I've been reading these posts with interest. Some of my practices are a bit
different from those discussed here.
(1) HV wiring: I use ordinary red-jacketed RG-58 cable with Kings HV BNC
connectors to connect power supplies to amplifiers. My 8877 amp runs 4600V
(no load) and the cable has been in place for the past thirty years and
disconnected/connected many times with no problems. For years the high
energy physics group at Ohio State used the stuff for photomultiplier tube
power supplies, where the voltage runs in excess of 5 KV. That's where I got
my cable and connectors. The worse case scenario is that it would short to
the braid and blow a fuse. RG-59 would be a bit better, though unnecessary
IMHO, but my available connectors were sized for RG58. The key point is
always to use shielded cable for HV interconnects, for safety reasons. Also,
aside from safety considerations, uninsulated wire, such as used in the
Drake L4B and L7, will get extremely dirty over time because the electric
field around the wire will attract dust and grime.
(2) I don't quite understand the rationale for back-to-back diodes between
B- and Ground. I've always used a single safety diode for protection, with
cathode tied to B-, since a B+ flashover will drive current from chassis to
B-. (Actually, I put one diode in the power supply and another in the RF
deck.) Back-to-back diodes would likely distort the grid current reading
because one of the diodes will clamp the voltage across the grid current
sampling resistor ( I use 5 or 10 ohms.) An internal tube arc will channel
the current surge directly into the B- lead (if the arc is to the cathode),
or to the chassis (if to the grid), and in either case a single diode will
provide adequate protection.
(3) For B- wiring, I just use one of the conductors in the multi-conductor
control cable to the power supply. The B-voltage is never more than a couple
of volts above chassis ground, even with a flashover, so there's no need for
a separate B- wire.
73,
Jim W8ZR
_______________________________________________
Amps mailing list
Amps@contesting.com
http://lists.contesting.com/mailman/listinfo/amps
|