Here is the exact method for determining the length of hose needed for a
particular water cooling system having High Voltage DC across the hoses.
This is how i do it for big amplifiers with high voltages usually over
10 kV. For ham use, one can be more relaxed with this, but I wanted to
demonstrate the right way to do it for those interested.
Water resistivity is measured with meters (Barnstead, Foxboro, Omega),
in ohm-cm or megohm-cm. Sometimes it is measured in inverse quantities
with mhos. This all depends on ionic content in the solution. If you
start with deionized or distilled water versus tap water, its a lot
easier to control. Tap water can have large variations in resistance,
usually too low unless you account for it, or keep the voltage low. It
is good practice to try and keep the hose leakage current under a
milliampere if you are planning to run in this configuration for a long
time (years). Higher current will cause corrosion of metal fittings
(electrolysis) and move metal to other places where you don't want it,
such as in the hot cooling jacket of the tube, the inside of hoses, and
filters. Iron pipe is terrible choice, use copper, stainless steel, or
quality plastic.
Figure out the hose cross sectional area in cm^2 using A = pi x radius^2
R (per cm length) = resistivity/area
Think of it as a cylinder of this resistance, 1 cm in length.
Using Ohms law, you know the HV you want for a tube, and use 0.001 amps
or less for I. This tells you what Resistance is needed, in each hose.
Length is easily determined now that you have the R per cm of length.
Convert everything to inches, feet, miles, if you want.
Commercial systems use resin beds (hey Culligan man?) to keep the
resistivity of the solution high. Not too high (theoretical limit is 18
Meg ohm-cm) though as it is very hard on metal components.
For the experiment I have been running all week, have 1.5 inch ID hoses
to the anode of a 4CW250,000 and the water is measuring between 1 and 2
Megohm-cm. I need 9 feet for inlet and outlet hoses from the tube to the
chassis to keep current around a mA.
73
John
K5PRO
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|