How would you go about stepping up dc power supply voltage?

For Short Term Learning Discussions ONLY. This area is for CURSORY questions and connecting with other users ONLY. ALL technical contributions need to be made in the appropriate forums and NOT HERE. All posts are temporary and will be deleted within weeks or months. You should have already search the extensive FAQs in each of the forums before posting here as your question may already be answered.
Ryan Oddie
Posts: 11
Joined: Wed Feb 15, 2017 6:10 pm
Real name: Ryan Oddie
Location: London, United Kingdom

How would you go about stepping up dc power supply voltage?

Post by Ryan Oddie » Sun Apr 16, 2017 8:37 pm

Hi all,

As appears to be the case for many the budding fusoioneer I am currently struggling to find / build a HV power supply for my fusor. Recently I noticed a couple of Hitec 1kW power supplies on eBay for a reasonable price, the only issue is the output is configurable only between 1kV and 10kV. Could someone please explain if / how I could go about stepping up the voltage (so that the output is configurable between 10kV and 100kV)?

Any feedback / guidance is appreciated.

Thanks, Ryan

ian_krase
Posts: 415
Joined: Mon Nov 28, 2016 7:48 am
Real name: Ian Krase

Re: How would you go about stepping up dc power supply volta

Post by ian_krase » Mon Apr 17, 2017 4:46 am

I think that the most common strategy is to start with high voltage AC from a transformer and/or a solid state switching driver.

AC can then be multiplied with a voltage multiplier. Conveniently, the individual elements of the voltage multiplier often only need to withstand the input voltage, not the output voltage. Finally it is rectified (or the voltage multiplier rectifies it for you).

Whether there's a way for HV DC to be easily boosted higher, without incurring a technical challenge larger than making the HV DC in the first place, is a question for somebody with more skills than me.

User avatar
Dennis P Brown
Posts: 1424
Joined: Sun May 20, 2012 2:46 pm
Real name: Dennis P Brown
Location: Glen Arm, MD

Re: How would you go about stepping up dc power supply volta

Post by Dennis P Brown » Mon Apr 17, 2017 3:46 pm

Reading the FAQ's on fusor power supplies can clear up just such issues as you are asking. But I will quickly summarize - a fusor requires not just voltage but current; exactly how much total power and ultimate voltage that will be needed depends on your ability to detect neutrons - these issues go hand in hand and cannot be easily separated.

Aside: one cannot "Up" the voltage of any power supply to much above its normal operating voltage (if you think otherwise, please read up on electronic power supplies - paying especial attention to transformers and how they work) and it is not useful to try. Thinking to do such is not for anyone who isn't an expert with electronics and very familiar with both transformers and high voltage systems. Current is another matter but that is for you to figure out using the FAQ's.

Ryan Oddie
Posts: 11
Joined: Wed Feb 15, 2017 6:10 pm
Real name: Ryan Oddie
Location: London, United Kingdom

Re: How would you go about stepping up dc power supply volta

Post by Ryan Oddie » Mon Apr 17, 2017 4:51 pm

Thanks Ian, I have seen a number of mentions of using a voltage multiplier / rectifier combination on these forums and I agree that seems to be the norm. I was just hoping I could short cut some of the complexity by using a factory built power supply and modifying the already DC output. As you suggest this may not be plausible but I thought it was worth asking just in case (as its a really good deal).

Hi Dennis, I think maybe I have not explained my situation in enough detail or I am simply misunderstanding your comment. The power supply is 1kW output and configurable from 1kV to 10kV (so the current is between 1Amp and 0.1Amp respectively). From some of the other posts on this forum I understand that around 600Watts is necessary for detectable fusion (in a well designed fusor). It is my understanding however that as a fusor design uses the potential difference between the negatively charged inner grid and the outer grid (held at 0V) to accelerated ions, the larger the voltage used the greater the potential well and therefore the more kinetic energy the ions will have at the centre of the grid. As the potential for fusion (fusion cross section) is proportional to the energy of the colliding ions my expectation is that increasing the voltage to 50kV/100kV will increase the likelihood of fusion. I am therefore looking to increase the voltage output of the power supply (I am looking to buy). My issue however is that although it generates the required power for fusion the supply is limited to 10kV. I would therefore like to know if / how I can step up the dc output (by feeding it into some dc equivalent of a voltage multiplier circuit) rather than building a high voltage power supply from scratch using a transformer / voltage multiplier / rectifier combination).

Hope this clarifies my question but please correct me if I have misunderstood the science or your point.

Kind regards,

Ryan
Last edited by Ryan Oddie on Mon Apr 17, 2017 5:00 pm, edited 1 time in total.

Ryan Oddie
Posts: 11
Joined: Wed Feb 15, 2017 6:10 pm
Real name: Ryan Oddie
Location: London, United Kingdom

Re: How would you go about stepping up dc power supply volta

Post by Ryan Oddie » Mon Apr 17, 2017 4:57 pm

Hi Dennis,

Sorry I think you added the aside whilst I was writing my previous comment. If upping dc to dc voltage is not possible then I guess its back to the drawing board.

