[Amps] re: baking out tubes

[Traian]

Steve Thompson wrote:

> The breakdown voltages are
> > increasing by "debarnacling" for the NOS tubes but it seems that is not
> > for the heavy used ones? (at least not for a used 4S040 and 4CX1000 I
> > have tried).
> For what it's worth, my theory is that a well used tube with an oxide cathode
> has minute amounts of cathode material spread around the inside, and this
> leads to a leakage level that won't reduce.
>
> Steve
>

Hi Steve,

Your theory is true for sure.
Anyway, the 4S040 is a filament tube (similar with the 4-400C) so no
such problem, but material deposit may still exist, maybe because of
a previous arc inside the tube, during the use.

I got improvement of the breakdown voltage for all of the NOS tested
tubes (more than 15 tubes until now), no matter if oxide or filament cathodes.
The largest improvement was for the NOS russian tubes (up to doubling
the breakdown voltage) while for the west made ones this was lower
(a 4CX1500B, six  8122 and three 8121, all NOS have been tested).
For some GU43B as for the 4CX1500B the first sparks occured at
only 7 kV, while at the final they withstand up to 14 kV, a 40% more
than for the 4CX1500B, while they having similar anode DC ratings.

Some negative experience: further conditioning by runing the filament on
for a GU43B I have tried (for only 30 minutes, as I found recommended
somewhere) got no further improvement at all. Maybe I should keep
the filament on for more time, as recommended on the n2dx website.
I have verified a NOS 500 pF 5 kV russian vacuum variable also.
Its voltage breakdown  "headroom" was much lower, breakdown
occuring at little over 6000 V and no much improvement after the
"conditioning". Maybe the spark energy was to large because of
the vac cap's capacitance itself?! Its spark noise was quite diferent...
A risk exist, as the breakdown voltage may get lower if the sparks
have very large energy, and these devices are not cheap at all...
Variations may be caused by different optimal spark energy needed,
other than the available here, determined by the construction and
materials used inside the devices, as by the manufacturing quality as well.
Will have to try for another vacuum cap, and I have another filament
cathode used tube in line for testing which to eventually confirm (or not)
the previous finding about the used tube, when my spare time will
allow for this.

I have started experimenting after I saw the only 8 years of storage
allowed period specified on the GU43B data sheet, and I didn't
thought that I will go so far. And I had luck that a ionisation glass
ball have been accidentaly broken here at the QRL and could build
the hi volt tester using its remains, hi, hi...

Have a nice day,
73,
Traian