Criticize my HV supply design

This forum is for specialized infomation important to the construction and safe operation of the high voltage electrical supplies and related circuitry needed for fusor operation.
Silviu Tamasdan
Posts: 147
Joined: Wed Jun 28, 2006 7:17 pm
Real name: Silviu Tamasdan
Location: Connecticut

Criticize my HV supply design

Post by Silviu Tamasdan »

This will be a long winded post, or perhaps even several posts, so bear with me as I go through this. I've posted this somewhere else already (but this won't be a literal copy/paste of the material from there, it's mostly re-written including some unchanged snippets). I've edited it quite a bit for posting here, eliminating irrelevant stuff, and adding new material as well.

NB as of 11/10/17 no HV has been generated from any of the designs below; currently the practical build is on hold until sometime next week due to other obligations.

Essentially it comes down to this: I have been unable to obtain anything suitable for a HV power supply for my fusor, so I decided to make my own from scratch. It's the only major subsystem for my fusor in which I haven't been able to make any progress with in years, including scrounging or buying major components. Notably the HV transformer of adequate voltage and power specs.

My aim is now to achieve a power supply that can deliver a minimum of 30kV at a minimum 30mA, power 900W or more. That's goal #1.

I was unsuccessful in finding an affordable power supply with the required parameters. While I am still looking, I've decided to make my own power supply. For efficiency, it will be a switched mode power supply (SMPS) with a ferrite core transformer.

All calculations below are for a frequency of the SMPS of 20kHz.

A quick back-of-the-napkin calculation gives me this for the requirements of the ferrite transformer. For a supported power of 900W, it would need a core area of at least 35cm^2. The power supported increases with the square of the core area.

Cores that I can find at reasonable prices online are EE140 type, which have a core area of 4x4=16cm^2. One of these supports about 180W power transfer. Not enough.
But if I stick 2 of them together in parallel that will give me a core area of 32cm^2, or a power of about 750W. Still not enough.
How about 3 cores? That is an area of 48cm^2, and maximum supported power of about 1700W. Check.

The material of these cores is a Mn-Zn ferrite, PC40. Datasheet gives for PC40:
-initial permeability 2300 at 23 degrees C
-Curie temperature above 200 degrees C
-saturation flux Bs of 0.5T at 23 degrees, or 0.38T at 100 degrees C; this is important.

I will assume that it will be pushed to its limits and calculate for max temperature of 100C.
From previous failures I have acquired the habit of over-engineering things. This not only gives me a safety factor, but also allows me to upgrade stuff in place if I need higher specs later. Some speculations later in this post regarding upgrades.
So I will multiply the Bs at 100C with a factor of 0.3 and use that for calculations as maximum flux.
That gives me a Bs of 0.38 * 0.3=0.114T, or 114mT or 1140gauss.

Calculations for primary winding.

B(gauss)=V*T(on)*10^8/2*Ae*N

V=primary voltage
T(on)=duty cycle divided by frequency
Ae=core area (cm^2)
N=number of turns
Or expressed otherwise
N/V (turns per volt)=T(on)*10^8/2*Ae*B

T(on)=0.5(square wave with 50% duty cycle)/20000Hz=0.25*10^-4
Ae=48cm^2
B=1140gauss

thus N/V=0.25*10^4/2*48*1140=0.25/10.944=0.0228 turns per volt

For the primary I will use mains AC (120V in the US), passed through a 2KW autotransformer for power control, rectified with a full-wave bridge and filtered, then switched on/off at 20kHz/50% duty cycle with power MOSFETs or IGBTs (probably a half-bridge configuration) into the primary.

120V is the Vrms of the mains; Vpeak is 169V (Vrms/0.707); Vaverage is 108V
I will aim for an average voltage of 12000V in the secondary, thus use the 108V value in the subsequent power calculations. The Vpeak in the secondary will be 18837V, and I will use this to select components tolerances. I won't calculate a transformer that gives me more than this for practical reasons (insulation etc), and will raise the voltage to the desired values using a Cockroft-Walton voltage multiplier in the secondary.

I will derive the current needed in the secondary from what the CW multiplier needs to achieve the desired output parameters.
For an input of 12000V, a 1-stage CW multiplier using 4nF capacitors at a frequency of 20kHz, in order to supply 30mA output will require at least 83mA input current. The output voltage will be 33kV with a ripple of 380V. That is acceptable.
Will round up required current to 100mA. This is what I will use in subsequent secondary calculations

(subsequent update and edit: I now have 10nF/30kV capacitors and if I use 2 in parallel for the stages that gives me 20nF capacity; plugging that in the CWM calculations I get for a 1-stage multiplier:
Iout=30mA, Iin=84mA, Vout=33.6kV, ripple=75V, total power 1005W. That is even better. Will use that)

Primary turns.
N/V=0.0228 (from above) V=108V (average), 168V(peak), Vrms=120
N=2.46 turns for average voltage, or 3.69 for peak. Round up to 4 turns.

Secondary turns.
Vavg=12000V, Vpeak=18837V. Turn ratio=12000/108 + 10%(compensate for losses)=111.11+11.11=122.22. Round up to 123
Turns=123*4=492.Check.