Thanks for your advice.

User avatar
Garrett Young
Posts: 123
Joined: Sat May 27, 2006 1:51 am
Real name: Garrett Young

Re: How would you go about stepping up dc power supply volta

Post by Garrett Young » Mon Apr 17, 2017 5:10 pm

What you are looking for is a switch mode DC-DC converter, but designing such a device to operate at 10kV input is nearly impossible. Boost topologies require a transistor switch and this would need to withstand 10kV plus "ringing". Your effort would be better spent on redesigning the multiplier section in the 1kV 10kV power supply to achieve a the high potential you desire, but this would also require considerable technical expertise. Your options are really two-fold - make your own supply using an X-Ray transformer or find an off-the-shelf supply with the exact specifications you want.
- Garrett

Rex Allers
Posts: 212
Joined: Sun Dec 30, 2012 8:39 am
Real name:
Location: San Jose CA

Re: How would you go about stepping up dc power supply volta

Post by Rex Allers » Tue Apr 18, 2017 2:40 am

Ryan, I don't think you are completely off base with the idea of upping the output voltage from a supply. I also think your observation of 600W or more of output is a reasonable size for a fusor supply.

Most of the modern HV supplies I am aware of are switchers running at 10s of kHz (maybe 30 - 40 kHz). The heart of the supply drives a transformer (or two) that produces two opposite phases of low KV (maybe 5 kV) relative to a center ground. These two phases feed a full-wave multiplier that ups the voltage to the input kV times n, where n is the number of stages in the multiplier. Of course as the output voltage goes up, the available output current goes down proportionally. But you seem to understand this, the output power should only go down a little if the multiplier is designed well.

The two main vendors of these types of supplies, I would say, are Spellman and Glassman. All the Spellman supplies that I am aware of have their multiplier sections potted in some kind of "rubber" so they aren't very hackable. The glassman supplies that I am aware of have their multipliers air insulated so they can be worked on. There's a detailed post somewhere on these forums where one member changed the output of one type of Glassman supply from positive to negative by reversing all the diodes in the multiplier section. A bit of jumper changing in the monitor portion was also required.

I have looked at a couple other Glassman supplies and found a couple where the multiplier circuit board had one or a few sections that were not populated so the output could theoretically be increased by adding diodes and capacitors to populate more stages. There would also need to be additions to the sensing HV divider circuit, current monitor and metering circuits.

For some arbitrary switcher supply with a too-low output voltage, it might be possible to remove the multiplier section, bring out the two phases of lower drive kV and build an external multiplier box to suit. Some of the higher power supplies are designed this way (2 separate boxes) from the factory.

I looked on ebay for a Hitec HV supply as you described and found nothing, so I have no idea what their design might look like. Any of the kind of modifications I have described here wouldn't be easy and shouldn't be attempted by anyone who lacks good engineering skills or HV safety concepts.

So -- maybe. But not easily if at all.
Rex Allers

User avatar
Dennis P Brown
Posts: 1424
Joined: Sun May 20, 2012 2:46 pm
Real name: Dennis P Brown
Location: Glen Arm, MD

Re: How would you go about stepping up dc power supply volta

Post by Dennis P Brown » Tue Apr 18, 2017 10:56 am

It was not clear in your post that the ebay power supplies were adjustable in voltage - that does change things. That said, a 10 kV DC output supply can't be further modified to increase its voltage in the range needed by a fusor in any simple manner as a poster pointed out. Your later post also indicates the "up"ed PS still only reaches 10 kV which is far too low for real fusion. Increasing a DC output isn't something I'd ever suggest trying (an AC system would be another matter and opens up other issues, however.)

As you now realize, the issue isn't just "up"ing the voltage. As Rex points out redesigning a spellman/glassman system to operate in the 20 - 30 kV range is possible; I consider that to be a major task that I am not aware of anyone here doing (haven't searched so could be wrong.) That said, building one's own voltage multiplier to operate in the 20 - 50 kV is both straight forward and not out of reach of even a novice; however, as the previous poster explains, the issue is frequency - only a rather high frequency driver can operate a voltage multiplier fast enough to provide the required power that a fusor requires for detectable neutrons (std 60 Hz will not work for that level of power.) Getting diodes and caps for a multiplier that can handle those higher frequencies could be pricey; also, issues of over-voltage spikes must be handled or the entire diode stack can easily be blown (with rather costly results.)

Your question about upping a DC power supply with a voltage multiplier indicates you do not have much experience with these systems - the short answer is, no. The long answer is that only AC can do this since one needs the oscillating voltage to charge a cap/diode system to get the voltage to double. Look at the theory and operation of such multipliers - they are very simple. High frequency power supplies are significantly more complex - I built a few and find that voltage multipliers are easy in comparison.

