I keep blowing D-2/ B minus 5408 to ground, suggestions?

A while ago, I lost plate current and found it was the single 5408 diode from B- to ground ("D-2", bypassed w/ low voltage .01 ceramic). The diode went open/No Connection.

I replaced the single 5408 with two in series (same batch) bypassed w/ 2x .005 Z5U ceramics). I got zsac back but could not try it for awhile as I was waiting for the Termination Resistors to short unused turns on Roller. A few weeks later I fit them onto a heatsink/fan combo and was ready to try it again. Appeared to be working and after 4-5minutes I had a nice 40m AM carrier into DL from 5w drive. I began checking Temps and then one of the 5408's opened up again. Fun stopped no plate current.

This time I replaced the 5408's with a single 6A10 and it’s working to the extent I have zsac again but I’m starting to realize this may be a symptom of something else. How would I know if one of the HVPS diode stacks went bad? It’s a FWB. I’ve never had a problem with D-2 before and going to 6 amps is double the 5408. Can I check the FWB diodes without removing them from the HVPS board?

I’ve never needed anything larger than a 5408 for D-2 even when driving 4x 572b’s

I apologize for making 2 threads about the same issue. This is the same 2x 4cx300a amp w/ Tetrode boards I been tooling around with for awhile.

Brondon 7861 suggested decreasing the Termination resistance and I'll try that next if the 6A10 diode doesn't survive. Nomad HVPS board surviving the stress and I needed awhile to realize the B minus diode going open is from shorting the unused turns of the roller. I ordered another 100 ohm 250w RF Termination resistor to bring down total to 33R. Heatsink/resistor combo pic attached.

I had to reorder the Term resistors because the first R's used the flange as a terminal placing heatsink in circuit. The new ones (pic) have terminals for each end of resistor. Heatsink & fan decreased the temp a lot so maybe I'm good to proceed.

I remain surprised that I all meter readings stayed the same despite the RF Term R's getting so hot from shorting the unused turns. I guess it puts more stress on the HVPS and the B minus diode to ground was 1st to open from the added current. ZSAC unaffected too.