I am going by your description. Somewhere there is a bad ground connection or maybe a leaky heather/cathode interface. Turning the attenuator affects it. So either it's the attenuator contacts not making good contact or the tube - either its pin contacts or something internal - is being affected by the mechanical vibration created in the chassis by the clicks of the attenuator switch. The stock pot might not set the problem off, but of course we can't know that unless you put it back in.
The short story is if it was on my bench I could probably tell you exactly what is going on in a few minutes. But when a circuit has been modified in a multiplicity of ways and we have to rely upon written description rather than hands on testing the best we can offer is educated guesses. This whole notion of being able to talk a person though troubleshooting on a forum didn't even exist 25 years ago, and though we are pretty good at it, it's still a rather awkward and uncoordinated method of fixing something. At least it is far better than trying to troubleshoot over the phone.
I know we seem a little impatient at times trying to help people who - from our perspective - decide to improve our products, but find that the circuit does not work right after they do. That may be rooted in the fact that we spend a lot of time deliberating over an aspect of a component that is often not understood by those who substitute other parts, and we (mostly meaning Josh these days) also spend a lot of time making the assembly manual as specific as possible to guarantee proper function. Everyone hates it when we say "return the circuit to stock". But it always works right if you do, and we will know exactly what you are looking at on your bench when you need assistance. At the same time this is DIY and we believe strongly that people should experiment and learn. That includes learning how to systematically go through something that isn't working as expected and using some logic to eliminate possibilities until you arrive at a solution. Changing one thing at a time is a very good way to maintain control in the modification process, and being willing to go back to where you were before the problem started to help identify what is going on is a very good habit to have acquired. I laud your efforts in that direction.
To be a little more blunt about one aspect of this conversation - if the guys who sell those shitty adapters would support them, we wouldn't have to waste so much time trying to help people with them. This noisy 6SN7 adapter opera has been going on for years now. You would think by now they would have figured out how to make the damned things.
