My CT-100 lost HV (from losing one B+ line).

[benman94]

Originally Posted by dtvmcdonald

Bad thoughts. Could it be arcing in the CRT? ....
This has me worried for my 15GP22.

It's a very real possibility, and I'd start looking for ways to rule it out.

My Great-Uncles Neil and Skip told me that the 15GP22 was a notoriously unreliable tube even when they were new. Many of them failed under warranty, costing RCA even more money, with arcing as a primary symptom. That's part (albeit a small part) of the reason that the 21AXP22 was rolled out in late 1954.

That said, I wouldn't worry too much. Tom's scenario sounds the most likely IMHO: all circuits on the +375 rail drawing excessive current with a tiny spike being the straw that breaks the camel's back.

[dtvmcdonald]

Originally Posted by benman94

It's a very real possibility, and I'd start looking for ways to rule it out.

My Great-Uncles Neil and Skip told me that the 15GP22 was a notoriously unreliable tube even when they were new. Many of them failed under warranty, costing RCA even more money, with arcing as a primary symptom. That's part (albeit a small part) of the reason that the 21AXP22 was rolled out in late 1954.

That said, I wouldn't worry too much. Tom's scenario sounds the most likely IMHO: all circuits on the +375 rail drawing excessive current with a tiny spike being the straw that breaks the camel's back.

Its clearly not a tiny spike on the 375 volt line. It normally runs in the
30-40 mA range, rock solid. I had a meter on the line for many turnons.

The solution is probably a big box of 250 fast blow mA fuses and don't use it
except for demos. A blown fuse is not a common happening.

But what tube CRT elements are arcing? The convergence electrode
is 1 nF fin series with 820K. The focus electrode is 7200 ohms in series with
the substantial inductance of the vertical convergence transformer and
the 10 nF to ground. That leaves the HV itself, which is about .002 to .004 nF
to ground, which is a lot of energy. But they would have to arc to the
screens.

The CTC2B has 100K resistors in series with the screens, filtered by
0.1 uF caps. 0.1 is 25 times .004, so even an instantaneous connection
would only put 800 volts on the cap.

Perhaps I should add that to each screen. But where? The caps are rather
large, and the resistor, during an arc, will have up to say 18 kV across it
(2 watt resistors, even modern ones, are plenty long enough to withstand
that though two 47K ones in series might be better). There's room in the
LV power supply cage, or underneath where it is in the 21CT55.

[dieseljeep]

Originally Posted by benman94

It's a very real possibility, and I'd start looking for ways to rule it out.

My Great-Uncles Neil and Skip told me that the 15GP22 was a notoriously unreliable tube even when they were new. Many of them failed under warranty, costing RCA even more money, with arcing as a primary symptom. That's part (albeit a small part) of the reason that the 21AXP22 was rolled out in late 1954.

That said, I wouldn't worry too much. Tom's scenario sounds the most likely IMHO: all circuits on the +375 rail drawing excessive current with a tiny spike being the straw that breaks the camel's back.

On a side note, regarding a set using a 15GP22. The set was stored in an attic for something like 25 or 30 years, that was freezing in the winter and extremely hot during the summer.
The CRT looked really good, after the set was restored. I guess, some things are meant to be, either pro or con.