# Re: A 20 kV DC tank supply

```

tesla-at-pupman-dot-com on 15.12.98 11:18:35

To:   tesla-at-pupman-dot-com
cc:    (bcc: Marco Denicolai/MARTIS)
Subject:  Re: A 20 kV DC tank supply

Original Poster: Scott Stephens <Scott2-at-mediaone-dot-net>

>At 02:04 PM 12/14/98 -0700, you wrote:

>What the advantage of a DC supply, if it is not used to integrate many
>cycles of a power supply to charge a cap to a higher voltage than would be
>otherwise possible. Such as is done with a camera strobe light. The
>milli-watt average power inverter oscillates hundreds or thousands of
cycles
>to integrate watt-seconds power in the photoflash cap, which is then
>discharged in milli-seconds for kilowatt power levels.

>You get to use a much cheaper, smaller, compact supply because you are not
>going at a high repitition rate. If you tried, the solid state drivers
would
>burn, inverter transformer saturate and capacitor would explode.

My idea is:

- to have a high repetition rate, up to 800 BPS
- before each bang the primary capacitor must be charged at a specified,
precise voltage

Using 3-phase rectification that means the power supply own capacitor must
store enough energy for about 3 bangs and 3 short circuits performed by the
spark gap.

I am going through transient calculation and I will mail the list with my
findings and my equations. At the moment it looks like:

- a rectified tank voltage of 20 kV DC
- a 30 uF tank capacitor
- a 0.1 uF primary capacitor
- a 1.6 Henry, 10 ohm limiting inductor (feeding spark gap and primary
capacitor)
- a maximum quenching time of 80 us

will limit the maximum tank supply current to 1 ampere (on the DC side).
The tank voltage will drop 1% maximum from its nominal value.

>Radar and linac's are DC appliances.

In my knowledge radars use precisely shaped DC PULSES, not plain DC.

```