Page 1 of 2

Oil Transformer Demo Fusor Diagram

Posted: Sun Mar 04, 2018 9:11 pm
by Josh Smith
Hello everyone,
Here is my power supply system diagram for my demo fusor, using Richard Hull's neon transformer diagram in the FAQ. Would this diagram be feasible, especially with the way I used the USB dual channel data logger in the diagram? I'm pretty sure everything works correctly, I just want to double check with you guys that I'm not making any careless mistakes. Here are the parts I'm referencing to in the diagram:

dual channel voltage logger - ... Swp-RaXpeD
diodes - ... VVM1&psc=1
10 ohm 5W resist - ... t+resistor
100 meg resist - ... 0810356012
1k ohm 1% resist - ... k+ohm+1%25

Re: Oil Transformer Demo Fusor Diagram

Posted: Mon Mar 05, 2018 4:49 pm
by Rich Feldman
Josh, it's good to see you attending to instrumentation from the outset.

You need to find out if your 2-channel USB voltage logger has isolated differential inputs, as implied in your sketch.
The specs sheet says two single-ended inputs, 0 to +30 V measuring range.
I bet their negative terminals are common, and probably common with the USB ground connection to the computer.
(What is the input impedance? What's the input voltage range that won't damage the instrument?)

If that's the case, and you already have the logger, there's a work-around. With op amps you can easily convert the measured current and voltage
into positive voltages referred to the computer ground, which should be the same as the fusor ground.

You show an alligator clip connection in a high voltage circuit. If the clip were to become disconnected, would there be a high voltage arc between it and the thing it's clipped to? Maybe the high voltage would find a path through the USB instrument and computer instead.

Not a bad start. Good luck.

p.s. For some non-fusor applications I've been using Lascar EL-USB-3 voltage loggers. Single-channel, isolated & battery powered. You can't see the data until after the fact: unplug unit from circuit being logged, plug it into computer, upload time and voltage list. The nominal scale, 0 to +30 V with 50 millivolt resolution, is set by a voltage divider inside the USB-to-clip-leads logger head. With that bypassed, it's 0 to +2.4 V, as demonstrated in my thread here about solid state hydrogen storage. I just got 3 used ones on ebay for about $70 total.

Re: Oil Transformer Demo Fusor Diagram

Posted: Mon Mar 05, 2018 6:57 pm
by Richard Hull
This circuit looks good. You are using this system as a full wave system with ground splitting the 10kv to 5 KV RMS which will give to a peak maximum output voltage of around 7KV at the fusor input. This is fine for a demo fusor. You might expect about 5ma max at this voltage before the core shunting cuts in and start dropping you voltage.

Richard Hull

Re: Oil Transformer Demo Fusor Diagram

Posted: Tue Mar 06, 2018 7:39 am
by Rex Allers
As others have indicated, the general layout of your scheme looks ok but I see a three areas that look like there could be problems.

1) In your drawing the output indicates two diodes (wish you had drawn proper symbols because direction matters) in a configuration that implies the transformer secondary between those two outputs has a center tap that is grounded to the case. I've never played with any oil burner transformer but I did a little web surfing tonight. One key thing I learned is that most of the newer ones (last 15 - 20 years) contain switching supplies. They make outputs in a frequency of 20 kHz or more. If you used one of those types of source the microwave oven diodes probably are not fast enough to use with the 10's of kHz frequencies.

Trying to understand the ones that have two output terminals (on many they are springs) I never found any information that was detailed enough to say if the two outputs have a center ground reference like a center tapped transformer would. This is necessary for the circuit you drew. Maybe others here have more detailed information on these oil burner transformers. Without that you could get in trouble.

On ebay I did see one listing that I would trust -- "Allanson 542 Type H Ignition Transformer, Oil Burner". It has only one big output insulator knob with a screw terminal. The label says the output is 10 kV at 23 mA (like your diagram) but says one end of the secondary is ground. It looks to be a conventional 60 Hz transformer. So you could use this but the output would tie to only one diode. The peak HV would be about 14 kV, so nice -- but it was about $100 in the listing.

So give us more detail on the transformer you have or are planning to use. Maybe you or someone here can find more details about how usable it might be.

2) You show using one 100 Mohm resistor as your main part in the voltage measurement divider. Just as a gut feel, that seems low as a HV measurement sensor. But, more important, the ebay listing you gave lists the voltage rating for this resistor at 3500 V. So even if you are only measuring 7 kV this one resistor won't cut it safely.

The ebay listing said 10 pcs of the 100 M resistors. If it was me, I'd put all 10 in series for a 1 Gohm resistor that is rated for 35 kV. Then you'll have something safe that will be usable later on a real fusor if you have a supply in the low 30 kV range.

If you use all ten resistors, you'll need to have a bigger resistor on the bottom of the divider, but more about that to come.

3) Rich mentioned some ground reference potential problems with your digital data acquisition. That's correct but there is more.

The listing says the device is 30 V full range and 8 bits. So each bit is a step of 30/255 or about 0.118 V. I think the listing mentioned something like that too. That is very coarse and especially if we look at what you will be measuring with your sensing circuits.

Let's assume your 100M resistor was a 20 kV device, so not overloaded. The divider ratio is ( 100 M / 1 K ) + 1 = 100001. So each 1 kV measured produces about 10 mV that we are measuring. So bad bad bad -- each 1 kV step of voltage is less than 1/10 of a bit on your ADC device.

Let's assume you expect to measure current in the 1 to 20 mA range through the fusor. The 10 ohm resistor you chose as the sensing device will produce 0.01 V/mA, so only a sensed voltage of 0.2 V for 20 mA. That's about 1 bit of your ADC = no measurement across the range.

