You tested each wire individually (separate) and other prob to chasis?
So are you say the wire from board negative wire grounded?
There is one positive tube(circuit) and one negative tube(circuit).
The tube that had the burnt resistor should be the wire but we tracing the location of the short by disconnecting the wires from the XLR jacks.
You only needed to do this on the side you had problems (right).
Anyways we cannot rule out that there could be a soldering mistake there.
The reason for testing wires is to locate the area of short so if the wire grounded is coming from the board side then ,
Before you pull out, recheck the soldering and connections on the resistor anode area.
I disconnected the ground and -ive output wires from the right hand XLR, these were the 2 wires that were showing continuity at the XLR when there shouldn't be continuity.
I then tested at the right hand XLR with the other probe at the +ive output connection on the board.:
1 - The +ive pin tested correctly. This output wire was left on because it was not causing the problem.
I then tested at the right hand XLR with the other probe at the -ive output connection on the board, the connection from pin 3 of the faulty tube circuit.
2 - The ground pin only showed continuity. This indicates that it wasn't the output wire causing it, because I had disconnected it. So the right hand -ive output connection from the board was obviously going to ground.
I then tested at the right hand ground XLR pin with the other probe at the chassis ground:
3 - It showed continuity correctly and none of the other pins showed continuity which was correct.
I then tested the ground output wire at the XLR end with the other probe at the chassis ground. This wire I had disconnected:
4 - It showed continuity correctly showing ground.
Then I tested the right hand -ive output wire at the XLR end with the other probe at the chassis ground. This wire I had disconnected:
5 - It showed continuity indicating a short somewhere before, or at the board -ive output connection.
Then I tested the disconnected right hand -ive output wire at the XLR end with the other probe at the -ive output connection on the board:
6 - It showed continuity which is what you would expect in a correct circuit but....
Then I tested the distconnected right hand ground wire at the XLR end with the other probe at the -ive output connection on the board:
7 - It also showed continuity indicating that the -ive wire was grounded.
Not much point checking the soldering around the anode resistor because it melts every time I switch on due to the power draining out. But this was a pre-existing problem, it was not the soldering that caused it. The other connections look good.
Edit:
I also tested the suspect right hand circuit tube pins 3 and 5 with the other probe at chassis ground, and both pins were grounded which was incorrect.
These results taken with the tests on the bad circuit continuity I did on the last page indicating continuity on many parts of the bad circuit which should not have been the case, I believe indicate that
somewhere on the circuit on the board is the problem, not in the wiring. That could explain the incorrect continuity readings in the bad circuit that I found during testing.