Secondary wire gauge.
The wire needs to carry 100mA at least, so a minimum of 31AWG, or 0.226mm diameter. That will take a current of up to 113mA. Check.
If I later would require the secondary to provide more current I could use 30AWG (up to 140mA) or AWG29 (up to 180mA). It would be pointless to increase the wire gauge above that, since at 180mA the power would be 2100W, which is more than the transformer core supports. AWG30 would support up to 1680W which is within core spec. Make a note to not exceed 140mA in the secondary.
(update: due to practical considerations will use 28AWG wire. Supports up to 230mA, much more than needed)

Primary wire gauge.
Current in primary: 100mA*123(turn ratio)=12.3A Will apply the same overengineering concept and round up to 15A minimum.
Wire size for 15A is 10AWG, or 2.58mm diameter.

But hold, at 20kHz there will be a skin effect. The skin depth at 20kHz is 0.5mm, so any wire over 1mm diameter will be affected by a reduction in its current carrying capacity. For 10AWG, this leaves a surface area equivalent of 21AWG (0.72mm diameter) which can only carry 1.2A. Not good enough.

Quick area calculations taking into effect the skin effect shows that I will need primary wire of at least 6.89mm diameter. 1AWG(7.34mm diameter) would be needed.

But hold again. Working with 7.5mm copper is going to be a nightmare, and wasteful of conductor area utilization. Better plan is to wind the primary with several strands of thinner wire in parallel, each carrying less current but the sum of the currents meeting my specs.

How about 12AWG wire? Diameter 2.05mm; the skin depth is 0.5mm so subtract 1mm (2*0.5) leaves 1.05mm unused in the core. That's equivalent to 18AWG which means I will derate the 12AWG current by a 18AWG current, thus 9.3A-2.3A=7A per strand at 20kHz. 2 strands in parallel would carry 14A which is close to 15.

How about 14AWG wire? applying the same calculations I need to derate it by the equivalent of a 22AWG due to the skin effect, thus 5.9A-0.92A=4.98A. A secondary wound with 3 of these in parallel would carry 3*4.98A=14.94A. I think I will chose this version which means the secondary will be 3*4turns 14AWG wire.


So overall I will need:

3*EE140 cores in parallel
3*4 turns of 14AWG wire primary wound over all 3 cores at once
492 turns of 31AWG wire (or better-28AWG chosen for practical reasons) wound over all 3 cores at once.

End transformer calculations.

Input: AC mains through autotransformer, full wave rectifier bridge, filtering capacitor (all specified for at least 200V, 15A); half-bridge MOSFET/IGBT plus driver at 20kHz.
Output: 1-stage CW multiplier, 33kV at 30mA.
Power required for output including CW losses: 1005W, approx 1kW.

Can later add a second CW stage for higher output. With the second stage, output would be 66kV/30mA with 530V ripple, but the secondary current would rise to 161mA and total power required rise to 2000W. That is out of spec.
However if I only require a current of 18mA the output will be 66.8kV with 315V ripple, and secondary current required 100mA for power of 1200W. That is within spec. So this upgrade in place would be viable without needing the transformer to be rewound. More voltage and more power=better fusion efficiency.
The maximum I could extract from this design with 2-stage CW multiplier, based on the 140mA limit for the secondary current set above, would be 25mA at 66.7kV with 440V ripple. That gives a power of 1669W.

The weight of the transformer, based on the weight of the cores plus that of the wire, is estimated at about 14kg. Size 14x14x12cm.

Please feel free to check my calculations and make appropriate corrections.


------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

While I plan forward about a big power supply, I'm starting to build a small one, just about 2W of power (later revised to 40W, see below), for the specific purpose of testing my designs and components. I have a (large for other purposes, but small for a power transformer) ferrite toroid of unknown material (I think however that they are more or less an equivalent of FT290-43, http://toroids.info/FT290-43.php ) that I got a while ago from a Chinese seller; it's 76x38x10mm (ODxIDxh)
The front-end will run at 12V and will have a 20kHz square wave generator, a gate driver, a gate driver transformer that will provide out-of-phase gate signals to the 2 power MOSFETs/IGBTs that drive the transformer primary in half-bridge configuration. If that works well in small-scale (12V primary, 1000V secondary) then I can scale it up with the same drive circuit but bigger MOSFETs and power transformer.

This will allow me to test
a. my driver design
b. my skills at winding toroidal transformers, which I've done in the past but not to a great extent
c. components, especially MOSFETs, HV capacitors and diodes.
d. RF interference on my neutron detection system. A small HV arc set not far from the detection tube should provide ample amounts of that.

------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

I have decided to use IGBTs instead of MOSFETS in the power driver. It won't change much in the other electronics part, but I expect to have smaller power losses in the driver stage, and thus less heat generated.
Why? A MOSFET is essentially a voltage-controlled resistor. Even when fully "on" it's still a resistor; a small one, but the voltage drop on it increases with the voltage it switches. An IGBT is essentially a BJT which gives it a Vcesat in the parameter list - essentially a fixed voltage drop which varies little with the switched voltage.
At low voltages, say 12V, the MOSFET is better. Say we have a MOSFET with Rds(on) of 100mOhms. That means it will drop 1.2V at 12V. IGBTs have a Vcesat on the order of 1.8-2.5V. So at 12V the

MOSFET will waste less power and generate less heat. But at 120V, the MOSFET will have a voltage drop of 12V, while the IGBT will still only dissipate 1.8V.
IGBTs are a little slower than MOSFETS, but that would only matter if I were using a switching frequency of 400kHz or more. Since I use 20kHz that won't matter.

