Continuing this thread, we used deionized/deoxygenated water for our RF
power tubes at the accelerator plant. I designed the anode cooling
tubing to have no more than 500 uA of leakage current per hose, based on
old RCA and Eimac application notes. This is at 23 kV DC plate voltage.
The resistivity always stays above 2 Megohm-cm for this. We have
Culligan bottles that are always bypassing a fraction of the 900 GPM
that cools the four transmitters, all of the tubes.
On the other hand, we use long water column dummy loads (no resistors
inside) that are about 5 wavelengths long at 200 MHz. They need some
conductivity, so we use the same deionized water, then dope it with
about 0.5% of a particular corrosion inhibitor that has a molybdate
salt. This yields a DC conductivity (1/resistivity) of about 490 uS/cm.
With this there is a 20 dB return loss in the load. It is tuned with the
concentration of solute.
Both the ultra pure deionized system and the ionized doped system are
closed loop with pumps and a heat exchanger, that has cooling tower
water on the other side.
We have some air cooling for things like filament stem under tubes, and
heat sinks/resistors in various power supplies, but the majority is
water cooled 24/7.
73
John
K5PRO
FYI: I'm afraid I have to disagree on the statement below about
distilled water being conductive.
In industrial experiments in the use of distilled water, I found it to
be very non-conductive. If I recall correctly, it has a measured
conductivity of 50 to maybe 100 micro-Siemens per centimeter squared.
That's not much. Pure water becomes conductive only if it becomes
contaminated with salt like contaminates. For a comparison, seawater
typically measures in the thousands of micro-Siemens per centimeter
squared, because of the high salt content.
While working on a clients project to do so, I spent many hours
attempting to inject RF into water of various solutions. It was very
difficult in the lab trying to get RF to propagate thru pure water.
Other more contaminated solutions not so much..
73,
Ray, W4BYG
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|