FAQ - Measuring Fusor Voltage Part III - HV Divider

If you have a question about this topic, the answer is probably in here!
Post Reply
User avatar
Richard Hull
Posts: 11302
Joined: Fri Jun 15, 2001 1:44 pm
Real name: Richard Hull

FAQ - Measuring Fusor Voltage Part III - HV Divider

Post by Richard Hull » Fri Sep 13, 2013 3:26 pm

This is the third part of an original FAQ related to high voltage metering for the amateur fusor. If you have not read parts I and II, then I suggest you start back at part I and follow the threads sequentially. Go to…..


In this part of the discussion we will assume you have a proper metering situation related to parts I and II and require a High voltage divider network to take your low voltage metering to a point where it reads high voltage. This divider will consist of two fixed resistors, 1. (High voltage resistor), 2. (Precision shunt resistor) and a small, multi-turn calibration potentiometer.

The high voltage resistor is the most critical of all the components in any high voltage metering system. It is here where most tears and frustrations are generated. The high voltage resistor must answer three criterions:

1. Be able to handle the voltage you impress across it without breaking down, (arcing), end to end or have significant leakage current that will foul its accuracy.
2. Be able to dissipate the waste heat generated, as most of the entire high voltage appears across it to ground in the divider chain, thus, more power is concentrated here.
3. Be of a size that can be comfortably fitted into the experimenter’s enclosed power supply. (This resistor will be deadly when the circuit is energized and should be contained in a tamper proof environment.)

High Voltage resistors are very special in that they are physically long resistors and are often encapsulated with epoxy and fashioned on special ceramic bodies. They are extremely costly and difficult to obtain, in general. They are rarely found surplus or in old gear or at hamfests or tech flea markets. Often, the voltage is such that a series string of resistors must be fashioned in order to make the HV divider resistor not arc over. A real fusing fusor uses voltages at a minimum of 25,000 volts and many better system use voltages up to 80,000 volts! The average system uses up to about 40,000 volts. Most all of these systems will require a “series string” of resistors to help stand-off the voltage and dissipate the energy spent within the string.

A demo fusor rarely has a voltage impressed across it of 10,000 volts. If one is to make only a demo fusor, then a single high voltage resistor can suffice in the divider chain.

For the sake of this discussion, we will focus on the average working fusor of 40 kilovolts and figure on 100 megohms in the series HV string for each 10kilovolts we are to have up to our maximum. This will yield a 100 microampere divider current. (I=E/R) 10e-4amps = 10e4 volts/10e8 ohms. The power dissipated in such a resistor is P= I x E or 10e-4 x 10e4 = 1 watt.

Analog metering……….
Based on the above, a single, 2 or 5 watt, 100 megohm resistor of about 3 inches in length would be ideal for a 10kv power supply metering system and in parts I and II the analog meter of 100ua did, indeed use a single 5 watt 100 megohm resistor in series to read 0-10kv.

A 40kv power supply would need 4 of these same resistors in a series string to achieve our ideal of 1 - 100megohm resistor per 10kv of power supply potential to drive the analog 100ua meter to full scale at 40kv. (Remember, in this case,the meter scale would have to be altered to read 0-40)

Digital metering…………
Our digital meters were set up to read 0-20 volts in Parts I and II and we calibrated them accordingly. These meters are 3-1/2 digits and can read to 19.99 volts. If we try to use them for 40.00 volt in the string to read 40.00 kilovolts the ½ digit can only be the numeral 1. So, we will use the 19.99 volt meter to read 2 volts 1.999 so with a proper dropping resistor chain it can read up to 199.9 kilovolts instead of 19.99 kilovolts. This is why we activated the decimal point at 199.9.

With 100ua flowing in any resistor chain we use, 40kv will have to read 40.0 or 4 volts on our 20 volt meter. To have 40.0 volts, (indicating 40.0kv), on our meter at 100ua we will need a second “shunt resistor of R=E/I or 4x10e4 ohms = 4volts/10e-4 amps. Thus, a 40kilohm resistor is needed for our shunt resistor. I would select a 33k ohm 2 watt 5% fixed resistor and a 10k ohm, 10 turn calibration pot in series to be the shunt resistor. This would allow precision adjustment of the 40kv meter as the shunt could be adjusted between 33kohms and 43 K ohms.

Now this meter will read as high as 199.9 kilovolts! Don’t try it though!!! Your 4 - 100megohm resistor HV string was set up for only 40kv max! You could probably use it to about 50-60kv with ease though.

The images attached will show how it works in actuality. I used a 0-30kv spellman supply to supply the HV for calibration (super nice rig, but limited to 2ma). I chose a 100meg, 3 watt, 15%(ugh) resistor rated at 20kv for the 10 kv metering system HV dropping resistor. The temperature coefficient, (tempco), on this resistor is lousy!.... Meaning its value varies with its temperature. I needed ~10k ohms as the divider "shunt" and chose a 7.5k ohm flamproof I had laying about and a nice 5k ohm 10 turn Bournes helipot for the calibration adjuster. (7.5 K to 12.5k). This thing worked right out of the chute and adjusted perfectly! It tracked with no variation from 1 kv to about 13kv. As I went higher up to about 20kv the tempco of the resistor and its 15% rating showed itself as the resistor warmed. At 20.0 kv on the Spellman, I was reading 20.2kv and after 2 minutes 20.3kv. Turning the unit off and feeling the resitor, I noted it was not hot but quite warm to the touch. After cooling, I turned it back on and all tracked fine again until the 13kv range and resistor heating took its toll again of accuracy. You will notice in the final image at 9.9kv we are drawing about 98ua and are right on track for the 100ua / 10kv rule.

I have done a bit of leg work and as of September 2013 you could purchase from Allied Electronics, a single Ohmite, high voltage, Super Mox, 100 megohm 10 watt, 1% resistor rated at 45kv for $19.99 resistor! None were in stock and a minimum order of 10 would be required. They did have in stock # 7002308, 50 megohm 10watt, 1% resistors at $17.58 each so you could series those, but you see how fast it all adds up in bucks to make up these special HV resistors.

Could we use cheaper common resistors? Yes! 50 - 2 meg, 1 watt, 5% resistors in series would work great to make a 100 meg resistor. This would only put 200 volts on each resistor....Well within spec., but that would be one huge resitor. 20 - 5 meg 1 watt resitors in series would also be OK, but you would have 500 volts across each one and that would be a bit of stress, yet OK I would think. Shorter, but still a big puppy! If you go this route, avoid using carbon composition resistors here.

This concludes your "take-you-by-the-hand" instruction on high voltage metering of the fusor. Good luck.

Richard Hull
Here is the KEY component! The 100 megohm, 3 watt resistor. It is only a 15% unit. You should strive for at least 5%
Top view of the layout. Meter is wired across the shunt series string of the pot and a 7.5k resistor. The key HV 100 megohm resitor is in series to the hot lead of the supply
5.4 kilovolts applied and the 10 turn pot has been adjusted to match the meter in the Spellman supply
A view of the network and meter. Note HV resistor is on a thick insulator for testing.
Final test up to the string's rated 10kv input. The resistor was barely above ambient after 5 minutes.
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
Retired now...Doing only what I want and not what I should...every day is a saturday.

Post Reply