I found in my electronics stash some 600V/60A IGBTs left from a previous project, and will use those. If needed, I have a few 1200V IGBTs on order which I expect to get in a couple of weeks.

------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

So I sat down and spent my Sunday by winding a transformer. I'm testing the "stacked cores" hypothesis. The medium ferrite toroids that I have, if used individually should support 2.5W, but if stacked by 4 would reach 40W or so. If this tests good I can scale up to the stacked extra-large EE cores.

Here are pictures (pics 1,2,3). Testing another day, I'm too tired today. This business took me 2 hours at least. It reminded me how much I hate winding toroids. I had avoided doing it for 20+ years until today.

Anyway, specs are as follows.
4xtoroid (cheap, unknown parameters from China), 76mm OD, 38mm ID, 13mm height
secondary: 170 turns 28AWG wire (I didn't need to use 28AWG, 35AWG would have been enough but it would have been technically more difficult); hopefully will give around 1000V, 40mA.
primary: 2 turns, 2 strands 18AWG wire; operating at 12V, 20kHz, 3.25A if the power rating is correct
Need to build an appropriate driver.

As far as the generator/driver goes, I don't need any advanced PWM features, just a fixed frequency oscillator, an intermediate amplifier/gate driver and the half bridge itself. A 555, a few transistors, a small gate transformer and the IGBTs are all I need beside the large transformer itself. And if I calculate everything right I can build it once, and use both with the test transformer above and with the final transformer.

------------------------------------------------------------------------------------------------------------------------------------------------------------------------------


Here's a quick and dirty schematic (pic5). It's not original, I lifted a large part of it from http://www.stevehv.4hv.org/FBD/FBschematic.JPG (but even that isn't really original, it's a classic 555 oscillator with 50% duty cycle, an integrated driver which I replaced with a couple transistors, and a classic half bridge). The component values are based on what I found in my boxes and could be replaced with anything equivalent or better. The gate drive transformer should be wound with at least 24AWG wire to allow for currents up to 500mA. One of the secondaries is wound in the opposite direction compared to the primary and the other secondary. Pic4 is the gate transformer finished.

I didn't add over-current protection, but that could go like this: small current-sensing transformer in series with the big transformer primary in the bridge; secondary from this would be rectified and fed into a comparator, then the comparator output into an optocoupler; the optocoupler output is used to reduce the drive in the driver, e.g. by switching different resistors instead of the one in the base of the second transistor. This resistor (which I marked 100-1k) is what determines the strength of the gate drive. It should be determined experimentally, and will depend on the power components in the bridge and their gate drive requirements.

Or even better, a small signal MOSFET such as BS170 connected between the base of the first transistor and the ground could do the job; perhaps even working as an AGC/voltage-controlled resistor to reduce drive signal if output exceeds a certain limit.

The IRF640 should be good enough for operating the power section at 12V; but for 120V either higher-rated MOSFETs or the IGBTs noted will be necessary.


-----------------------------------------------------------------------------------------------------------------------------------------------------------------------


AGC /overcurrent protection idea. (pic6)

For a modular build, everything from the 555 up to and including the gate transformer, and the optocoupler would be in one module (the signal module); everything downstream and including the power MOSFETs, and the current sensing circuitry would be in the power module. The current sensing part would be dimensioned to the power transformer and voltage in use, and would provide the same signal to the OC. I could then switch power modules depending on the transformer I use and keep the same low-signal part.


--------------------------------------------------------------------------------------------------------------------------------------------------------------------------


Let's discuss the secondary of the final large core EE transformer. I actually have a question (or several) for the collective mind of this board at the end.

Per calculations above I will need 492 turns of 30AWG or heavier wire. I have been shopping around, and high-voltage-rated 30AWG wire is hard to find in appropriate lengths (200m is what I need) not to mention fairly expensive even from the usual Chinese sources. I was able to find some 28AWG wire in 300m lengths that's fairly affordable (thus the updates in earlier parts of this post with 28AWG used). The catch is that it's rated for 3000V not 10000 or 20000 as I was hoping. I will come back to that.

Now assuming that I use this wire, based on the number of turns, the physical properties of the wire and the space I have available in the core window, the most reasonable geometry would be to wind it in layers of 41 turns per layer, 12 layers total. 41*12=492.
The voltage in the secondary will be 12000Vrms. That comes to 1000V per layer. That would mean the rated voltage résistance of the wire isn't reached unless you consider the difference of potential across 3 layers deep (e.g. from the 1st layer to the 4th layer). Even when considering the peak voltage of around 19000V you still have to look across 2 layers deep (1st layer to 3rd layer etc).

Would this be safe? What if I use some extra insulation between layers, say some thin polypropylene/PVC or silicone film? What if I encase the whole secondary in silicone or epoxy resin?

Also, what if I submerge the transformer in mineral oil as it's usually done with HV transformers? Actually I have some reservations regarding that. The wire insulation is silicone. I don't know if this silicone is resistant to mineral oil, or if its mechanical and/or electrical properties wouldn't be degraded by the oil.

