Microwave Oven Transformer

[Quentief]

Hi everyone,

I have recovered a transformer from an old microwave oven. I would like to use it to make some very high voltages and if it is possible, to melt some glass with it. Actually, I have already seen on YouTube some people used these devices to melt some glass to make fulgurites.

I would like to experiment that. However, I have been told than these transformers need to be ballasted, otherwise they will draw too much current, until to burn. Can I use a 100 mH coil in serie with this transformer to limit the current and preserve it from overheating ?

[Hydron]

I'm going to have to start with the obligatory warning about microwave oven transformers - they are one of the most dangerous readily available electrical items, and can easily kill you with a single touch (they will easily push 100s of milliamps to amps through a human body). WAY more dangerous than almost any solid state tesla coil for example. If you're going to muck about with one treat it as a box of instant death!

As for ballasting, as a crude approximation you can just work out the reactance of the ballast (for an inductor this is simply 2*pi*f*L) and then use ohms law to work out the maximum current that could flow at a given voltage. This will not take into account any resistance or inductance of the transformer (microwave oven transformers, or MOTs, will have _some_ designed-in leakage inductance for ballasting, but not enough to stop them being overloaded when run straight on mains power), but is useful to get an upper bound.

[Quentief]

Yes you absolutely right, these transformers are very deadly. I am curious to experiment my own but I am also very scared to use it.

Actually, I would like to make some kind of resistance furnace, using liquid sand as resistance. It should work : the high voltage make the current able to pass through the sand and because of its very high resistance, the sand will melt and I would like to place above it a crucible with the metal to melt.

According to my calculations, the coil should limit the current to 5.4 Amps (effective current). I think that should be suitable for the transformer. Furthermore, I did not consider the resistance of the coils and the reactance of the MOT, so the current should be lower than 5A.

[klugesmith]

Yes, MOT's come with risk of lethal electric shock.  One thing that's even more dangerous, I think, is large multi-kV capacitors.  MOT core has shunts that limit the current on HV side to order of 1 ampere.  No such limit in the capacitor, which at same initial voltage could hit you with 10 amps or 100 amps depending on resistance of your skin and meat.  Or 0.5 or 0.2 A, if that's worse for your heart.   And MOT is inherently safe when its power cord is unplugged.

Limiting MOT current with a ballast in series with the primary is generally easier than ballast in series with secondary.  The required ballast impedances are much lower.  You could use a heating appliance or incandescent-type light bulbs.

Another option is to use a similar MOT with one winding shorted.   For example, both MOT primary windings in series, and one secondary short-circuited. One reference is Adam Munich's page here: https://adammunich.com/microwave-transformers/

Re. sand as a heating element.  Don't you need to get it very hot before it will conduct appreciable current, even with voltage very high?  What will contain the sand?


Do you have a variac?  It would be educational to chart the MOT's no-load primary current as you ramp the input voltage from zero to nominal.
Could repeat that with shorted secondary, up to the maximum current of your variac.

[davekni]

Just emphasizing the safety issue!  A man died here in my little town about 3 years ago, using a microwave oven transformer to char Lichtenberg patterns on wood.

A ballast on the transformer input is not much help until the output current is reduced well below 0.25A.  That's roughly the 50/50 lethal current at 60Hz when flowing from one hand to another (or to a leg).

Large capacitors can be quite dangerous.  The ~1.5-2uF 2kV capacitor from a microwave oven is much less dangerous than the transformer, however.  Humans are less sensitive to DC than to AC.  Peak sensitivity is around 82Hz (from a study I read decades ago, so there may be updated info now).  The single pulse discharge from a 2uF 2kV cap (4 Joules) has some risk, but not that much.  It's less energy than a defibrillator, and especially less charge.  (Long ago in college, I experienced this across two points of one hand.  Painful and left a burn mark.  Not a relevant data point for safety, however, as it didn't pass through my heart.)

[Quentief]

Actually, when I was saying that I would like to ballast the transformer, of course I was thinking about the primary coil. I would like to connect in serie the primary coil with an inductance. However, you said that these transformers are already ballasted thanks to the shunts. I did not remove them, so do I still need to ballast my transformer with an external component ?

