Anyone know a commercial source for HV, HF transformer windings? / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

Anyone know a commercial source for HV, HF transformer windings?

previous 1 2 3 4 next

Move Thread

LAN_403

Wolfram

Sun Mar 17 2013, 06:28PM

Registered Member #33 Joined: Sat Feb 04 2006, 01:31PM
Location: Norway
Posts: 971

Loads of good info in this thread.

I would be very interested in some figures on the transformer in the first post, particularly the dimensions, SRF, leakage inductance and turns count. Do you use a resonant driving scheme, and do you have any other resonant components in the primary circuit aside from the leakage inductance and the reflected secondary capacitance?

This thread has inspired me to try to make some similar transformers. I'm thinking of making a small microcontroller driven winder for these sorts of coils, and if I ever get around to it I will surely post details and code here.

I found an interesting book on coil winding, "Coil Winding" by "Geo. Stevens Manufacturing" from 1954. It has a chapter on these "Universal" windings. I haven't read all of it yet but it looks like it contains some good info. The book can be found online, but I'm not sure if it's out of copyright so I won't post it here.

Ash Small

Sun Mar 17 2013, 09:34PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

Wolfram wrote ...

This thread has inspired me to try to make some similar transformers. I'm thinking of making a small microcontroller driven winder for these sorts of coils, and if I ever get around to it I will surely post details and code here.

I've spent the past few hours researching stepper motors with a view to doing this as well (I mentioned above that I was considering building a manual one first, but I'd want it to be easily converted to stepper motor control).

I'm currently researching the feasibility of using a linear motor for the cross-feed.

The coding should be pretty straightforward (I've not done much coding for well over 30 years, but it's just a case of setting up a few counters, from what I can make of it), and running some transistors (presumably MOSFETS) from the IO pins of a RasPi.

As there are three of us here who've now expressed our intention to build one, should we start a new thread, where we can compare ideas, etc?

(My immediate requirement for one of these machines is for my HV supply detailed recently elsewhere on this forum, but the ability to wind coils similar to the OP's would also be useful for other projects)

Thanks for the details regarding the book. I'll look for it later.

Carl A. Willis

Mon Mar 18 2013, 03:04AM

Registered Member #9640 Joined: Wed Jan 16 2013, 07:53AM
Location: Albuquerque, New Mexico
Posts: 26

Wolfram wrote ...

I would be very interested in some figures on the transformer in the first post, particularly the dimensions, SRF, leakage inductance and turns count. Do you use a resonant driving scheme, and do you have any other resonant components in the primary circuit aside from the leakage inductance and the reflected secondary capacitance?

My equipment is a little bit limiting, since I don't have a good impedance bridge or analyzer that can measure high inductances reliably. If anyone has a good suggestion for such a piece of test equipment, I am sure I could put it to use. The primary inductance, off core, is 0.9 uHy; on-core it is 10.6 uHy; the secondary inductance, off core, is 40.0 mHy; the core is an ungapped 3C81 material. In the Glassman supplies, they gap these cores with a plastic shim.

My driver is very similar to this ol' favorite here. It's an SG3525 PWM chip driving a pair of fets in push-pull. I use a 15V / 8A power supply. The circuit's publisher, Bob Iannini, sells a version of it on his Information Unlimited web store as the PVM-12, albeit with a couple features to "nerf" it and a random selection of a fragile, badly-performing, and worthless HV transformers (at least in my experience!). What I have done on the three drivers I've built is to drive a 1:1 isolation transformer (8+8 turns on the primary, 8 turns secondary) with fairly ideal characteristics. The secondary of this transformer is then used to drive the modified Glassman transformer, or some other transformer according to what's available.

Here is a video I took today showing how an Info Unlimited PVM-12 is re-purposed for high power output and for driving an alternative transformer (albeit with a different transformer than the Glassman).

Hope this meager information is helpful.

-Carl

Ash Small

Mon Mar 18 2013, 09:30AM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

Carl A. Willis wrote ...

My equipment is a little bit limiting, since I don't have a good impedance bridge or analyzer that can measure high inductances reliably. If anyone has a good suggestion for such a piece of test equipment, I am sure I could put it to use. The primary inductance, off core, is 0.9 uHy; on-core it is 10.6 uHy; the secondary inductance, off core, is 40.0 mHy; the core is an ungapped 3C81 material. In the Glassman supplies, they gap these cores with a plastic shim.

You can measure inductance quite accurately using a signal generator, known value capacitor, and' scope. I usually use ~100 Ohm resistor in series with the sig. gen. output for protection. Leakage inductance of the secondary can be measured by shorting the primary, and vice versa.

Stray/self/distributed capacitance of the secondary would presumably affect these results. I assume this could be measured either directly, or indirectly using two or three known value capacitors, and then extraplpolating.

Maybe someone else can either confirm this, or suggest improvements/alterations to this method?

