Just need the tube and a floating grid driver and a 4500 volt isolated
filament transformer. All surface mount of course.
I just got some Ixys mega-dpak parts to play with, so they seem to be
available. I'll crank up the old Spellman and blow some up.
Reply by Joerg●July 15, 20212021-07-15
On 7/14/21 9:19 PM, jlarkin@highlandsniptechnology.com wrote:
But a tube costs less and, if picking a Russian one, is more available
right now :-)
--
Regards, Joerg
http://www.analogconsultants.com/
Reply by ●July 15, 20212021-07-15
On Thu, 15 Jul 2021 05:01:43 +0000 (UTC),
DecadentLinuxUserNumeroUno@decadence.org wrote:
>jlarkin@highlandsniptechnology.com wrote in
>news:8gdveg1188en4vg0bphd53vsicu588ce42@4ax.com:
>
>>
>> https://www.digikey.com/en/products/detail/ixys/IXTT02N450HV/373763
>> 3?s=N4IgTCBcDaIJYA8AuSAMYB2IC6BfIA
>>
>> Rds_on is < 625 ohms! More like a tube!
>>
>> It sure Spices nice.
>>
>> https://www.dropbox.com/s/wlpc73azpm0iyzn/T855_Totem_2.jpg?raw=1
>>
>>
>>
>>
>
> Ya saw how far apart them thar legs are.
>
> At those potentials everything would need to be potted.
>
> Incorporate at full potential very carefully (planning and
>engineering).
At 40 volts/mil creepage, that fet is good for 8 KV or so. One could
put a little slot in the board too. Or a dab of conformal coat around
the pins.
I'd expect the silicon to break down before the pcb insulation does.
I'll have to test one and see what actually happens. I have an old
Spellman 30 KV supply that might still work.
--
John Larkin Highland Technology, Inc
The best designs are necessarily accidental.
** OK - which tube then ??
How about the venerable 6BQ5 ??
With 100V on the plate, it will pass over 150mA = 600 ohms Rpk on.
Max plate V is speced at 1100V - it will easily take double that without breakdown.
Transconductance is about 12mA/V = well less than a MOSFET, but not bad thing with kV of output signal involved.
Dissipation is 12W long term, 10 times that ( 150mA at 800V) for a few seconds without failure.
That big lump of steel ( the plate ) takes some time to get red hot.
OTOH the MOSFET chip will fail in sub 1 mS if abused like that.
Tubes can actually arc over internally from extreme plate voltages without failure, particularly all glass types like the 6BQ5.
AKA the EL84 to non Yanks .....
FYI:
I have a test rig here that uses a pair in P-P, 650V on the plates and 325V on the screens.
Operated in Class AB.
Outputs 60W of clean, sine wave power without distress.
...... Phil
Reply by Rhydian●July 15, 20212021-07-15
On Wed, 14 Jul 2021 21:19:37 -0700, jlarkin wrote:
That data sheet is a bit sneaky. The DC specs (fig 13 and 14) are
nowhere near as impressive as the headline figures.
But, given that a lot of your designs are about making crazy-fast pulses,
that might not matter.
Reply by ●July 15, 20212021-07-15
jlarkin@highlandsniptechnology.com wrote in
news:8gdveg1188en4vg0bphd53vsicu588ce42@4ax.com:
Ya saw how far apart them thar legs are.
At those potentials everything would need to be potted.
Incorporate at full potential very carefully (planning and
engineering).