mwhouston wrote:
I replaced them with 100uf 400V. Yes over-kill but all I had. I don't think the 100uf will hurt anything.
This is correct. All that changing the 50µf to 100µf will do is lower the low frequency half power point on the output stage to 4.0Hz (-1.6dBv@20Hz). The output transformer is still the limiting factor on the low end rolloff.
Glad to hear that the Dragon has risen again.

B.T.W.
mwhouston wrote:
To protect these very expensive and limited edition tubes, fuses in the plate cct. may have been a good idea. Most say you can "hear" a fuse.
I believe that this complaint is usually due to the mechanical contact issues with the way most people mount fuses. In this situation I would generally use a pigtail fuse
like this one, with some heat shrink tubing over it, wired directly into the B+ leads at the output transformers (one for each channel). This requires a soldering iron to fix but they should only blow when something else fails so it shouldn't matter too much. These fuses should be "silent" in the amp.
The key is finding the right value. The current rating is simple, it's the I^2*t value that is difficult. This is a measure of how fast the fuse will open given a specific fault current. It needs to be long enough to handle the initial transformer core magnetization inrush current, but small enough to open circuit quickly in the event of a real circuit fault. It's a balancing act. Let me know if you'd like to go this route and I'll see what I can find. I'll warn you now, it will probably take a little experimentation to find the right value.
Keep us informed.