View Single Post
  #5   Report Post  
Old March 19th 12, 11:36 PM posted to rec.radio.amateur.homebrew
highlandham[_3_] highlandham[_3_] is offline
external usenet poster
 
First recorded activity by RadioBanter: Jan 2010
Posts: 63
Default How to measure high tube amplifier voltage

On 18/03/12 17:14, Kenneth Scharf wrote:
On 03/10/2012 05:56 AM, Michael wrote:

Most multi-meters only measure voltages up to 1000 volts. How do you
measure voltages often used with tubes like 3-500z or 3cx1500a7
tubes? Will the range extended from Fluke listed below work with a
1000v 10 mega ohm multimeter?

80K-6 High Voltage Probe
http://www.fluke.com/fluke/usen/acce....htm?PID=55358


The multimeter I have is the Radio Shack 22-801 which is rated for up
to 1000 volts.

Radio Shack 22-801 owner's manual
http://www.radioshack.com/graphics/u...0801_PM_EN.pdf


Thanks

Michael

I actually built a HV meter for an amplifier project years ago using a
surplus 0-1ma panel meter and a string of 10 meg ohm resistors. The
resistors were 1/2 W 5% and the entire string was covered with two
layers of heat shrink tubing for insulation (heated to shrink it around
the resistor string). I hand selected the resistors measuring their
value with my Fluke mutimeter so I actually had an accuracy better than
5% in the multiplier chain. The panel meter I used was intended to be
inserted into a HV circuit so the case was insulated to withstand the
voltage with the meter panel mounted. While I connected the meter across
the full plate supply voltage, if you are using series connected
capacitors in your plate supply (with divider resistors across each cap)
you can connect your voltmeter 'tapped down' on the divider chain and
multiply the reading by the divider ratio. This will be safer than
measuring the full voltage directly, but you will lose some accuracy
depending on the tolerance of the divider/bleeder resistors across the
capacitors. In most cases you don't need to know the plate voltage to
better than 5% anyway.

=======================
Perhaps it is useful to mention that non-HV resistors should not be
´exposed´ to more than 500 V , meaning that for a measuring range of for
example 5000 V any string should have not less than 10 resistors
(whatever the resistance values required )
Personally I would tap a relatively low voltage from a loop string with
on the tap a ten-turn trimpot in series with a fixed resistor.
If the DVM would have a max range of 1000 V I would first by means of a
variac adjust the PSU´s output to 1000 V ,directly measured with the
DVM . Thereafter with the string in place I would connect the DVM to the
tap and adjust the ten-turn trimpot for the correct multiplication
factor with the DVM range set at a range lower than 1000V . The ten-turn
trimpot would be exposed to a low voltage differential and serve only
for fine adjustment. All this would result in a measuring uncertainty
of better than 5%.(assuming the DVM has a measuring uncertainty equal or
better than 2% )

Frank GM0CSZ / KN6WH