Where can I get a non-UL solid state neon sign transformer?

[Ben321]

Most of the non-UL neon sign transformers are those big clunky metal ones (literally just a conventional electromagnetic transformer in a metal box), while most of the UL compliant transformers are the electronic / solid state type of transformers. In fact, I have yet to find one of these electronic transformers that's NOT a UL compliant transformer. For those who don't know, for a neon sign transformer to be UL compliant, it must incorporate a secondary GFI (that is a ground fault interrupter on the high voltage output). Basically if the output arcs to any grounded metal, the transformer should sense that while current is flowing, it's only flowing through one of the high voltage wires, not both of them, and respond by shutting down.

This may be great for safety on neon sign installations, but not great at all for high voltage experimentation. One might want to intentionally allow the output to complete the circuit to ground for half of the normal output voltage (most neon sign transformers have a grounded center tap, so hot-to-hot is the full voltage, but hot-to-ground is half voltage). This wouldn't be possible with a UL compliant transformer. While the simple solution might be to buy one of the old bulky electromagnetic transformers, that's not what I'm looking for, specifically because they are quite large and heavy. So I'm wondering if there's any companies out there who ever at any point in their operations manufactured electronic neon sign transformers that were NOT of the UL compliant variety. Like maybe something they made shortly before the UL rules for neon sign transformers went into effect, when companies were migrating to solid-state transformers, but before the UL rules went into effect. Or at the very least, maybe there's a company out there who manufactured a UL compliant solid-state neon sign transformer, but didn't make it hard to modify the circuit to bypass the secondary GFI (most of these UL compliant solid-state neon sign transformers have the entire secondary circuit and the GFI potted in epoxy so it's basically impossible to modify, but I'm hoping there may be at least one company that DIDN'T do this).

Or maybe better yet, maybe there's a company which sells products specifically intended for high voltage hobbyists, who maybe able to actually sell non-UL solid-state high voltage transformers as a main product (either by using professional techniques to un-pot the circuits with chemicals made to dissolve epoxy that wouldn't be available to the average consumer, or by making their own solid-state high voltage transformer designed for the hobbyist market designed from the ground up for this purpose).

My 2 interests in a solid-state neon sign transformer are that they are small and light weight, and that they output high frequency high voltage (more like 30kHz, that's 30000Hz, instead of 60Hz) so they have the added margin of safety that they won't electrocute you to death (even if their hot arc does give you a painful burn) if you have an accident while using it.

So if anybody knows where to get a solid-state neon sign transformer made without UL compliance (or one that has been modified, or one that is easy to modify, in this manner), or alternatively one that's built from the ground up for hobbyist use (so it never would have the secondary GFI in the first place), please let me know.

[Stray Electron]

  Wow, that's a long story.  Q: why don't you just find one of the old transformer type neon transformers?

     OK I reread your tome and I found this "that's not what I'm looking for, specifically because they are quite large and heavy."

   FWIW: a friend of mine used to do a lot of experimentation with high voltage and he used old TV flyback transformers to generate HV for what he was doing. At the time, he was able find them in surplus stores and from mail order surplus sellers like Electronics Goldmine. In fact we were talking about them (the transformers) the other day and he mentioned that he still has a case of them. But one problem that he had was that most of them had some sort of internal diode in them that interfered with what he was trying to do. But he figured out that he could short a charged (400 volt?) cap across one winding and that it would burn out the diode without harming the coil.

      I think that you'll find in both machinery and electronics that "large and heavy" usually translates "low cost, easy to use and reliable", especially when compared to a lot of the more modern replacements.

    Edit #2. "My 2 interests in a solid-state neon sign transformer are that they are small and light weight, and that they output high frequency high voltage (more like 30kHz, that's 30000Hz, instead of 60Hz) so they have the added margin of safety that they won't electrocute you to death (even if their hot arc does give you a painful burn) if you have an accident while using it."

    That sounds like a pretty good description of a flyback transformer.  They were designed to operate at ~16 kHz and you could probably push them to 30 kHz.  The HV in black and white TVs was about 16 kV as I recall but the HV in color TVs was closer to 25 kV so take your pick.

   The other possible candidate is to use an automobile ignition coil.

[Ben321]

  Wow, that's a long story.  Q: why don't you just find one of the old transformer type neon transformers?

     OK I reread your tome and I found this "that's not what I'm looking for, specifically because they are quite large and heavy."

