High Voltage Power Supply Design

This forum is for specialized infomation important to the construction and safe operation of the high voltage electrical supplies and related circuitry needed for fusor operation.
Post Reply
J16
Posts: 23
Joined: Wed May 04, 2005 9:42 pm
Real name:
Contact:

High Voltage Power Supply Design

Post by J16 »

Welcome to part 2 of my attempt to check my fusor’s design. For the sake of efficiency I have broken it up into multiple parts. Here is the power supply design I intend to use with my reactor. And yes, I have read through almost all of the information available on neon sign transformers on the forum, and I have read the FAQs and other important postings in this section.

In the document below I think I have covered all the major problems associated with using a NST to power a fusor. One major goal for this design, and all of my others, was low cost. These designs are made to be cheap and still exceed the minimum requirements for the reactor.

Known problems:

At the moment the reactor lacks a method of manipulating the output of the power supply. This is easily solved by incorporating a 500-watt variac. This detail was not included in the design document below, because it is not necessary for the basic testing of the power supply.

Also no information was included in this document on how to measure the current output or the voltage output. This will be covered in a separate document once this design has been completed.

The capacitors being homemade is not critical to the design, but it is something that I really want to try. Any suggestions on how to safely test them would be appreciated.

As a final note, I would like to thank everyone on the forum who helped me in refining the design of the valveless deuterium supply. :) All of the completed designs will be available online once they are all completed.
Attachments
Power Supply.doc
(59 KiB) Downloaded 288 times
User avatar
Richard Hull
Moderator
Posts: 14991
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: Simplified High Voltage Power Supply

Post by Richard Hull »

A voltage multiplier will not cut it for powering a fusor without some very large capacitors to hold the voltage up under a 10ma drain and a neon sign transformer won't work either for a true fusing fusor. All of these items are only suited for a DEMO fusor.

You will need an actual tranformer rated for full voltage of a minimum of 20,000 volts, preferably 30,000 volts (X-ray or potetnial type) that can supply a current of 20ma or more. In this fashion a simple high voltage diode will be sufficient for rectification and no capacitors are needed.

Remember it is not just voltage tht runs a fusor but power and that means a specific current at full voltage. multipliers just can't do this without making them lethal with very large, high voltage energy storage caps at 60hz.

There are many solutions which can use a multiplier but they involve a bit of expense, cutsom wound HF transformers and a bit of electronics to drive them.

The simplest solution is a variac controlling an x-ray transformer through one suitable diode or diode string with decent metering and that is it.

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
J16
Posts: 23
Joined: Wed May 04, 2005 9:42 pm
Real name:
Contact:

Re: Simplified High Voltage Power Supply

Post by J16 »

The voltage multiplier DOES use very large capacitors; the model I'm working off of used 2500 picofarad, 30 kVDC, ceramic capacitors.

Furthermore the neon sign transformer should be up to the task. The NST running at 15KV @ 30mA puts out 450 watts. The 20kV @ 20mA power supply you suggested would only put out 400 watts. The NST I would prefer to use is 15kV @ 60mA, which puts out 900 watts. The second power supply you suggested was 30kV @ 20mA, which only puts out 600 watts. In both cases the NST is more powerful, at least by comparing predicted wattage.

Additionally I went back any checked all my numbers again. Based on the information obtained from by references (included in the document) the power supply will start to become more unpredictable with more stages, but while only using one stage, the calculations should be well within a 5% margin of error.

As shown in the document, the power supply should be capable of 21.2kV @ 21.2mA. Those numbers would be within the acceptable range for fusion work.

Please at least read through the documentation, and look at the CW multiplier reference. Most of the information included in the first paragraph of this post was included in the document. I have been reading through posts for about a month and a half, I know what it takes to create fusion, and I designed these various systems to be able to meet the requirements.
apollo
Posts: 100
Joined: Wed Jan 28, 2004 1:59 pm
Real name:
Contact:

Re: Simplified High Voltage Power Supply

Post by apollo »

With a two stage power supply, it should supply supply about 40KV at 20mA (assuming that that transformer can be pushed to 1200W) if my VDrop calculations are correct (you did have a two stage design in your doc, didn't you?).
J16
Posts: 23
Joined: Wed May 04, 2005 9:42 pm
Real name:
Contact:

Re: Simplified High Voltage Power Supply

Post by J16 »

Actually a two stage in this configuration would only provide about 42.4 kV @ 10.6mA. Currently the design is only using one stage, but based on the availability of parts, that could change.

For example if given the option of using a 12kV @ 60mA NST, a second stage would be required to use it.
DaveC
Posts: 2346
Joined: Sat Jun 30, 2001 1:13 am
Real name:

Re: Simplified High Voltage Power Supply

Post by DaveC »

Joe -

Richard is correct. The NST will be a bit of a disappointment, for the following reasons. (These are all dealt with in various message threads over the past couple years, too )

The NST ratings are misleading... They are 1OPEN CIRCUIT, voltage and SHORT CIRCUIT current, which is essentially ZERO output voltage.