As always I appreciate your opinions.

------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

That's all she wrote for now.

I know we have HV experts on this board. Please point any flaws in the plan above. Better I find out now than later.
Above has been proof-read twice, but if any spelling errors are still left please use your imagination to correct.
Attachments
primary 80% completed
primary 80% completed
completed secondary
completed secondary
finished 40W transformer
finished 40W transformer
gate transformer
gate transformer
oscillator/driver
oscillator/driver
AGC/overcurrent protection
AGC/overcurrent protection
Last edited by Silviu Tamasdan on Fri Nov 10, 2017 1:08 pm, edited 1 time in total.
There _is_ madness to my method.
User avatar
Richard Hull
Moderator
Posts: 14991
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: Criticize my HV supply design

Post by Richard Hull »

Interesting desin calc's. Designing a switcher to do what you need is a real job. You are to be commended. I think you will need to build the small system first and see the results vs. calculations. You might be in for a real awakening in the testing of the result. I was hoping to do a half-bridge design, but have little impetus to do it as I have a good supply already in use.

Richard Hull
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
The more complex the idea put forward by the poor amateur, the more likely it will never see embodiment
Silviu Tamasdan
Posts: 147
Joined: Wed Jun 28, 2006 7:17 pm
Real name: Silviu Tamasdan
Location: Connecticut

Re: Criticize my HV supply design

Post by Silviu Tamasdan »

Richard Hull wrote: Fri Nov 10, 2017 12:57 pm Interesting desin calc's. Designing a switcher to do what you need is a real job. You are to be commended. I think you will need to build the small system first and see the results vs. calculations. You might be in for a real awakening in the testing of the result.
I may very well be, that's why I plan to build and test the small one first (with the 2 transformers done, that's more half-way through). If anyone can tell me specifics on that awakening please do as to make it less "real". :)
I'm still unable to test high voltages even if I were to complete the build today (unless making some arcs in air is considered testing). I'm waiting for some high voltage resistors in the mail to make a proper divider. Should be here in a few days. Gives us plenty of time to discuss about it.
There _is_ madness to my method.
User avatar
Richard Hull
Moderator
Posts: 14991
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: Criticize my HV supply design

Post by Richard Hull »

You should never spark a switcher system! All manner of deadly pulsed energy can be forced into the solid state stuff. Build a suitable metering system first. Test it on a simple 15kv rectified neon transformer system hooked to a variac to see that it is performing as expected.

Richard Hull
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
The more complex the idea put forward by the poor amateur, the more likely it will never see embodiment
User avatar
Finn Hammer
Posts: 298
Joined: Sat Mar 05, 2016 7:21 am
Real name: Finn Hammer
Contact:

Re: Criticize my HV supply design

Post by Finn Hammer »

Silvio,

Impressive study of the transformer, but the size of that transformer core and associated windings seem awfully big.
36cm^2 is the size I would expect from an iron core at 50-60Hz, not a ferrite transformer switched at 20kHz.

I will have to dig up my copy of Snellings "Soft Ferrites" to substantiate this claim, hopefully I can do that tomorrow.

You will want to make the primary coil out of either Litz wire or flat copper strip.

As an example of modern switching supply transformers, Here is a photo of the 2 transformers in a Spellman 30kV/20mA supply, although it switches at 100kHz, the secondaries are smaller than a silver dollar.
Are you aware of the forum at www.4hv.org ?. It might be the right place to document your build.
The two ferrite transformers, with silver dollar sized secondary windings.
The two ferrite transformers, with silver dollar sized secondary windings.
Another thing you want to have at your disposal, is a high voltage power load. I am in the fortunate position to be payed for building one for a project and to keep it afterwards. I built it just today.
160pcs. 10kOhm 7W power resistors.
160pcs. 10kOhm 7W power resistors.
Cheers, Finn Hammer
User avatar
Richard Hull
Moderator
Posts: 14991
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: Criticize my HV supply design

Post by Richard Hull »

Nice 1.6megohm 1120watt load setup. At 35kv that would drag about 22ma from the supply about 720 watts. That load would just be getting hot. No problem. Turn to turn gradient well outside of arc over issues. Probably a lot of inductance in there, but with DC applied no real issue. Good design.

Richard Hull
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
The more complex the idea put forward by the poor amateur, the more likely it will never see embodiment
Silviu Tamasdan
Posts: 147
Joined: Wed Jun 28, 2006 7:17 pm
Real name: Silviu Tamasdan
Location: Connecticut

Re: Criticize my HV supply design

Post by Silviu Tamasdan »

Richard Hull wrote: Fri Nov 10, 2017 1:30 pm You should never spark a switcher system! All manner of deadly pulsed energy can be forced into the solid state stuff. Build a suitable metering system first. Test it on a simple 15kv rectified neon transformer system hooked to a variac to see that it is performing as expected.