So if you want to use digital acquisition of these measurements (not a bad long-term goal) I'd suggest a measuring device that can be set to no more than 5 V full scale and 12 or more bits of resolution. Then I would pick low end resistor value of the divider and a current sensing resistor value to give outputs big enough for easily sensing with decent accuracy the expected high voltage and current you might see.

In the short term, I'd skip the ADC and just use two voltage meters that can measure at minimum to the millivolt.

I think you have read enough of the FAQs to have something that is close to a proper design but I think you have a lot more you should work out before trying to build it.

Also, maybe, this is more of a New User Chat topic, as a best place to post.

Re: Oil Transformer Demo Fusor Diagram

Posted: Tue Mar 06, 2018 8:21 am
by ian_krase
I use an OBIT for demo fusor and sputtering purposes.

While some OBITs are actually switching supplies, it is still fairly common to find iron-and-copper actual transformers. Much more common than in neon-land, and they never have GFCI which helps. They are indeed center-grounded to the case. I have four diodes providing two full-wave rectifiers, one for positive output and one for negative output. All microwave oven diodes.

At least on my OBIT, the springs are easily removable -- they are threaded onto screw studs. I removed and replaced with nuts and ring terminals.

Re: Oil Transformer Demo Fusor Diagram

Posted: Tue Mar 06, 2018 3:22 pm
by Josh Smith
Here is the OBT I have - ... 2025!US!-1.

I don't have the loggers yet, but I'll wait to worry about them after I get the rest of my circuit working. But, based off the information you all gave, I should be able to figure out how to digitialize my data now, so thanks for helping on that end.

I actually already have the circuit set-up, and did a test to see if I could get some plasma last night. Unfortunately, I got nothing, but It's more likely than not a power supply issue than anything else. I brought down the vacuum in my chamber to around ~30 microns, but even after turning the variac all the way up to 120V, I still got nothing. I tested the wire connections with a battery hooked up to the wire directly after the diode, and I got the exact same voltage all the way through the wire on the grid with a multimeter. I have also previously tested the transformer at max applied voltage with a screw driver, and was able to get sizable arcs, so I don't think the transformer is the issue either. Thus, I determined it must be an error with the diode. Could it be the frequency issue as you said - or could I be putting the diode in the wrong direction? Here is a more detailed diagram of my power supply setup - maybe it will give more insight. I can also post pictures of my set-up as well if need be. Also, thank you guys for your feedback - I'm definitely learning a lot but there is much more that slips through my uneducated mind, and you've all been able to point out things I didn't even consider.

Re: Oil Transformer Demo Fusor Diagram

Posted: Tue Mar 06, 2018 6:36 pm
by Richard Hull
Rex brought up a good point on the A to D converter. I am sorry but 8 bit for current measurement across the resistor is terrible. 10 bit is marginal, 12 bit is a start at decent resolution. The 10 bits of the Arduino is barely OK as its span is only + 5 volts (non-differential).

When choosing an A-D system converter one must do the math and look at the input span vs the resolution of the bits over that span.

Richard Hull

Re: Oil Transformer Demo Fusor Diagram

Posted: Tue Mar 06, 2018 9:10 pm
by ian_krase
I feel like a logarithmic amplifier might be desirable here

Re: Oil Transformer Demo Fusor Diagram

Posted: Tue Mar 06, 2018 10:52 pm
by Rex Allers
Your circuit looks ok to me, except it doesn't show a ground wire from the fusor shell back to the transformer case. I assume you actually had it and just left it off the drawing. Without it, it can't work and is unsafe.

I looked on the web for the transformer manufacturer and found these two pages. ... ansformers ... ers_lr.pdf

They didn't come out and say it, but I saw no indication that this one is a switcher type and they say this, "Mid Point Ground", which is what you want for the way you have the diodes connected. As Richard said, your 5 kV RMS out (1/2 of 10 kV) should make 7 kV peak as DC.

I would put together a divider so you can measure the HV with and without the fusor connected. I assume you did purchase 10 of those 100M resistors. I would use at least 4 of them in series which should be good to 14 kV (3500 x 4). Assemble something like this...
HV divider.png
That resistance ratio will get you close to 1 V out equals 1 kV measured. You'll need to get a 400K resistor for the bottom end. A 420 K may be easier to find and would only be off by about 5% high output.

Do you know how to solder? If not, that is a skill you should learn to make safe and reliable stuff like this.

Do you have a DMM (Digital Multimeter)? If not, you should get one, or better several. Here's one common source for a cheap but good enough meter. ... 90899.html
A set of leads that have clips on the ends is helpful too, especially for a connection like this.

In the divider drawing I gave I connected the meter so that it will indicate (-) readings for (-)HV. An old-style meter that can only swing positive would need the lead connections reversed.

Re: Oil Transformer Demo Fusor Diagram

Posted: Wed Mar 07, 2018 2:01 am
by Richard Hull
I have written extensively about the Harbor Freight meters. I now have about 106 of them here at the lab and didn't pay one penny for any of them. They are regularly listed as free on coupons sent to my home, received in the mail and in the news papers and any number of men's magazines. I also get free coupons on sales flyers when I go to Harbor Freight at the checkout counter as well. I started picking these up in 2013 and now have a huge stock of them. 12 meters have been converted to 2kv meters, meters on my Arduino test stations and other fixed non-portable uses around the lab. I have so many that I remove all batteries which are included inside the meters so I can store the meters without fear of leakage.

Richard Hull