From: Mike Harrison
Subject: Re: High Voltage Power Source from Monitor Third Attempt
Organization: White Wing Logic
X-Newsreader: Forte Agent 1.9/32.560
Date: Sat, 16 Nov 2002 10:59:19 +0000
NNTP-Posting-Date: Sat, 16 Nov 2002 10:59:42 GMT
Resistors have voltage ratings as well as resistance ones. Normal
resistors are usually rated no more than 500V, and in the extreme case
will flash over as you have observed.
For high voltage use, you really need metal glaze resistors, designed
for high voltage use.
For the 25KV rating you need, you will need to use a number of
resistors in series. The easiest resistor to obtain with a HV rating
is the Philips/BC VR37 series, rated 3500V, so about 6 of these in
series will get the required voltage rating, although I think these
are only avaiable in values above about 1M, so you will need a lot of
series strings in parallel to get the resistance you want.
I've had success using 1 & 2 watt resistors at about 6KV for short
periods, so this might be worth trying - maybe use 4 in series (4 x
47K to get the 200-ish K). It will also help to coat them in silicone
sealant or conformal coating to reduce the risk of flashover.
Heatshrink sleeving is better than nothing. Snug-fitting silicone
sleeving is also really good.
If you can get them, solid carbon (as opposed to carbon film)
resistors are also good for high voltage use, but they are not very
A few simple things you need to know about combining N resistors, of
resistance R, power rating W and voltage rating V, assuming they are
the same value :
Resistors in series : resistance is R*N,, wattage is W*N and voltage
rating is V*N
Resistors in parallel :
Resistance is R/N, wattage is W*N, no difference in voltage rating is
Power rating will probably not be a major issue for you , so I
wouldn't worry too much about it.
On 15 Nov 2002 17:45:05 -0800, Thorfirstname.lastname@example.org (P!erCer)
>Some of you may remember me from my previous posts...my goal has been
>to create a high voltage power source out of a monitor that I can use
>to power my lifter (http://jnaudin.free.fr/html/lifters.htm). I am a
>complete electronics newbie, and my first two attempts ended in a
>broken monitor. The first time, it worked well for awhile but then I
>shorted the high voltage and burned out the HOV transistor (I think).
>The second time, I was all set with a resistor in the line
>(220,000ohms). The resistor was only rated at half a watt, but I
>bought 10 so when it blew I could replace it. Unfortunately, my
>device shorted and I saw the spark arc AROUND the resistor (frying yet
>another HOV transistor..actually this one tests fine. But something
>is wrong). The third time, I am going to get it right. I have been
>reading a bunch about electronics, and here is my design:
>I take the high voltage output from the flyback and run it out of the
>monitor through a drilled hole to a glass bottle filled with sand,
>towering high above the table surface. On top of this bottle is a 3x3
>grid of 220K Ohm resistors rated at half a watt, which I understand
>effectively forms a big resistor capable of dissipating 4.5 watts. I
>will completely insulate these resistors with 100% silicon goop,
>avoiding the arcing problem that ruined my second monitor. From the
>resistor array a thin wire trails to my device. The ground from my
>device then goes into a wire that goes into the monitor and is
>soldered to the metal braid around the CRT. Another wire is soldered
>to this braid, and goes to the end of my discharge probe (a yardstick
>with about an inch of bare wire on the end).
>From what I understand, the resistors should protect the HOV
>transistor a lot more than before (I was supposed to get ones of 240K
>ohms, but couldn't find that particular value). According to Ohm's
>law, a current of .11 - .15 amps will flow through my resistor
>(depending on voltage which I cannot measure or predict, should range
>from 25K - 35K)...will that protect the circuitry enough? I will make
>great efforts to avoid shorts this time, in any case.
>I have to go to karate now, please let me know what you think.