Not everyone is aware of how the gas discharge neon signs actually work. They are very much like a fluorescent tube. It takes a fairly high voltage to initiate the arc (discharge) and then the voltage drops to around 10v/inch while the tube draws rated current from the transformer. The unit is not really rated for a volts x amps, volt- ampere rating, the way a standard power transformer is, the true volt amp rating of a fully loaded NST is probably around 100 VA., rather than 450 .

This is a rude awakening for many of our experimenters. Now it doesn't really shut you down for experimenting and building the demo fusor.... But it will put a big crimp on what you can expect if and when you get to trying real, Deuterium fusion. Our unfocussed spherical grid structures, just eat up too much power, in low energy gas discharges, to give much joy with a low current high voltage supply.

Also, BIG capacitors, for this type of operation are not measured in pico farads, such as 2500.. which is a modest one, but in micro farads. To get an accurate idea of the electrical energy storage needs, look at an old microwave oven supply... It uses a type of voltage clamping circuit ti effectively double the peak of an AC wave and shift the center so that it is all positive. It has about a 2 amp rectifier, and an HV secondary that gives about 3kV pk. It uses 0.25 to 0.5 micro farad (500,000 pf!) capacitor for voltage conditioning. Those are mean energy storage devices, usually having a bleed down resistor attached across their terminals, to give careless technician half a chance at survival. The wise experimenter treats them very carefully, indeed. If can find an appliance repair shop, you may find a few useable caps in otherwise blown out microwave ovens.

Hope this helps a little.

Dave Cooper
J16
Posts: 23
Joined: Wed May 04, 2005 9:42 pm
Real name:
Contact:

Re: Simplified High Voltage Power Supply

Post by J16 »

Oh well, back to the drawing board. Does anyone know why neon transformers act so oddly? I already know how it drives a regular neon sign.


Would plate transformers work well with a CW multiplier?
User avatar
Richard Hull
Moderator
Posts: 14991
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: Simplified High Voltage Power Supply

Post by Richard Hull »

I assume you read the complete exposition, with experiments, on neon sign transformers in the files section.

viewtopic.php?f=19&t=7945#p57349

This really tells the story on neon sign transformers.

The 450 va rating of a neon transformer is real provided you short the output leads together. This will give you zero volts out and the primary will demand 450 voltamps from the line as the primary drives the shorted condition. The 450 va is not what the transformer can supply at rated voltage! Instead, it is a worst case, catastrophic failure, current draw on the mains should a dead short in the neon system occur!!!

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
J16
Posts: 23
Joined: Wed May 04, 2005 9:42 pm
Real name:
Contact:

Re: Simplified High Voltage Power Supply

Post by J16 »

The data in the document you referenced showed that the 5kV @ 30mA NST was able to perform up to it's nameplate voltage, if there was no load. So how much of a load does the reactor put on the power supply?
User avatar
Richard Hull
Moderator
Posts: 14991
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: Simplified High Voltage Power Supply

Post by Richard Hull »

That is absolutely correct.....With no load......

Does anyone here think that to force deuterons to fuse one needs no energy to do that! The deuteron plasma IS A LOAD!

The normal load on a running fusor would be about 10ma at full fusing voltage, whatever that might be. Usually, (20kv-60kv). Fusors demand a minimum of about 200 watts of power to fuse with some of the more successful fusors needing 500 or more watts.
This limit is more a function of grid survival than some plasma limit as about 1/3 of all the energy winds up in grid heating.

Remember, also, that in any voltage multiplier you LOSE current to get the volts. You multiply potential, not power.

Thus, if a super large 5kv @ 60ma Neon xfrmer can honestly produce 5ma near nameplate voltage, (and it can), then that is 25 solid watts at the full 5kv. If we now quintuple the voltage in a multiplier with appropriate filtering to get 25kv, then the most we might expect for our 25 watts input power to the multiplier is 1ma which would not even come close to running a fusor. If we now increased the current drain to 15ma at the neon, (more plasma load), and it fell to 4kv giving us 60 watts to play with, the voltage at the output of the quintupler would now be only 20kv, but we would have 3ma available! If we dragged 30ma out of our transformer and had the voltage sag to 3.0kv, (90 watts), the quintupler would now be down to 15 kv with a current of 6ma available, etc.,etc, and so on.

As Don Lancaster says,

The game is rigged,
you can't win,
but it is the only game in town.

Watts 'is' watts. A voltage multiplier does just that and nothing else. There is no such thing as a power multiplier with only one input.

