Ballast resistor construction

For posts specifically relating to fusor design, construction, and operation.
Post Reply
Dan Knapp
Posts: 402
Joined: Wed Aug 06, 2008 8:34 am
Real name: Dan Knapp

Ballast resistor construction

Post by Dan Knapp »

After failing to find any suitable large high wattage resistors at hamfests and flea markets, I finally ordered some from China on Amazon. I bought two 100K/100W ceramic wirewound resistors for $10.67 each (see attached photo). These aren't specifically rated for high voltage use, but I figured the physical length would be adequate. True high voltage resistors of this size are usually custom made and extremely expensive. For example, those shown in the attached photo "Osaka Ballast" (the brown rods) from a lab I visited in Japan were custom manufactured and I was told cost "many, many Yen." The resistors I ordered took two weeks to arrive from Hong Kong.

I drilled out the included ceramic bushings and used them to mount the resistors on a piece of 3/8" glass tubing. Using short pieces of larger bore glass tubing to center them, I mounted the resistors inside a 20" length of 4" diameter lucite tubing (photo ballast1). The whole assembly was mounted vertically on the vacuum chamber rack to facilitate convective cooling (photo ballast2). If needed, I'll add a blower at the bottom to facilitate cooling. I plan to add a liquid crystal thermometer strip inside the lucite tube to monitor the temperature, since I've managed to fry some previous ballasts.

As wirewounds, these resistors could also add some inductive snubbing of arc pulses. I tried to measure the inductance of one of the resistors, but it was off scale on my device, which is only designed to go to 700 uH (reading was 2500, but I have no idea if this has any meaning.) A possible embellishment could be to add an iron or ferrite rod inside the glass tubing to increase the inductance, but this might create a breakdown path.

I thought this might be useful info to others building a ballast. The price is not bad for resistors of this size. The mounting materials actually cost more than the resistors if you don't already have suitable materials on hand. I have not actually used the ballast yet; I will update this report when I do.
Attachments
100K100W resistors.jpg
Osaka Ballast.JPG
ballast1.jpg
ballast2.jpg
Roberto Ferrari
Posts: 364
Joined: Tue Mar 02, 2004 12:21 pm
Real name: Roberto Ferrari
Location: Argentina
Contact:

Re: Ballast resistor construction

Post by Roberto Ferrari »

Hi Dan

Excellent design!
How far will you push it in terms of kV?
Roberto
Rex Allers
Posts: 570
Joined: Sun Dec 30, 2012 3:39 am
Real name:
Location: San Jose CA

Re: Ballast resistor construction

Post by Rex Allers »

Nice work, Dan. One minor suggestion, possibly not necessary... At minimum, cut or file the sharp square corners of the resistor terminals to make them rounder. Better would be to also add a metal shape or build with solder a rounded or blob shape to avoid easy arc flash starting points (pun intended).

My experience: I was testing a 50 KV supply for voltage and current. I have a load I made from a bunch of power resistors in series to a total of 900 kohm. They are mounted on 4-inch ceramic standoffs above a ground plane. Not enough resistance to fully take the supply output. So first I hooked up the load and slowly cranked up the voltage to 10 mA out at about 9 kV. That proved the supply could make expected current. Then I wanted to check the supply unloaded max voltage. The easiest way to disconnect the load was to disconnect (float) the ground end of the load. As I cranked up the supply to near 50 kV I had a few sparks (short snapping arcs) at least a couple inches long to a grounded wire near the end of the resistor I had disconnected. These arcs came from the sharp corner of the resistor terminal I had floated. I didn't expect it to be able to arc that far.

It made me consider ways to avoid any sharp corners on the HV wiring. No harm done in my test and I had accidentally configured things so the HV was passing through the .9 M resistor to the arcing point so it happened to have some limiting too.

Just a note on safety, my load resistor and a HV divider for measuring voltage were well away from my operating point of the supply and were on a ceramic tile floor. I was never close to these parts when voltage was present.
Rex Allers
User avatar
Richard Hull
Moderator
Posts: 14992
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: Ballast resistor construction

Post by Richard Hull »

It is always important to figure the load and voltage gobbling power of your ballast. the two 100k in series is 200k. Regardless of voltages fed, be it 10kv or 100kv, it is the current that is the player here. At 10ma into the fusor, you would loose 2kv in the ballast here and 20 watts of power. Two 20 watt 100K resistors would have handled this. At 10kv that means you are down to only 8kv! The higher the voltage at any given current, the less deleterious effect the ballast will have.

The beauty of the large wattage is physical length of the resistor and the ability to swallow 1 amp arc conditions in the multi-millisecond range. At 45KV, my max voltage, and a 15ma run current, I use a single 100 watt 63Kohm power resistor placed inside the oil filled x-ray transformer that I modified. This means about a 675 volt, 11 watt loss in the ballast while in stable operation. I figure I could suffer a 500 watt arc for about 1 second. This might certainly blow my HV rectifiers via their current limits before the resistor opened up.

Placing the resistors in a plastic tube for HV protection without at least HV putty wadded over the resistor tabs, you might suffer an invisible brush discharge creating trapped conducting ionized ozone or air ions causing arcing.