Richard Hull
Duly noted. No sparkies. When I get the voltage divider done I'll test it with an oil ignition transformer I have laying around. I think it only goes to 7500V but should be enough.
Finn Hammer wrote: Fri Nov 10, 2017 3:31 pm Silvio,

Impressive study of the transformer, but the size of that transformer core and associated windings seem awfully big.
36cm^2 is the size I would expect from an iron core at 50-60Hz, not a ferrite transformer switched at 20kHz.
Actually one part that I cut out when I posted here was the similar calculation for an 60Hz iron core transformer. It ended up about twice as big.
You will want to make the primary coil out of either Litz wire or flat copper strip.
Litz is expensive especially if you need that amount of current. But flat strip, that's a genial idea. I have lots of flat copper, 0.7mm. I'm not familiar with calculating the current load capacity of flat copper. Or have a defined idea on how to insulate them.
As an example of modern switching supply transformers, Here is a photo of the 2 transformers in a Spellman 30kV/20mA supply, although it switches at 100kHz, the secondaries are smaller than a silver dollar.
Are you aware of the forum at www.4hv.org ?. It might be the right place to document your build.
If you noticed I linked above to a design from 4hv.org
Another thing you want to have at your disposal, is a high voltage power load. I am in the fortunate position to be payed for building one for a project and to keep it afterwards. I built it just today.

IMG_20171110_155852.jpg

Cheers, Finn Hammer
I envy you for that. I have thought about a power load but I'm sure my wife would divorce me if I started _yet_another_project_ on top of everything else I've got going on.

I would be interested in your opinion about the last part of the post where I discuss the insulation of the secondary.
There _is_ madness to my method.
User avatar
Finn Hammer
Posts: 298
Joined: Sat Mar 05, 2016 7:21 am
Real name: Finn Hammer
Contact:

Re: Criticize my HV supply design

Post by Finn Hammer »

Silvio,

Litz for 20kHz operation is really easy to make yourself. As you stated, 1mm wire would be fine, so you just bundle enough 1mm wires together and twist them with your power driver. To get access to all separate wires afterwards is also easy: You heat the end of the bundle with a gas torch, untill red hot. When cold, Fondle the scorched wires to rub the black residue of warnish off.
Finally, dip ends in hydrochloric acid to bring a nice shiny solderable surface forward.
A nice inductor for a 200A buck converter I made.
A nice inductor for a 200A buck converter I made.
With regard to the choice of transformer core, I took a look in Snellings book, just to realize that I could not follw him, so instead, turned to the Magnetics website.
Magnetics is a company that takes the Grand American Approach to marketing their products. This means that they will support you with a comprehensive material to walk you trough the selection process, and even give samples.
I have here in my hand a core that I know for a fact will support up to 11.7kW at 20kHz
11.7kW @ 20kHz ferrite core from Magnetics This core is strip wound primary with mylar insulation.
11.7kW @ 20kHz ferrite core from Magnetics This core is strip wound primary with mylar insulation.
By the time that I aquired it, I asked Magnetics and Phillips here in Europe the same question: "Do you have a core that can transfer 10kW?"
Magnetics sent me samples, Phillips bluntly suggested that I contacted a transformer manufacturer that knew what he was doing.

The latter is an example of the Lousy European Approach.
-Anyway, go here:
https://www.mag-inc.com/Products/Ferrite-Cores

And click on "Power design with ferrite cores" to get this PDF:

https://www.mag-inc.com/getattachment/P ... lang=en-US

And you will be well advised on transformer core selection.

For a 1600W design, you could use this core:
https://www.mag-inc.com/Media/Magnetics ... 6420UC.pdf
And for a whopping 2800W design, use:
https://www.mag-inc.com/Media/Magnetics ... 9316UC.pdf

These are all 20kHz figures, and if you want higher power troughput, up the frequency.

You ask about insulating the secondary, and if you make a multi layer wound coil, you will under all circumstances want to put interleaving between each layer, simply in order to keep the winding pattern tidy. The interleaving material could be Mylar, a polyester material made by Dupont©, and the thickness 0.05mm. Intuition might suggest you take up excess space by using interleaving, but this is not the case, just one overlapping turn will take up much more space.

Here is an example of a 20kV transformer I made 15 years back, for a classic Tesla Coil, and it runs to this day in Museum duty:
trafo.jpg
This transformer has 10000 windings on each secondary coil, I seem to recall, in 50 layers, so interleaving was an absolute must.

But there is an issue that needs adressing first, and that is the self capacitance of the secondary winding. This capacitance can ressonate with the leakage inductance of the transformer, and create a peak that limits the power troughput, and this is the reason that many high voltage, high frequency transformers stop at around 5-7.5 kV. And also the reason for the narrow pie shaped bucket wound coils shown up in the Spellman supply.
This is an issue you will have to investigate, before you get to a proper working secondary coil.
From your previous postings I see that you take a solid engineering approach to things, doing the calculations needed.
I cannot be of much further help here, because I am mathematically challenged, but perhaps Rich Feldman will tune in here?

Cheers, Finn Hammer
Silviu Tamasdan
Posts: 147
Joined: Wed Jun 28, 2006 7:17 pm
Real name: Silviu Tamasdan
Location: Connecticut

Re: Criticize my HV supply design

Post by Silviu Tamasdan »

Thanks for your help! This will take some time to digest and I don't think I will be able to do it this weekend as I'm at work.

I will probably write a more detailed post later tonight, but for now I want to address secondary capacitance. TBH I don't think it will be much of a problem because the spacing between layers and between the turns in the same layer will be substantial. The wire I plan to use has a 0.5mm silicone insulation so the minimum distance between conductors will be 1mm plus whatever extra insulation I will add. I wasn't planning on interleaving the layers but rather overlap them; I don't think the capacitance will add up to much this way. But of course I may be proven wrong in practice.

Mylar is a good idea, don't know how it didn't occur to me as I have used it before (as a side note, do you know of a practical way to weld or glue together 2 sheets of mylar? I have been unsuccessful by using either heat or a range of usual adhesives. I'm talking about putting 2 sheets end-to-end, not stacked)

And making "litz" from ordinary wire is a great hack as well. When you talk about litz the first thing that comes to mind is the true RF stuff made from dozens or hundreds of 40AWG. :)
There _is_ madness to my method.
User avatar
Finn Hammer
Posts: 298
Joined: Sat Mar 05, 2016 7:21 am
Real name: Finn Hammer
Contact:

Re: Criticize my HV supply design

Post by Finn Hammer »

Silviu Tamasdan wrote: Sat Nov 11, 2017 7:34 am Thanks for your help! This will take some time to digest and I don't think I will be able to do it this weekend as I'm at work.
Yes, transformer core selection is a mouthfull. BTW, i cannot seem to find an example of your back of an envelope calculation of the core size, would you mind to publish it. A 2kW core in a 20kHz PSU should fit in the palm of your hand.
Silviu Tamasdan wrote: Sat Nov 11, 2017 7:34 am
I will probably write a more detailed post later tonight, but for now I want to address secondary capacitance. TBH I don't think it will be much of a problem because the spacing between layers and between the turns in the same layer will be substantial. The wire I plan to use has a 0.5mm silicone insulation so the minimum distance between conductors will be 1mm plus whatever extra insulation I will add. I wasn't planning on interleaving the layers but rather overlap them; I don't think the capacitance will add up to much this way. But of course I may be proven wrong in practice.
Perhaps we don't have the same definition of interleaving. My understanding is this: You wind a full layer, then wrap a layer of mylar, to form a new smooth base to wind the new layer on (and add insulation too).
I don't understand what you mean by overlapping the windings
The self capacitance can be calculated, but a better approach is to wind a test coil and measure it's parameters.
Silviu Tamasdan wrote: Sat Nov 11, 2017 7:34 am Mylar is a good idea, don't know how it didn't occur to me as I have used it before (as a side note, do you know of a practical way to weld or glue together 2 sheets of mylar? I have been unsuccessful by using either heat or a range of usual adhesives. I'm talking about putting 2 sheets end-to-end, not stacked)
One of the advantages of mylar is it's voltage standoff properties, due to the integrity of the material. I think you will be better off by buying mylar in a roll instead of sheets, so you can have he lengths you need.

Cheers, Finn Hammer
Silviu Tamasdan
Posts: 147
Joined: Wed Jun 28, 2006 7:17 pm
Real name: Silviu Tamasdan
Location: Connecticut

Re: Criticize my HV supply design

Post by Silviu Tamasdan »

LOL the question about mylar was about a completely different project - one that uses gases but not for fusion but rather for floating things up in the air. :p
There _is_ madness to my method.
Nnnnnnn
Posts: 58
Joined: Tue Jun 13, 2017 6:25 pm
Real name: Niels Geerits

Re: Criticize my HV supply design

Post by Nnnnnnn »

Hi Silviu,

Great post! I might be repeating what others have already said. I also built my own working HVPS using a half bridge, a ferrite toroid and a CW Multiplier (40kV. 10kV @ 50mA is the most I have used for now). Now I am going to change the Inverter to a push pull type circuit with a center tapped Primary. This allows me to use only N-FETs in addition I will be lowering the primary voltage from 250V and 4A to 30V at 20A (these FETs are cheaper). Drain Source ringing on my P-FET (half bridge) was a pain. It is good you are taking more time to think of a smarter way to switch those FETs.

I think your power size estimations might be off. My supply uses a smaller core than what you are proposing. In addition commercial low voltage supplies I own have smaller core sizes (e.g. Computer PSUs, Welder supply). What did you base your estimations on?

Also save yourself some money and start with low voltage and low current so you can use cheap FETs. You don't want a bucket of dead power FETs like me.

Don't forget flyback diodes. I was told the diodes built into mosfets aren't that great for this purpose, but this may be wrong.

To measure the voltage you really don't need any special resistors. I got 100 0.25W 10Mohm resistors and put them in series (this takes time). They don't have to have a high precision, because the high amount of resistors averages out errors. After the 10Mohm resistors I put a single precise 100k ohm resistor. I measure the voltage across this resistor. Cost 15€ IIRC.