If one want's to go the route of upping a Spellman/glassman and modifying it is possible (and with your experience, I would strongly suggest you do not.) The key is getting a unit that has a significantly high current ratting and upping its voltage as Rex & Garrett both point out. Just remember, when you open the supply in that manner you are on your own - safety is a very important issue and when the system is at a higher voltage it is now well outside its safe operating parameters and you alone must ensure that your mod's remain safe. Do remember that current output falls very rapidly with extra stacks so the original supply will have high current since a fusor in the 30 - 20 kV really needs 10 - 20 ma to be detectable by run-of-the-mill neutron detector systems.

This brings up my last issue in my previous post - neutron detection. The better (more sensitive) your detector, the less current and lower voltage one can use in a fusor and still detect neutrons. Richard's FAQ covers this in great detail and is an essential read. And I will add that in my experience, building high voltage and current power supplies for a fusor was easy compared to the efforts I put into neutron detection. That may very well just be my limitation but do not over look that part of a fusor - no power supply will do you any good if you do not have (1) deuterium, and (2) a neutron detector that is sensitive enough for your fusor's flux level. Don't over-look these two rather critical issues.

Keep an eye out for working (!) power supplies that can handle at least 700 watts in the voltage range you decide upon. Higher wattage's are, of course, better and will make neutron detection easier. This brings up the issue of voltage so read Richard's FAQ on power/voltage and fusion to get a feel for this topic. Unless you are good at electronics, do not buy unproven supplies unless the seller will refund your money if the unit does not work. Which brings up a point - having the equipment to measure the required high voltage safely is also a critical ability one has to have before buying any HV supply!

As for x-ray power supplies you need to realize that these are not turn key systems and require both a good bit of electronics knowledge and experience with HV (often high freq.) systems to be successful with. Read both the FAQ's on this subject and various people's threads on their attempts with systems they bought.

User avatar
Richard Hull
Site Admin
Posts: 10692
Joined: Fri Jun 15, 2001 1:44 pm
Real name: Richard Hull

Re: How would you go about stepping up dc power supply volta

Post by Richard Hull » Tue Apr 18, 2017 4:23 pm

Hi-Tek typically makes monster supplies. (Very high voltage, very high current) They typically use heavy, potted, multiplier discs and use a 20lb flyback style 25khz xfmr. They also contain two parts. The driver is usually three phase AC input and is a large rack mount item. The output multiplier stack is a separate item and can be configured to most any voltage based on the number of large stacked multiplier discs. None of the high voltage stacks can have their polarity changed, ever, unless completely new multiplier discs of the selected polarity are purchased. Needless to say, only a negative HV output supply will function with a fusor. Surplus negative output supplies are just not found, typically, and the ratio of positive:negative supplies are inthe 40:1 range!

In the end, unless you are an electronic genius and high power, high voltage circuit designer, there is absolutely no way to effectively and economically take 10kv DC at 1 KW and up it to 30-40 kv. Normal Hi-Tek supplies take 208 three phase AC power. This is converted it to 400VDC, then, using a IGBT electonic switcher/Driver, this is converted to a pulsed DC (High frequency 25khz). This is fed to a giant flyback transformer to make 25khz AC at very high current and moderately high voltage. This is fed to a multiplier stack to produce HV DC. If you do not fully understand every thing I just said, leave behind ye, all hope of modifying such a supply. (AC to DC, DC to Pulsed DC, Pulsed DC to high frequency AC at moderately high voltage AC to a multiplier to produce very high voltage DC of single specific polarity)

I currently own the proper Hi-Tek, giant flyback and 4 positive discs that were in a +60kv @ 60ma stack. I recently purchased three Hi-Teck negative stack discs for the super low price of $150 each. At some point, I hope to make an IGBT 240v single phase circuit to drive the flyback and pump the new negative disks. Simple inertia and 71 years of age holds me back as I already have a fine linear X-ray supply which I reconfigured to negative output. This ancient supply has faithfully driven fusor IV since 2004.

If it ain't broke, don't struggle to fix it....or change it out!

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
Retired now...Doing only what I want and not what I should...every day is a saturday.

Ryan Oddie
Posts: 11
Joined: Wed Feb 15, 2017 6:10 pm
Real name: Ryan Oddie
Location: London, United Kingdom

Re: How would you go about stepping up dc power supply volta

Post by Ryan Oddie » Wed Apr 19, 2017 7:01 am

Hi All

Thanks for your comments. I did in the end buy the power supply (for £200) but as you have all pointed out its a fools errand to attempt to modify it given my experience. I do have some other ideas for using it for ION injection (at its rated output) so it will come in handy further down the line (just not as I had hoped).

I realise from all your feedback that it would have been so much simpler to give you the actual details of the power supply so that you can give more informed feedback. My apologies for that, I will make a note to provide such information in the future. For anyone interested the power supply I bought is a slightly older version of the Hitek OL1K/103 (labelled as the OL1000/103). The spec sheet for it can be found here: http://www.advanced-energy.com/PDFdatas ... pplies.pdf

Thanks again for all your input it has been really insightful

Post Reply