Yes, if I can find the information. As I mentioned before, the whole purpose of the thing (as far as we use it) is to do ESD testing, in other words to blow semiconductors, so I want to keep semiconductors out of its food chain. Those sparks are energy that has to circulate back to the tester, and that means through the power supply. So the way I see it, I can wind for the same amp-turns product (output is weak, current and heat is high) or I can try to achieve the same number of turns in a different gage, but that will push the operating voltage up inconveniently, and would quickly exceed the available space for windings. It seems to me that there used to be tables for such things, worked out probably at the cost of a lot of experience and copper, but I'm having no luck finding them. Here's my current reasoning: The magnet coil measures 70 ohms resistive, and I can't measure the inductance with my gear, so I have to work with the resistance number. At 70 ohms, assuming no switch, the max current would be 1.6A (assuming RMS voltage number). Sanity check, the device draws about 40W nominal, so that would indicate a current of 340mA, so it looks like a 338 ohm resistor (ish) from a total power drain point of view. So the switch must be opening well before the point where I have 1.6A flowing. The original wire is #29, at 82 ohms per 1000 feet, so there must have been about 850' 3344 turns (counted) at 0.34A is 1137 amp turns. Assume that in the end, the rest of the device should not be able to tell that I've changed my source voltage, so power in should be roughly proportional to power out for both voltages: At 12V, 40W input would be 3.33A or 3.6 ohms nominal. Seems like about 17 ohms of wire would work. #23 wire would give me 20 ohms per 1000 feet, or about 17 ohms per 850 feet= .. 1137 / 3.33 =3D 341 turns. I tried an experimental coil with #19 wire (to minimize resistance loss) and 400 turns and the output was very low. Interestingly there is paper insulation between each layer in the original coil. I don't know if it's there for heat, to reduce inter-winding capacitance (possible) or for additional electrical insulation. The voltage across the opening switch can be between 1kV and 2kV. That's high but the start and end of the winding is separated by a lot of layers, with a total stress per layer of about 400V (start of layer X to end of layer Y) which is not unreasonable for modern magnet wire. What am I missing that makes this seem so hard? Is it in fact actually hard? They did make 32V models back at the turn of the century, so it would seem to me that a 12V version would not be unreasonable. If I try to get the same number of turns, then I have a fun guessing game for length of wire and what gage will fit on the bobbin. The same number of turns of #23 vs #29 would be four times the physical volume. The total bobbin area is 2.65 inch by 0.495 inch winding depth with a 0.6 inch core containing a 0.5 inch iron (steel?) pipe which has been cut down it's length to cut eddy current losses. I thought about putting a ferrite core down the middle, but I don't have anything on hand that fits. On Wed, Apr 22, 2015 at 9:10 AM, Dwayne Reid wrote= : > Seriously cool! > > Any chance that you can share pictures and build information when you are > done? > > dwayne > > > At 07:23 PM 4/21/2015, David VanHorn wrote: > >Yup. The main solenoid is wired like a doorbell buzzer, and the resonat= or > >is a series LC across the switch. > > > >So the current starts flowing, and after a while the switch opens. The > >current then has no choice but to charge the C in the resonator (about 0= ..1 > >or 0.15uF through the 10 turn winding of the resonator. When the curren= t > >drops below the critical point, the contacts close, shorting out the > >resonator series LC circuit. There is a secondary on the resonator of > >about 1100 turns, which is responsible for the HV output. > > > >I just want it to work at 12V. Seems like not too big a demand. Total > >power consumption is about 40W. > > > >Around the turn of the century, they made these things in 220V, 110V and > >32V (for farm users) but that's when they were sold as "medical" devices= .. > >Now it's my ESD tester. The 50kV output is impressive, and any > >non-disposable electronics should stay well clear. An inverter would > >almost certainly get fried as it's in the path of the return current. > > > -- > Dwayne Reid > Trinity Electronics Systems Ltd Edmonton, AB, CANADA > (780) 489-3199 voice (780) 487-6397 fax > www.trinity-electronics.com > Custom Electronics Design and Manufacturing > > -- > http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive > View/change your membership options at > http://mailman.mit.edu/mailman/listinfo/piclist > --=20 http://www.piclist.com/techref/piclist PIC/SX FAQ & list archive View/change your membership options at http://mailman.mit.edu/mailman/listinfo/piclist .