One last thing. Now that you know you can probably get away with a smaller core you will probably need more primary windings. Winding the secondary of a toroid is a pain (takes very long). You may want to bump up the frequency a bit to save on windings. (I am at 50khz and 10 primary windings). It is up to you though higher frequencies come with other problems (switching losses, skin effect)
Silviu Tamasdan
Posts: 147
Joined: Wed Jun 28, 2006 7:17 pm
Real name: Silviu Tamasdan
Location: Connecticut

Re: Criticize my HV supply design

Post by Silviu Tamasdan »

Niels,

Thanks for the advice. It's the first time I design a SMPS, so I used some formulas I found that were used in other similar projects. One of them was P=(Ae(cm2)/1.152)^2 which I have seen used in a similar circumstance. I probably need to do more reading. However, I am OK with using a larger core area, it makes much less likely that I will overload it and gives me the potential for in-place upgrades. The cores I'm using for the transformer aren't toroidal but E+E so much easier to wind for. The toroid in the pictures is just for the initial tests. I'll re-calculate the transformer when I have some time to do some serious reading, but I'm still planning to use E-cores and leave a large power buffer in the calculations. No magic smoke should be lost here. In addition it will allow me to not complicate the primary and secondary windings too much.

I already got my HV resistors, that's 3 pieces of 1.5Gohm each, about 17cm long each, rated for 30+kV. I won't push each that far obviously. They cost me about $20 together, so the same ballpark as your setup, and will save me soldering 100 resistors in series. The nice thing about these resistors is that they screw end-to-end into each other so placing them in series doesn't even require any soldering.

Cost of the power MOSFET/IGBT is no concern. I have a source for obtaining them fairly cheaply. My current version of the design uses 600V/60A IGBTs which cost me about $1 a piece. And in fact I have received yesterday a few 1200V/40A IGBTs which had cost me about the same.
There _is_ madness to my method.
Silviu Tamasdan
Posts: 147
Joined: Wed Jun 28, 2006 7:17 pm
Real name: Silviu Tamasdan
Location: Connecticut

Re: Criticize my HV supply design

Post by Silviu Tamasdan »

I had very little time to get stuff done lately, but here are the latest developments.

I've read a lot more about SMPS topologies and practical implementations, had help from some people on 4HV.org, and realized a fundamental error I was making in the calculations above.

Using the information I got now I will be adapting a known working design to my parameters. I will use proper gate drivers for the IGBTs, and a comparator-based overcurrent protection scheme. Since the OCP will use only half of the 4 comparators in a LM339, one of the unused ones will become the oscillator instead of a separate 555. I have increased the design frequency from 20kHz to 40kHz, to get more efficiency out of the ferrite transformer and save me a lot of wire in the secondary. The oscillator will actually generate double that (80kHz) and the signal will be divided by 2 by a T flipflop made with a CD4027 to achieve near-perfect 50% duty cycle. In series with the primary of the transformer I will have an inductor to provide enough reactance to briefly absorb any current spikes (shorts or arcing in the secondary) until the regular overcurrent protection kicks in. The protection inductor will be made on an identical ferrite core and same primary as the main transformer, but without a secondary.I've also reduced the secondary voltage to 9000V to avoid overstressing the CWM components, and increased the CWM stages to 2.

For the transformer itself I will use a U+I ferrite with a core area of 8.4cm2 which should have ample reserve for a design power of 1000W. The primary will be 4 turns of litz (28 strands of 22AWG enameled wire, should support 25A current and the calculated peak current needed is 23.57A; I may make it 30 strands which will support up to 28A). The secondary is 450 turns of high-voltage 28AWG wire with 0.5mm of silicone insulation. This accounts for approx 10% core losses (hysteresis, flux leakage etc). Max secondary current based on wire gauge is 200mA, but will not go over 100mA because that's what my HV diodes are rated for (30kV, 100mA, 100ns) and the primary driver is dimensioned for that. The secondary geometry, based on the core window area available minus space needed for the primary, will be 25 turns per layer, 18 layers total. Layers separated by 2x0.1mm mylar sheets, one above and one below the connections between layers.This will give me a core window utilization of about 92%.

The above followed by a 2-stage CWM should reach 50kV at 18mA.

Below picture of some HV parts I got in the meantime.
Top row, 30kV and 20kV caps for use in the CWM.
Middle row, large E+E cores for a later, higher power version of the SMPS. The core area for these is 16cm2. Next some high voltage resistors (20x2.2Mohm and 3x1.5Gohm) to be used in voltage dividers for measuring. And a roll of 28AWG HV wire.
Bottom row polypropylene caps, the U+I core for the transformer discussed above and caps for filtering the rectified mains voltage for the primary.

For size reference, the length of one of the E cores is 14cm (5.5 inches). Each half weighs almost 3 pounds.
Attachments
My new HV friends
My new HV friends
There _is_ madness to my method.
Rex Allers
Posts: 570
Joined: Sun Dec 30, 2012 3:39 am
Real name:
Location: San Jose CA

Re: Criticize my HV supply design

Post by Rex Allers »

Are those resistors from Russia? I bought some that look like that a good while back from a Russian seller. Mine are all 200 Mohm.

If they are like the ones I got, they have a threaded screw projecting from one end and a threaded hole in the other, so you can screw them together end to end.

If it helps, what I worked out is the threads are metric 4-.70. I wanted several resistors in parallel for a load so made two aluminum bars where I drilled holes in one of them and threaded 4-.70, and drilled slightly bigger holes in the other bar to accept 4-.70 screws.

parallel resistors
parallel resistors
This doesn't help your supply project, but maybe a bit of help in future work.
Rex Allers
Silviu Tamasdan
Posts: 147
Joined: Wed Jun 28, 2006 7:17 pm
Real name: Silviu Tamasdan
Location: Connecticut

Re: Criticize my HV supply design

Post by Silviu Tamasdan »

Yes, they are exactly the same Russian resistors. They do screw into each other and the thread is M4 indeed. I bought them largely due to this feature. It makes it very practical to construct all sorts of dividers easily.