   FWIW: a friend of mine used to do a lot of experimentation with high voltage and he used old TV flyback transformers to generate HV for what he was doing. At the time, he was able find them in surplus stores and from mail order surplus sellers like Electronics Goldmine. In fact we were talking about them (the transformers) the other day and he mentioned that he still has a case of them. But one problem that he had was that most of them had some sort of internal diode in them that interfered with what he was trying to do. But he figured out that he could short a charged (400 volt?) cap across one winding and that it would burn out the diode without harming the coil.

      I think that you'll find in both machinery and electronics that "large and heavy" usually translates "low cost, easy to use and reliable", especially when compared to a lot of the more modern replacements.

    Edit #2. "My 2 interests in a solid-state neon sign transformer are that they are small and light weight, and that they output high frequency high voltage (more like 30kHz, that's 30000Hz, instead of 60Hz) so they have the added margin of safety that they won't electrocute you to death (even if their hot arc does give you a painful burn) if you have an accident while using it."

    That sounds like a pretty good description of a flyback transformer.  They were designed to operate at ~16 kHz and you could probably push them to 30 kHz.  The HV in black and white TVs was about 16 kV as I recall but the HV in color TVs was closer to 25 kV so take your pick.

   The other possible candidate is to use an automobile ignition coil.

Ignition coils don't run at a high enough frequency to be really safe. I've used an ignition coil and above about 500Hz they are useless.

Flyback transformers run at 15kHz on their input but with the vast majority of them the output is pulsating DC (they have a built-in rectifier diode, and this is solidly potted in the epoxy of the transformer so it can't be removed). That's absolutely unsafe because even though there's a high frequency component to the output signal (unless it's filtered by a capacitor), there's also a high voltage DC offset. That high voltage DC will be pushing quite a few milliamps (I think the TV flybacks can output current well in excess of 10mA, which I think is considered the threshold of dangerous current) will lock your hand to it if you accidentally touch it, and you won't be able to let go, and it will then stop your heart, and you will die.

Neon sign solid-state transformers however are designed to AVOID having a DC offset on the output, because that will prematurely age one of the electrodes in the tube, shortening the life of the tube. The output of a neon sign solid-state transformer is intentionally AC, and since its oscillator operates at a high frequency (about 30kHz) that also means its high voltage output is at this high frequency.

Again, I think a solid-state neon sign transformer WITHOUT the UL mandated secondary GFI, would likely be considered the ideal high voltage source for a hobbyist. It has increased safety due to the high frequency output and it's smaller than the bulky electromagnetic type of transformer.

[Phil1977]

Why not wind it by yourself? For non-lethal power levels and a quantitiy of 1 the working time seems reasonable and you get exactly what you want.

You could e.g. start with a ZVS driver intended for inductive heaters. Just replace the heater coil with a high-ratio transformer like 10 windings primary and 5000 secondary. With 24V/10A input supply you may get 10kV/20mA output of pure AC with >10kHz.

There are LOTS of projects documented online for DIY-high-power supplies. I could tell you two german-language sites dedicated for that, but I´m sure there is something similar available in English.

PS: The rough calculation above is a worst case estimation. Even with a lower transformer ratio like 10/2000 you may get the same output voltage because of the resonant driver circuit. These things are quite tolerant. It´s fun just to try.

[Ben321]

Why not wind it by yourself? For non-lethal power levels and a quantitiy of 1 the working time seems reasonable and you get exactly what you want.

You could e.g. start with a ZVS driver intended for inductive heaters. Just replace the heater coil with a high-ratio transformer like 10 windings primary and 5000 secondary. With 24V/10A input supply you may get 10kV/20mA output of pure AC with >10kHz.

There are LOTS of projects documented online for DIY-high-power supplies. I could tell you two german-language sites dedicated for that, but I´m sure there is something similar available in English.

PS: The rough calculation above is a worst case estimation. Even with a lower transformer ratio like 10/2000 you may get the same output voltage because of the resonant driver circuit. These things are quite tolerant. It´s fun just to try.

Won't ZVS drivers differ by application? I've seen some that are designed to run an inductive heater which typically has 2 output terminals. Those designed to run flybacks (often sold for use as the high voltage power supply for CO2 laser tubes) typically have 3 output terminals, anticipating that you will be connecting a flyback transformer with 3 input pins, that are used for a tapped primary coil (where the tap is used for feedback into the driver circuit). These 3 output terminal ZVS drivers are unlikely to work with a home-wound flyback that has a simple single coil primary without a tap for feedback. I think the oscillator on these won't even oscillate without the feedback because it's designed to accept a signal from the feedback winding as part of the oscillation process.

On the other hand a ZVS driver designed to run an inductive heating coil, is not going to expect a feedback winding in the transformer, but is going to expect an air-core inductor (which is what an inductive heating coil is). However the core of a flyback is not an air-core. It's a ferite-core. This will have different magnetic properties, and thus the primary coil of the transformer will have different electrical properties than the inductive heating coil that the ZVS driver for inductive heating would be expecting.

I'm not sure how out-of-spec these ZVS drivers can go without being damaged. I'd worry that unless it was explicitly designed for hobby use over a wide range of applications, then the driver should be assumed to be an application-specific ZVS driver that could easily be damaged if used for any other application.

The other problem is if it operates at 10kHz like you were saying, that's still too low of a frequency to avoid a painful shock. Typically about 18kHz and higher is safe in terms of avoiding an electric shock. This is because our nerves don't sense electric oscillations at this high of a frequency or higher. Below 18kHz is going to give you a shock, and the lower the frequency the more painful, and if low enough actual injury can occur. I've always made it a rule of thumb to try to use high voltage for experiments only if the output is AC with a frequency of 20kHz or higher.

You can measure the frequency by just connecting a wire to the input pin of a sound card to make a crude antenna, and then recording and analyzing the signal with audio software (I use the commercially available software Goldwave for this). A very strong RF output from the high voltage oscillator will easily be picked up by a 192kHz sample rate sound card (acting as a crude VLF radio receiver) using a long wire antenna.

[Phil1977]

I can tell you from own experience that these 2- or 3-pole ZVS drivers for small "hobby"-induction-heaters are fairly robust as long as you drive them with not more than 36V DC.

Caps and IGBTs therein seem to be rated for induction coils driven with up to 300V mains, but regarding the prices for these modules they are probably fake and I don't trust their print-on type and value.

If you play with these things you will from time to time burn an IGBT or MOSFET - thats part of the fun 

Switching frequency depends on your configuration. I wrote >10kHz just to be on the safe side, as far as I remember these switchers usually run more on the 100kHz side.

From my point of view a heavily downrated switching module, supplied with eg. 24V and 2A idle current and 10A short circuit current may be quite stable. If you have a thermal imager then use it to see components getting hot before the magic smoke escapes.

PS: Personally I´d use a 100MHz-scope and a TinySA spectrum analyzer to check the switching frequency and for indications of EMI. Even if the switching frequency is low, you sometimes have really nasty emissions in the higher MHz spectrum. I don't say it´s worse than some other SMPS or especially cheap LED drivers, but anyhow, as long as I have the equipment to check it I don't want to unintentionally operate a radio station.

[Ground_Loop]

Would a microwave oven transformer work for you?

[NiHaoMike]

How about a small Tesla coil? Could wind a secondary with fewer turns of thicker wire (ideally Litz) if you want more current at a lower voltage.

[Phil1977]

---

Would a microwave oven transformer work for you?


----
There's no point getting old if you don't have stories.

You definitely won't ge old if carelessly playing with microwave oven transformers.

Use a tesla coil, use some self made transformer with ZVS, be careful when charging capacitors. But please never use a microwave transformer if you are not 250% confident of what you´re doing.

230V mains AC can be deadly. A microwave oven transformer IS deadly at the first an only wrong touch.

[Ground_Loop]

---

Would a microwave oven transformer work for you?


----
There's no point getting old if you don't have stories.

You definitely won't ge old if carelessly playing with microwave oven transformers.

Use a tesla coil, use some self made transformer with ZVS, be careful when charging capacitors. But please never use a microwave transformer if you are not 250% confident of what you´re doing.

230V mains AC can be deadly. A microwave oven transformer IS deadly at the first an only wrong touch.

Nonetheless, my question stands. I don’t feel a burning obsession to save people from themselves or presume their ignorance in matters of safety.

[Ben321]

---

Would a microwave oven transformer work for you?


----
There's no point getting old if you don't have stories.

You definitely won't ge old if carelessly playing with microwave oven transformers.

Use a tesla coil, use some self made transformer with ZVS, be careful when charging capacitors. But please never use a microwave transformer if you are not 250% confident of what you´re doing.

230V mains AC can be deadly. A microwave oven transformer IS deadly at the first an only wrong touch.

Nonetheless, my question stands. I don’t feel a burning obsession to save people from themselves or presume their ignorance in matters of safety.

Nope. Microwave oven transformer isn't a good idea. I don't have high voltage gloves, so handling something like that wouldn't be safe for me.