Do not worry about the HV capacitor, I discharged it and I do not want to use it. I know to ballast the transformer will not reduce the electrocution risk, it is more like to preserve the transformer against overheating risks.

And concerning the sand, well the idea is to draw an arc through the sand by touching the electrodes. The electric arc should melt the sand. Then, because liquid sand is an electrical conductor, I would like to use it as a heating resistance.

[Hydron]

MOTs are already ballasted to a degree as part of their use in a microwave oven, however it is not sufficient to prevent them from overheating when used for drawing arcs etc. They can potentially be used for short periods without additional ballast, but will pull a _lot_ of current, and definitely need additional ballast for continuous use.

[MRMILSTAR]

MOTs are built very cheaply and right on the edge of proper operation. They have insufficient iron and copper to operate as a proper transformer should. They will draw at least 400 watts with no load! So they will get hot regardless.

I built a poor-man's filament transformer with one by rewinding the secondary coil to produce 10 volts at 10 amps for my VTTC. It worked but was horribly inefficient, drawing at least 400 watts with no load and heating up rapidly. Because of this I replaced it with a proper Thordarson filament transformer.

[Quentief]

You are right, these transformers are really inefficient. I have been told than they are tend to be replaced by Inverter transformers, which have a better efficiency. Unfortunately, the transformer I recovered is a classical one, with probably huge energy lost issues.

I have read the website that you shared : https://adammunich.com/microwave-transformers/
Adam Munich is suggesting that it is possible to design an inductive ballast to limit the current at the primary, by using a long wire wrapped around some steel rods. He claims that 50 turns around 25 or 5 mm iron rods is enough for this purpose. I would like to try that, however I do not understand what are the core dimensions : what is the length and the section of the core ? He means 25 mm diameter or length ? I cannot understand.

[davekni]

If you can find two more microwave oven transformers and a steel-capable band or chop saw:  Cut the two transformers in half (easier after removing the shunts) and join the two primary halves at their cut surfaces with a shim.  Parallel the two windings.  (Check phasing to avoid anti-parallel.)  The two in parallel can handle the line current reasonably efficiently (where a single one is over-driven as Steve mentioned).  Adjusting the shim controls the inductance and therefore the amount of ballasting.

[kilovolt]

Klugsmith:

A ballast on the transformer input is not much help until the output current is reduced well below 0.25A.  That's roughly the 50/50 lethal current at 60Hz when flowing from one hand to another (or to a leg).

You mean 0.025A, right? Because 0.25A (250mA) is really deadly, no doubt!

Best regards
kilovolt

[davekni]

kilovolt,

There's obviously very limited data on human electrocution parameters - a few case studies and extrapolation from animal experiments.  The operative words before 0.25A are "well below".  0.25A at 60Hz is what I recall reading in a book in high school (45 years ago) as the current having a 50% chance of causing death.  (Or perhaps it was 0.25A at 50Hz and 0.22A at 60Hz.  82Hz was listed as the peak human sensitivity frequency.)  Of course, that depends on duration, size of the person, etc.  The limited bit of information I can find from web searching now is summarized on this page:
https://hypertextbook.com/facts/2000/JackHsu.shtml

Their range of 0.06 to 0.2A appears to be more of a 99% survival threshold than 50/50%.  There are cases of humans surviving (with severe burns) thousands of amps, perhaps because the plasma path engulfing them shunts much of the current after the initial few milliseconds.

I had a "friend" in high school who had a large 12kV 60mA NST.  He also had a very troubled home life.  One evening he decided to see what 60mA felt like, so grabbed the two output terminals, one in each hand.  60mA was beyond the let-go threshold, so he was stuck clenching both terminals.  Fortunately for him, he managed to pull the line cord with a foot.

Ballasting to 5ma would make it rather safe.  For 5mA, a microwave transformer is probably not the right starting point.  Ballasting to 0.1A would drastically reduce the chance of death on accidental contact, so could be valuable as an added safety feature.  Sounds like this project needs more current than that, so extreme care not to touch is the only option.  (Accidental touching with the back of a hand is much less hazardous than the inside of a hand, as the current-induced muscle contraction pulls away from contact.)