EDIT: I know the accuracy of this method isn't ideal, but I imagine it could be improved, any suggestions would presumably be useful to a number of people reading this thread. (Maybe a simple bridge, using 'known value' components?).

klugesmith

Mon Mar 18 2013, 07:37PM

Registered Member #2099 Joined: Wed Apr 29 2009, 12:22AM
Location: Los Altos, California
Posts: 1716

Carl & Ash,
When measuring inductance with coils on cores, don't forget the nonlinear permeability and hysteresis at low flux levels. For meaningful measurements on power transformers, you need a strong enough signal to properly magnetize the core.

Yes, it isn't hard to get the linear parameters of inductors using a sine-wave generator and a 'scope to measure voltage amplitudes. With a known capacitor and/or resistor (perhaps Z_out of the signal generator) connected to the device under test, sketch out a BODE plot. Generally you can skip the phase measurement.

The value added by fancy RLC meters is automation of the frequency sweep, V and I measurement, and computations.
But I find it enlightening to do the last part myself, as in this chart of measurements using an HP4285A RLC meter (75 kHz to 30 MHz).

The DUT is a circuit board Rogowski coil (described in another thread) with just its own self-capacitance. At the low frequency end we can see the knee where DC resistance takes over. The phase measurement nails that pretty well. I bet it would be easy to use the coil by itself as the tank in a 21.7 MHz oscillator.

I'd love to measure one of Ash's or Carl's coils on that machine.
-Rich

Ash Small

Mon Mar 18 2013, 10:22PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

klugesmith wrote ...

Carl & Ash,
When measuring inductance with coils on cores, don't forget the nonlinear permeability and hysteresis at low flux levels. For meaningful measurements on power transformers, you need a strong enough signal to properly magnetize the core.

Meaning you just need to amplify the signal?

I seem to remember coming accross some similar graphs (especially the blue line) when researching Smith charts and Pi circuits (or similar matching circuits)

There is a wealth of useful information in the book that Wolfram linked to. I've not finished reading it yet, but I'd recommend it to anyone who is considering winding their own HV transformer. (or any other coil, for that matter)

Tony Matt

Sun Mar 24 2013, 03:57AM

Registered Member #3700 Joined: Sat Feb 19 2011, 12:59PM
Location:
Posts: 107

Ya...

Some time ago I used honeycomb windings imerse in oil, commecial transformer oil from Shell, Mobil, etc . I have learned that is a unnecessary complication.

Now I used to make the transformers air dry. I wind my own HF and HV transformers using a simple crankcase and a turn counter, allways powered by hand. The wire is also guided by hand. The machine base is made with 2"x4" pine, clamped in the table.

The secondary coil (that one that carries the high potential) has several wire layers. The insulation between the wire layers is Mylar (polyester) film, 0.1 or 2x0.1 mm thick film. The insulation thickness is dictated by the max voltage between two adjacent layers. The number of turns in each layer is calculated to produce not more than 1200 or 2500Vpeak between two layers ( not more than 600 or 1250Vpeak per layer).
Also the insullation film is 6 to 10mm wider than the wire layer wide, allowing proper creepage insulation between the layers.

After the winding done, current injection heathing for dry at air, alkid varnish imersion, air dry again. The tests are done with double of frequency double voltage for one minute.

Such devices are running 24 hours /day, 360days per year.

Tony Matt

Sun Mar 24 2013, 08:59PM

Registered Member #3700 Joined: Sat Feb 19 2011, 12:59PM
Location:
Posts: 107

here is the machine...

1364158765 3700 FT151510 1301695125 3343 Ft0 P A Cranckcase And Turn Counter Dscf0058

Ash Small

Sun Mar 24 2013, 09:42PM

Registered Member #3414 Joined: Sun Nov 14 2010, 05:05PM
Location: UK
Posts: 4245

Tony Matt wrote ...

here is the machine...

I'm planning on comparing self capacitance of coils of the type you use, Tony, and honeycomb and basketweave coils, and also measuring leakage inductance, as part of my 'calculating self resonance of HV transformers' thread.

I'm playing around with NEMA23 stepper motors at the moment, for my 'universal coil winding machine'. I'll post on progress with drivers for the steppers soon.

Tony Matt

Sun Mar 24 2013, 10:15PM

Registered Member #3700 Joined: Sat Feb 19 2011, 12:59PM
Location:
Posts: 107

Hi Ash !

If you are going to produce 1 or 10 identical coils, the best is the crankcase & turn counter machine. This means " man have complete control of the work "

After you had a tested and aproved a coil, and are going in mass production, you may use the coil parameters to produce 100 or more similar coils in a fully automated power drive machine.

The turn counter is that one that allows regressive count >> that one that goes back when you turm the crank in the reverse...

Leaking inductance calculations are related with the geometry of the coils, not with the type of winding. see Handbookof Transformer Design and Application , Coupled Inductor Design, by Lloyd Dixon. Sometimes, the leaking inductance is desired , as the some wellding transformers.

previous 1 2 3 4 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob