Page 1 of 1

Voltage multiplier help

Posted: Sat Jan 20, 2018 5:18 am
by Leonardo Chen
Hello, I have recently been considering multiple ways to get 35-40kv at 10 milliamps. However, I recently came across a solution that is cheap and definitely do able. I am thinking of hooking my 12 kv neon sign transformer to a full wave rectifier Cockcroft-Walton multiplier which would then step it up to ~36kv, perfect for fusion. Now I am searching for diodes and capacitors to build this multiplier, because I could not find any pre-built multipliers. I am wondering if the Fusor community could help me find supplies or a multiplier to use. Besides that, I have located nearly all my parts and am nearly ready to start ordering them. I look forward to showing the community my Fusor.
Thank you.

-Leonardo

Re: Voltage multiplier help

Posted: Sat Jan 20, 2018 5:40 am
by Bob Reite
I think you meant 10 mA.

Google for "Full Wave CW Multiplier" for a schematic.

Google for "Full wave cockcroft walton voltage multipler calculator" for a handy site that will help you choose capacitor sizes based on frequency and how good or bad you want regulation and ripple. Ripple is not a big concern, regulation might be.

I finally bought some stuff from hvstuff.com The didoes I bought do work as advertised, I would assume that the capacitors also live up to their ratings. Only downside is the stuff is shipped from China, so you have to wait 3-4 weeks to get it.

Re: Voltage multiplier help

Posted: Sat Jan 20, 2018 11:57 am
by Leonardo Chen
Thanks bob, I will do that. The shipping doesn’t matter to me.

Re: Voltage multiplier help

Posted: Sat Jan 20, 2018 12:33 pm
by Dennis P Brown
A neon sign transformer (NST) will not work for a fusor grade (read detectable neutrons) voltage multiplier - NST units are designed for 60 Hz and to supply only an initial surge current and rather low sustained/continuous current. Read Richard's excellent studies on this very subject! Also, do read both the FAQ's and past posts on voltage multipliers (some have built usable units.)

This is why there is no FAQ on how to build a simple, cheap fusor grade power supply using a NST. People have been approaching this problem for a long time and NST's are not the x-former that one uses.

I have gone this route for another project (deuterium accelerator) using a NST and I exploited very large caps (HV door knobs) and high current diodes and it did not work. I strongly suspect if you use any NST (even coupled units) you will not get enough current since a 60 Hz x-former will not provide enough current fast enough to create the power/current you will need. My system was built for a electro-static supply and still couldn't produce enough current for that simple application (at least not the 100 kV I needed - lol.)

What you need is a very fast switching supply (20 - 40 kHz) to feed the multiplier and that requires a more complex build (full wave is a good idea here but does require more caps & diodes.) Search the posts on this subject before buying parts so you can build the correct supply.

Bob Reite, who just posted here answering your question, has built a successful fusor grade multiplier. Maybe he will share that design if you request it.

In any case, any high frequency/high voltage/power supply is not a simple build unless one has a fairly good knowledge of electronics or is willing to learn it (read - an extensive side project that could take a good bit of time. Not an easy approach for a novice.) Remember, the power output for a fusor is substantial and hence, the components for the switching supply must support this power level for a fairly long time.

This all said, building a NST based supply is a fairly low cost method to learn about both the limits of high voltage systems, voltage multipliers and the how of simple basic electronics related to HV. As long as you understand that a NST based supply will not make detectable fusion, this could be a good demo fusor supply - depending on the caps used, all the lethal voltage of a fusor with no detectable neutrons (lol.)

Re: Voltage multiplier help

Posted: Sat Jan 20, 2018 8:14 pm
by Leonardo Chen
Thanks Dennis, I will read up on voltage systems and consider other options.

Re: Voltage multiplier help

Posted: Thu Jan 25, 2018 8:04 am
by Bob Reite
I think once you run the numbers for a 60Hz supply, you will NOT want to spend the money on the large HV capacitors! Plus they will store lots of really destructive energy. As stated, you will need some source of 20-30 KHz energy to feed a ferrite core transformer. 30 KHz allows you to use reasonably sized capacitors. While the change on the capacitors will still be lethal, at least there won't be as much energy to damage equipment in case of an arc over, which you will get from time to time. I get my 30 KHz from a function generator that feeds a pair of 400 watt audio amplifiers of my own design that "I had on hand".

Re: Voltage multiplier help

Posted: Thu Jan 25, 2018 11:42 am
by Finn Hammer
Bob,
Am I correct in assuming, that you do not have a feedback network to stabilize the output of your supply?
If yes, this is great news, since the feedback network design is a challenging task.

Cheers, Finn Hammer

Re: Voltage multiplier help

Posted: Thu Jan 25, 2018 1:27 pm
by Bob Reite
No feedback network. To get the two parallel amplifiers to share the load, there are 2 ohm 100 watt resistors in series. with the 100:1 turns ratio of the transformer, that makes the output impedance of the supply somewhat high eliminating the need for a ballast resistor on the output. The amplifiers were designed to take abuse, as I did not want any warranty returns in spite of what customers might do (such as shorted speaker lines). Worst that ever happened using them in this service was blown rail fuses. Put in a new fuse and they are as good as new.