Author Options:

Connect transformers together in series? Answered

I've got 4 neon sign transformers, each outputting 12 000V at 30mA. I want to connect them together in series to increase the voltage to 48kV but I just wanna check on the right way of connecting them for safety reasons. The input for all transformers is the standard 220V AC.( I need such a high voltage for a fusion reactor) (the pic is one I found on google that's why the input voltage is 120)


The forums are retiring in 2021 and are now closed for new topics and comments.

5 years ago

Not enough because you lack some very basic electronics knowledge judging from your questions.

One problem with answering questions in isolation is an assumption of underlying knowledge. A lack of that understanding when dealing with high voltage, large currents or unstable systems leads to a good possibility of personal disaster.

Jack A Lopez
Jack A Lopez

Best Answer 5 years ago

I think you forgot to up the picture, or link to it.

However, I can almost picture it. I have seen neon sign transformers (NSTs) before, both IRL, and via Google(r) Image search.

Moreover, I can almost imagine four of them, all energized, with their secondary windings all wired together in series.

I think I would want to be on the other side of the room, or maybe in a different room, in a different building on the other town, before closing the switch that turned it on.

The first obstacle I can think of, to this wonderful dream of connecting 4 NSTs together, is that the secondary windings of your NSTs might be connected to the metal case of the transformer.

You should use an ohmmeter, or multimeter in ohmmeter mode, for to look into the windings of your NSTs for to try to discover what is connected to what.

As recently as like nine months ago, someone was asking this forum about how NSTs are wired, and how to examine one with a ohmmeter. My reply to that question, including some pictures of a NST, and a hand drawn diagram of what I imagine it looks like inside, are posted here, in this answer:


So yeah. That's the first obstacle; i.e making sure the transformers are not shorting to each other by way of the metal cases.

The next obstacle I can think of is, there may be places inside one or more of the NSTs where the insulation is subjected to much larger electric fields (by a factor of 4) than those experienced the original design.

By the way, have you heard of, read of, voltage multiplier circuits?



That might be another, better, way to get your high voltage, unless it just has to be 48 KV AC, rather than DC.

I did mention that you don't want to be anywhere near things like this while they're turned on, right? Yeah. Don't touch it. If you really must poke or touch some part of it while its energized, the best way to do this is by way of of a few meters of insulating plastic pipe, also known as a "chicken stick".


Answer 5 years ago

I certainly agree with you about being in another building when the switch is turned on. Thanks for telling me about voltage multiplier circuits, but do you think I'd be able to get around 30kV DC from 12kV AC at 30mA?

Jack A Lopez
Jack A Lopez

Answer 5 years ago

Voltage multipliers tend to work well when the output current from the last stage is small. One would probably work perfectly if the output current is zero.

But, you know, how much current do you need?

It's an important question. Maybe it is the question you should ask first considering this high voltage power supply, for your particle accelerator or fusor, or whatever it is. i.e: How much current must be supplied at 30 KV DC?

This is equivalent to asking how much power must be supplied, since power is voltage multiplied by current.

The reason it is an important question is because power supply design tends to become more challenging, more difficult, more expensive, for things that require more actual power, which is, voltage multiplied by current.

Well, that statement is pretty vague so far, kind of like saying the cost of a power supply scales with the amount of power, but assuming it is true, then taken to its logical extreme: The easiest power supply to build is a power supply that supplies zero power.

As a too easy example, consider a power supply which must supply 10000 volts DC, at exactly 0 amperes, or 0 microamperes, or 0 picoamperes, etc. The power it supplies is exactly:
(10000 V)*(0 A) = 0 W = zero watts

You could build this supply from just a single capacitor, that happens to be already charged, with a potential of 10000 volts across it.

You might notice some practical problems with this idea, like: How do you charge the capacitor in the first place? What do you if this capacitor is actually leaking charge slowly, like by ionizing air that touches its terminals? Is there even a way to measure the voltage on this capacitor without drawing current from it? Does that require a special zero-current volmeter?

Finally, is a device that does no work, actually useful for anything.

But of course these questions arise because the example is too easy, too simple to actually be practical, so now I have to try to give some examples of the next best thing: high voltage sources which supply almost zero current.

The first example is one of those air-ionizer gizmos. The actual module that makes the high voltage is typically a brick shaped thing, about the same size as a pack of cigarettes. The voltage at the output terminals is around 10 KV, and the current is, I dunno, microamperes? Some pictures of these,

Wikipedia article for "Air ionizer"

The second example I'm going to offer of an almost zero-current, high voltage, power supply, is that of the old Van de Graaff generator,

I mean this device is essentially a single high voltage capacitor (like the too easy example), but with moving rubber belt to constantly re-charge it.