Richard Hull
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
The more complex the idea put forward by the poor amateur, the more likely it will never see embodiment
John Futter
Posts: 1848
Joined: Wed Apr 21, 2004 10:29 pm
Real name: John Futter
Contact:

Re: Ballast resistor construction

Post by John Futter »

Dan
Do not put those resistors in the tube the ozone (read live ions) will build up and kazap

see this post I did a few years ago very similar resistors and corona rings at top and middle reduced corona current to acceptable

about 3 quarters of the way down the post

http://www.coultersmithing.com/forums/v ... f=45&t=757

other than that well done keep it up
User avatar
Dennis P Brown
Posts: 3160
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Re: Ballast resistor construction

Post by Dennis P Brown »

Like Richard's, my ballast resistor is under oil (simple synthetic motor oil - superior to even most x-former oils.) Besides the complete elimination of corona issues, it provides excellent cooling.
Dan Knapp
Posts: 402
Joined: Wed Aug 06, 2008 8:34 am
Real name: Dan Knapp

Re: Ballast resistor construction

Post by Dan Knapp »

I have a 60 kV / 60 ma power supply, but I don’t plan to go to more than about 40 kV. I’m only using a 30 kV feedthrough. I have solder blobs on the resistor tabs, but the putty is a good suggestion. I realize the resistance and wattage are probably twice what would normally be needed, but I was going for physical size and length. I appreciate that I’m wasting 2 kV per 10 ma, but I have ample margin with the supply, and I’m more interested in improving stability at this point. I’ve already fried some smaller resistors with current runaway. The problem I’ve been having is that a current surge will start the system into a runaway condition if I’m not fast enough backing off voltage. I don’t have an analog control knob for voltage (the supply only has a computer interface with 1kV increments) so I’m not able to develop the fine tuning skill Richard reminds us is essential. I’ve disabled the arc detector on the supply (by bypassing the current transformer at the bottom of the multiplier) and set the current limit high to make it work more like a brute force x-ray transformer supply, but without rapid fine tuning, it’s hard to keep it stable. I’m keeping the resistors inside the tube for safety and hoping convection will keep the ozone cleared, but I may need to add a blower at the bottom.
Thanks for the feedback on this.
Dan Knapp
Posts: 402
Joined: Wed Aug 06, 2008 8:34 am
Real name: Dan Knapp

Re: Ballast resistor construction

Post by Dan Knapp »

I failed to respond to John’s comment. If I encounter discharge problems, I will go to a larger acrylic tube and add three corona rings like you show in your photo. The photo from the Japanese lab shows two corona rings. They go over 100kV with this system in air. Incidentally, this is the system where they did the first tritium experiments in a fusor. I’m hoping to not have to deal with the hassle of an oil filled ballast.
User avatar
Rich Feldman
Posts: 1470
Joined: Mon Dec 21, 2009 6:59 pm
Real name: Rich Feldman
Location: Santa Clara County, CA, USA

Re: Ballast resistor construction

Post by Rich Feldman »

Was it Rex who first mentioned the badness of sharp points and edges here?
I would think that's important even when the dielectric is oil.

When one figures the maximum E-field strength from max voltage divided by distance between conductors, the answer is always too low. (Except between infinite parallel plates.) Real answer is somewhere between that, and max voltage divided by radius of curvature at a conductor edge.

Looking for design examples of rounded (or not) edges in oil immersed apparatus. Have seen rounded edges on x-ray transformer cores. Not so important when the adjacent coil winding layer is close to the core potential.

One project under way is a mains-voltage transformer with 30 kV of isolation on one coil. Starting with an existing transformer, replace one coil with a coil spaced much farther away from the core, and operate it under oil.
DSCN0606.JPG
Should core edges be rounded off with a file?

p.s. Also experimenting with an easy way to wind nice transformer coils: Unwind some inside and outside layers from a factory-built coil. But this is Dan's thread, about HV ballast resistors.
All models are wrong; some models are useful. -- George Box
User avatar
Dennis P Brown
Posts: 3160
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Re: Ballast resistor construction

Post by Dennis P Brown »

I have normal (semi-sharp) edges on my ballast but it is under oil, and has not been an issue for up to 33 kV. I have used drops of silicon sealant for that purpose when the device was first operated in air with success but again, 33 kV is not 40 kV (Nor the issues of rms.) A transformer core does not carry any voltage so not necessary to round those edges, I'd think.
Dan Knapp
Posts: 402
Joined: Wed Aug 06, 2008 8:34 am
Real name: Dan Knapp

Re: Ballast resistor construction

Post by Dan Knapp »

Finally got around to using the ballast resistor today. It worked fine, and I didn't see any signs of discharge (although I wasn't working above 20 kV). The idea of a putting liquid crystal thermometer strip inside the tube didn't work - too hard to see - so I just put a dial thermometer on top that I can read across the room. The temperature at the top got to 48 degrees C. max in two hours of running. After turning things off, the outer plastic tube was barely warm. It looks like it will work OK without forced air cooling.
User avatar
Dennis P Brown
Posts: 3160
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Re: Ballast resistor construction

Post by Dennis P Brown »

Glad the power load isn't too high. A word of caution: the difference between 20 kV and 30 kV can be significant relative to corona's and how the system behave's. When one gets over 20 kV, especially in fairly dry air, the arc over's can be unusual and even somewhat inconsistent.
Post Reply

Return to “Fusor Construction & Operation (& FAQs)”