High Voltage Probe
When I first seriously considered building a
Tesla Coil, one of the first things I bought was a high voltage probe (Fluke
model 80K-40) good to 40,000 volts. Now, with an operating Tesla
Coil, Cockcroft-Walton voltage multiplier and Van de Graff generator, 40KV
is way too low. Commercial test probes for higher voltages are available,
but are very expensive. So I decided to design and build my own.
The picture above shows a schematic drawing of
my design. The voltage divider circuit contains 25, 1 watt, 400 MW,
high voltage resistors and one 1.0 MW high voltage
resistor connected in series with crimp style butt connectors. This
gives a 10,000:1 division ratio and a voltage limit of 500,000 volts (to
dissipate 25 watts). The barrel of the probe is a 62" length of 1.5"
OD X 1.25" ID fiberglass tubing. The probe tip and the three banana
jacks (high potential, test ground and earth ground) [10/20/01: the
banana jacks were replaced with a BNC bulkhead jack] are mounted in
1/2" thick circles of polycarbonate connected to the barrel with 8-32 flat
head slotted nylon screws. Each butt connector is mounted in the
center of an 1/8" thick polycarbonate circle. The probe is covered
with PVC heat-shrink tubing (fiberglass makes me itch).
This picture shows the parts ready for assembly.
The polycarbonate disks were cut with a router and compass attachment.
The fiberglass tubing was cut with a carbide hacksaw blade and the end
trued and trimmed with a Dremel rotary tool with diamond cut-off wheel
(glass dust everywhere, seriously itchy too). The six holes through
the fiberglass tube for the end cap mounting screws were drilled with a
3/16" titanium nitride coated drill bit and counter sunk with an 82º
high speed steel bit.
This photo shows the parts being put together.
Most of the resistors are in the tube (except for the five assembled above
the tube and four which were still on back-order from Newark when this
picture was taken). Also shown are the two 24" insulated leads which
will run through the handle and are soldered to the banana jacks (which
are color coded so I can't screw up and put 500,000 volts through my multimeter)
[10/20/01: the banana jacks were replaced with a BNC bulkhead jack].
In this photo, you can see the ribbed handle of the probe and the read-out
of 80,400 (8.04 X 10,000) volts from the C-W.
A quick check of one of my small Van de Graff generators gave a reading
of only 18,000 volts, suggesting that the 10 giga Ohm impedance of the
probe is still drawing more current (1.8 mA
in this case) than the generator can deliver.
I'll try it out on my Tesla coil soon.
Tried it with the Tesla coil. Whenever the coil was running, the
multimeter display blanked out. I'll probably need an oscilloscope
and a Faraday cage to make this work. Time to cruise E-Bay.