How fast does the belt move charge? I mean what would be the approximate number, measured in microcoulombs per second, equivalently electric current in microamperes? Or maybe it is more like nanoamperes. I dunno. But if I had a number for this, then I could actually make a meaningful comparison to other high voltage sources, like the air ionizer module.

Regarding voltage, even a small VDG, like table top sized, as seen in a typical physics classroom, can produce potentials of like 100 KV.

Also the bigger versions have been used in the past (and maybe in the present too?) to drive actual particle accelerators.

Which sort of brings me back to where I started, with the assumption that you're driving something like a particle accelerator, some thing that is mostly vacuum on the inside so the only current is, I dunno, like small numbers of fast moving ions. Uh.... I guess you're the one who has to build the thing, whatever the details are.

By the way, I think the member here who has written the most 'ibles on high voltage projects is Plasmana. I think. I could be wrong about that, but it seems like his name comes up a lot.


Also BTW, there are forums out there, totally dedicated to high voltage, like,



Answer 5 years ago

well yes I am driving a 'particle accelerator' and the current would be very low. Fusion occurs at any current level but I'd like it around 20mA cause then you get a nice ball of plasma inside the inner grid which is inside the vacuumed chamber, but if I can't get it to that amt current it would be fine.

Jack A Lopez
Jack A Lopez

Answer 5 years ago

I am guessing a single neon sign transformer driving a some kind of rectifier network (i.e full wave, doubler, tripler, etc.) will get you approximately, into the ball park, of where you want to be for voltage and current.

Looking at the voltage supply for the Make Zine fusor, shown here,


in Steps 15 - 20, you can sort of get a picture of what they're using for a HV supply, and it appears to be just a full wave rectifier circuit using 2 diodes, the kind used with a transformer with a center tap. It is the same as the diagram in this picture,


from the Wiki article on "Rectifier"


The expected peak DC voltage for this circuit is 0.5 times square root of 2 (=2^0.5 = 1.414) times the RMS voltage of the transfomer. For example, for a 15 000 volt RMS NST, this gives 0.5*1.414*15000 = 10605 V

So, they're only using like 10 KV, rather than 40 KV, but they claim this fusor works. But, I dunno. Their criterion for "it works" might just be "makes glowing purple light"

Regarding proof that it works, Christensent's 'ible,


is notable since he apparently has some tools to detect neutrons, and for me, that's pretty convincing evidence some nuclear process is happening, compared to observations of heat and purple glow.

Also this article,


from fusor.net, suggests essentially the same circuit as used in the Make Zine fusor, also made from a NST, and rectified to about 10 KV DC.

Christensent's ible says he's using about 40 KV, but his description of the circuit he built to supply that voltage is pretty sketchy, or non-existent. The picture in Step 5 kind of looks, to me, like maybe he has a multiplier network of capacitors and diodes. It's in a plastic bowl on the left side of the picture. Also guessing he's got it drowned in mineral oil to keep it from arcing to itself.

Which reminds me, if you want to see a more clear picture of a homemade voltage multiplier network for producing several 10s or kilovolts, this page,


is an oldie (2001?) but a goodie. Some guy named, Fred, building a linear accelerator in his garage. The page says this supply produced 100 KV, at 5 mA!

Anyway, I can't tell you for certain what will work, or what won't, but it looks like there are plenty of people out there working on similar things


5 years ago

this young man is apparently 15 so i guess lacks some basic principles and knowledge he might be better off gaining before getting too ambitious.

NOTE I have nothing against youth.


5 years ago

The limit for "normal" HV transformers is around 18KV, finding something with a higher rating is next to impossible unless you want to spend serious money.
Reason for this is quite simple:
The higher the output voltage the more needs to get into the primary side of the transformer.
With your models you are looking at around 400-500W (with losses) on a single one already.
Next problem is the magnetic field, which needs enough windings - but too many windings required and the output current breaks down.

Although you can connect some types in series to increase the output voltage, your main problem reamains to be the power requirement and insulation.
4 in series and you need at least 1.5kW to power them....
I have done similar experiments with up to 6 Mots and even in those "low voltage" regions isolating and insulating becomes and issue.
And so far I have not seen a working fusion reactor in a residential area either ;)


5 years ago

If you are thinking of connection the output of one transformer to the input of the next one to boost the voltage it will not work. The insulation will break down in the primary of the second transformer and short out. High voltage transformers need to be built with enough turn to turn and layer to layer insulation to withstand the voltage.