Re: Cooling MOFSETS and IGBT's.

["Tesla list" <tesla-at-pupman-dot-com>]

Original poster: "Jim Lux by way of Terry Fritz <twftesla-at-uswest-dot-net>" <jimlux-at-earthlink-dot-net>

Typical distilled water resistivity spec for HV cooling would be anything
above 2 Megohm cm

So, assuming your cooling tube had a cross section of .1 square centimeter,
and it were, say 1 meter long, the approximate resistance would be 2 * 100
/ 0.1 or 2000 Megohms.  Not much current flow.....

This level of deionization is typically achieved by those cartridge
deionizers, which often are installed in a handy fixture that has a
conductivity probe with a light/relay contact and a set point at, say, 2-5
Mohm cm, to tell you to replace to cartridge.

Alternately, when water cooling something at a known voltage, you could
just measure the leakage current.  Using the above example, and running at,
say 20 kV, the leakage current would normally be around 10 microamps

David's observation of 400K with the DMM probes is somewhat unusual... But,
a very very small amount of contaminants can dramatically reduce the
resistivity.  You really need that cartridge deionizer in there for the
initial flush out of the system.

Tesla list wrote:
> 
> Original poster: "by way of Terry Fritz <twftesla-at-uswest-dot-net>"
<Tesla729-at-cs-dot-com>
> 
> In a message dated 3/19/01 6:53:45 PM Central Standard Time, tesla-at-pupman-dot-com
> writes:
> 
> >
> > > Distilled water as the coolant would work quite nicely, and hold off
> > > the 30 kV.
> > >
> > > or, run entire water cooling rig "floated" at the HV.
> >
> > But will it remain distilled in contact with various voltages and
> > metals? I don't think I'd have water anywhere near HV of any sort and
> > the thought of dripping condensation doesn't appeal either if for no
> > other reason than corrosive effects.
> >
> > Regards,
> > malcolm
> >
> >
> 
> Hi all,
> 
> Malcom's right! And BTW, I tried to measure the resistance of distilled
> water vs. tap water one time years ago and although the tap water obvi-
> ously had much more resistance that the plain tap water, the distilled
> water did indeed have some conductance. If my memory serves me cor-
> rectly, I think by just placing the ends of the multimeter probes about
> an inch into a drinking glass full of each kind of water, the tap water read
> ~ 20 kOhms and the distilled water ran around 400 kOhms. I don't think
> this would be near enough resistance to hold off any serious high voltage,
> but would allow for some current flow, which in affect, would short it out.
> Example: 10 kV thru 400 kOhms = 25 mA of current leakage or 250 watts.
> And at 30 kV, there would be over 2000 watts of power wasted this way in
> the form of heat, assuming 400 ks of resistance! And I'm sure there would
> be more surface area of exposed oppositely charged HV conductors in
> this setup than with the two slender multimeter probes.
> 
> David Rieben