clip one leads off of every resistor and test , I find that old tube gear suffers from drifting
resistors open , high or low, you name it.  Especially the old carbon composition types, they tend to get
cracks in them. resistors in tube gear fail more than capacitors , in my experience. It can go either way.

Also ground solder joints to chassis tend to seperate from chassis metal too.

Are the meters in line with any audio path , thats the MCI way ?

all the above are things I have found as issues in MCI equipment. Go after the simple and overlooked
issues first. I suppose you already know this but I see it time and time again.

 

[pucho812]

Nice MCI unit , I'm a fan.

Clip one leads off of every resistor and test , I find that old tube gear suffers from drifting
resistors open , high or low, you name it.  Especially the old carbon composition types, they tend to get
cracks in them. resistors in tube gear fail more than capacitors , in my experience. It can go either way.

Also ground solder joints to chassis tend to seperate from chassis metal too.

Are the meters in line with any audio path , thats the MCI way ?

all the above are things I have found as issues in MCI equipment. Go after the simple and overlooked
issues first. I suppose you already know this but I see it time and time again.

right... I fixed my fair share of mci machines and consoles. so I know what your saying

 

[lassoharp]

I fixed my fair share of mci machines

How many molex sockets have you replaced?  So glad my MCI doesn't use those.

 

[pucho812]

I rather not talk about it. It's embarrassing.

 

[pucho812]

tested without grid resistors connected, and replaced tube socket. with no tubes in the 6386 sockets, I test around 108VDC for plates. This makes as ther VDC is coming after a voltage regulator tube an 0B2. With a 6386 plugged into channel 1, the voltage drops for the plates down to about 74VDC.  The socket for channel 2 measures 108VDC for the plates. When CH1 and CH2 tubes are in, I get 74VDC for plates on CH1 and 37VDC for plates on CH2. With the grides connected, I get no changes in measurements.

 

[joe-electro]

Can you sketch out the 6386 stage and the power supply path that's feeding it? Usually when I run into a roadblock like this it's because I'm looking in the wrong place or I have a misconception about the circuit.  Maybe a quick sketch will shed some light on where to look next. Also, does either channel work the way it should?

Thanks,

Joe