20 years overdue for this DC3 trouble

Hi all, I'm a long time owner of an awesome DC3 that's had a gremlin since day 1. Side note, the loaner amp I had gotten in year 1 had the same problem.
I've got a bit of ELT education and I'm a tech for a living but not for anything like tube amps. So this is half for hobby and half because this amp is too important to me to get rid of.

Mesa DC3 (I believe first version). Channel "blending".
Since I've gotten this amp in the 90's it's had the most annoying intermittent issue. When on CH1, after the amp gets hot, CH 2 starts to "bleed" through. I'll hear a second sound with heavy distortion that sounds like a lower volume version of CH2. The Ch2 gain effects this sound completely. The CH2 bass/mid/high have a slight effect on the second sound. The original Rhythm channel sound is completely unaffected.

In the first year I sent the amp back for warranty service and they sent it back with new tubes. That worked for a week. I ended up putting in a fan to "fix" the issue. Now I want to really fix the issue. So here's where I'm at.


*I've swapped and rotated every tube. Can't isolate it to a faulty tube.
*I did the bias mod to no avail.
At first I thought one of the mechanical relay contacts might be getting sticky when it's hot but I can't get it to "unstick" when I tap them. With a DMM I see a DC voltage drop across the NC contacts when they're supposed to be open. So they seem to check out. Plus, I believe if the NC contacts are sticking closed then I'd get no sound at all when I drop the CH2 Master volume because that'd shunt the signal to ground, but the CH2 Master knob has no effect on the other sound.

I'm leaning towards the LDR's not fully resisting but I cant figure how to test them while the amp is running. What kind of reading am I supposed to get across those contacts? When I do get my Fluke 87 in there (volts or ohms) it seems to be introducing too much noise to get a good reading. What can I do to test them short of pulling them out and replacing them? It looks like it'll take a full day to get to the bottom of the board. I don't even know what's a good replacement part. I have the fluke and a simpson 360 at my disposal.

Any and all help would be appreciated. Also, I'm curious if anyone else had this issue with the DC3. Strange that I had this with both of the ones that I encountered.

PING! And a small update in case anyone sees this.
So I just ohm'd out the "switch" contacts for the LDR1 in question and it was reading at 23 ohms when physically hot and only went up to 28 ohms when cold. Both de-energized. I wasn't sure that I wasn't getting readings from the rest of the circuit so I cut it out and got the same readings. The thing is definitely shot. Off to find replacements.
I'm still open to all and any advice.

Good luck! Please continue posting updates here. Thanks!

Will do. It was an 11 year old thread that got me this far so I'll try to pay it forward.

So, is there anyone that can direct me to the right LDR I don't know if Mesa makes it hard to find some parts or they just want to have their own specs on their equipment.
I've heard different opinions on what will work and what should. Also I want to know the prevailing theory on why these keep failing.

According to an email reply from a tech at Mesa the replacements are 5C9.
And according to another thread from 2011 there were a batch of bad LDR's that were installed in a bunch of amps in mid 90's. Also that Vactrol is a more reliable manufacture for them.

Can someone tell me if you've had luck swapping these out from the top? Or am I looking to pull out the board?

Thanks!