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
Roberto Ferrari
Posts: 364
Joined: Tue Mar 02, 2004 12:21 pm
Real name: Roberto Ferrari
Location: Argentina
Contact:

Re: Simplified High Voltage Power Supply

Post by Roberto Ferrari »

Richard,
As the name of the game is low efficiency, cannot we handle the grid as the anode in an X-ray tube, cooling it?
I know it will require big changes, namely to mount the tungsten grid into a copper rod, internally water-cooled, and to choose which half of the fusor tie to ground. Any one of the alternatives would be cumbersome but would help to push more power into it.
Roberto
User avatar
Richard Hull
Moderator
Posts: 14991
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: Simplified High Voltage Power Supply

Post by Richard Hull »

All of this has been discussed before. Yes, we can most certainly cool the grid. As of yet no one has attempted it.

Again, lots of chatter, no action and the ball is left bouncing in the court by all the folks who had the great ideas.

Ideas remain a dime a dozen. Apparent genius is as cheap as rain water. Action remians as rare as the most precious of gems.

No big complaints at my end, just simple observations.

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
Richard Hester
Posts: 1519
Joined: Thu Jul 05, 2001 12:07 am
Real name:

Re: Simplified High Voltage Power Supply

Post by Richard Hester »

The problem is that cooling the grid may prevent the grid from melting and retard evaporation, but it won't do much to prevent sputtering, where metal particles are ejected from the grid via brute force ion bombardment. Hydrogen/deuterium is generally not a very efficient sputtering gas due to the low atomic weight, but the high average energy in a fusor environment sort of makes up for this.
J16
Posts: 23
Joined: Wed May 04, 2005 9:42 pm
Real name:
Contact:

Re: Simplified High Voltage Power Supply

Post by J16 »

The following takes into account the loss of voltage and amperage at the rated load, based on the paper by Richard Hull.

viewtopic.php?f=19&t=7945#p57349

It also takes into account the loss of amperage due to the CW multiplier.

Neon Sign Transformer 15kV @ 60mA
First the center tap on the secondary is removed to allow for full voltage output. Because of the 2/3 loss this transformer only puts out 10kV @ 40mA at the rated load. Thus using a 2-stage CW multiplier the output is changed to 28.3kV @ 14mA. This results in a 396-watt power supply.

As a final note, I do not plan to use a NST for my project unless I have no other alternative, which is looking likely. I just want to show that it is possible to use a NST to produce measurable fusion.
User avatar
Richard Hull
Moderator
Posts: 14991
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: Simplified High Voltage Power Supply

Post by Richard Hull »

Have you ever gutted a neon sign transformer? I have de-tarred about 30 of them! You can't effectively lift or rid yourself of the center tap!

All of the higher voltage transformers starting at 7.5kv output rating contain two (2) secondary windings in the form of a doughnut coil shape. The winding ends nearly touching the iron core, (close against the core-inside windings). Naturally, these ends nearly touching the iron MUST be tied to it. The two outer most winding leads, at the outside, are taken to the two high tension knobs.

The iron core would have to be disassembled and the coils removed. The core would probably have to be undercut slightly by a milling process in order to allow for insulation to be inserted under the coils. Remember, this is the half point of the transformer secondary and the inside of both coils will bristle with 7500 volts right next to the iron core The two inner leads of the, now insulated, doughnuts would be joined with HT wire across the primary winding that sits between them. While doing all of this you could remove the magnetic shunts and make the transformer run to full voltage and full current! You would then retank the whole thing in mineral oil.

A number of tesla coilers with good hands and "the right stuff" have done this. It is just a matter of a big tarry mess, a good deal of metal milling, reassembly and oil tanking.

Most folks just buy the transformer they really need and be done with the entire biz. You can burn off a lot of life trying to re-engineer neons.

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
J16
Posts: 23
Joined: Wed May 04, 2005 9:42 pm
Real name:
Contact:

Re: Simplified High Voltage Power Supply

Post by J16 »

BY removing the center tap I was reffering to the procedure outlined here

http://rtftechnologies.org/emtech/nst.htm

I also meant that I was going to remove the center-tap ground, I hope that I didn't create any confusion.
User avatar
Richard Hull
Moderator
Posts: 14991
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: Simplified High Voltage Power Supply

Post by Richard Hull »

This is the same thing, but the transformer core is now lifted to 7,500 volts and hopefully floats without danger in the tar. Note some cores are welded to the frame or separately brought to the frame in the tar. If this is the case you might create a health hazard. Also, the primary is at ground as the neutral input, effectively, is at one end and, thus, the 7500 volt core may arc to the primary quite easily through the thin tar at the primary to iron interface. It is all luck of the draw here as not all transformers are built like the allison or as easy to bypass.

Remember, the core remains hot forever as the secondary center points are NOT grounded to the case but to the iron core!!! The core is then separatly grounded to the case. Anyone who has fully depotted a neon xfrmer will know this subtle danger point.

You may get away with murder here, you never know.

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
Post Reply

Return to “High Voltage - Fusor Input Power (& FAQs)”