I'm not sure of the power rating for these but the seller states that they are all rated for 35kV. While for the 1.5Gohm that may be true (the dissipated power would be 0.9W), I seriously doubt it for the 2.2Mohm as that would make a resistor dissipate 560W. I however guess a realistic rating would be 20-30W maximum each.

I don't plan to make a power load out of them. For that purpose I have a whole bunch of 100k/10W and 20k/20W resistors on the way.
I plan to put 100 of the 100k resistors in series for a 10M/1000W load, with a midpoint tap (so it can be used as 10M/1000W end-to-end or 5M/500W one-half or with the 2 sections in parallel for 2.5M/1000W) and 100 of the 20k in series for a 2M/2000W load. These would allow me to test the PS with various loads from 5 to 25mA at 50kV.

I actually have 29 of the Russian 2.2M resistors. I guess I could make a power load out of them, if I connected 25 of them in a 5x5 series/parallel grid; if my guess of 30W for each is correct then the load would be 2.2M/750W max, so it could be used at up to 40kV/18mA or so. It wouldn't be useful for testing at 50kV though. Perhaps the power dissipated could be increased by submerging in oil, but I'm not sure how would the lacquer they're covered in resist the mineral oil.
There _is_ madness to my method.
User avatar
Richard Hull
Moderator
Posts: 14991
Joined: Fri Jun 15, 2001 9:44 am
Real name: Richard Hull

Re: Criticize my HV supply design

Post by Richard Hull »

I doubt the resistors are more than 5 watt unless they are ceramic. The question is....Are they composite or spiral wound resist under that coating? I am fairly sure they are not wire wound.

Richard Hull
Progress may have been a good thing once, but it just went on too long. - Yogi Berra
Fusion is the energy of the future....and it always will be
The more complex the idea put forward by the poor amateur, the more likely it will never see embodiment
Silviu Tamasdan
Posts: 147
Joined: Wed Jun 28, 2006 7:17 pm
Real name: Silviu Tamasdan
Location: Connecticut

Re: Criticize my HV supply design

Post by Silviu Tamasdan »

You can't tell how they are made, the red coat is very thick and you can't guess what's under it. You can't peek inside either because both ends are covered with metal caps.

I changed the primary wire to 60x24AWG wire, will support up to 30A and increases the usable frequency to above 60kHz should I decide to try it. I made 90cm of it, picture below. The end loops will be cleaned of enamel (molten aspirin works great for that, hope tomorrow is a warm day because I have to do it outside, aspirin releases very irritating fumes) and soldered to flat 0.7mm copper terminals. I will also place it inside a piece of PVC-based heat-shrink tubing and after the primary is wound that will get shrunk in place.
Attachments
20171222_094321.jpg
There _is_ madness to my method.
Rex Allers
Posts: 570
Joined: Sun Dec 30, 2012 3:39 am
Real name:
Location: San Jose CA

Re: Criticize my HV supply design

Post by Rex Allers »

A little bit more on those Russian resistors.

I forgot to mention that the length on the ones I have is not real consistent. If you look at the thin bar in the picture I posted earlier, it's a little wavy and there is a bit of a gap between the end of one and the bar.

Per Richard's question, I looked at the ones I have and there is a small imprint of the traces visible in the outer coating. I can see that the resistor is a helix pattern. The width of the resistive stripes is about .055" and the spacing is about .115". The length of the resistor in between the end caps is about 5".

They have a nice shiny brown surface that looked to me like ceramic, but looking closely I see small amounts of the brown on the metal end caps, so maybe more likely something like epoxy paint.

I also have some EBG 200 Mohm resistors that I found specs for. They are rated at 11.7 watts. Those are smaller in diameter and shorter. Just roughly I'd estimate 2/3 the size of the Russian ones. So my best guess for the Russian ones would be about 20 W.
Rex Allers
Silviu Tamasdan
Posts: 147
Joined: Wed Jun 28, 2006 7:17 pm
Real name: Silviu Tamasdan
Location: Connecticut

Re: Criticize my HV supply design

Post by Silviu Tamasdan »

I asked the Russian seller about power rating. He replied that the official power rating of these resistors is 5W. So definitely my 2.2M ones wouldn't work at more than 3kV each. Your 200M should be able to take 30kV.

At this point I only plan to use them for a HV divider. 3*1.5G +2*2.2M in series i.e. 4.5G + 4.4M. Not a perfect 1000:1 but close (1024:1). Or if I'm able to select 2 which add up to 4.5M that would give 1001:1
There _is_ madness to my method.
Rex Allers
Posts: 570
Joined: Sun Dec 30, 2012 3:39 am
Real name:
Location: San Jose CA

Re: Criticize my HV supply design

Post by Rex Allers »

Note that the tolerance printed on my resistors is 10% and I measured the 5 200 M in parallel (should = 40 M) at 38 M. So I suggest you have a pot on the low resistance leg of the divider to calibrate the division ratio.
Rex Allers
Silviu Tamasdan
Posts: 147
Joined: Wed Jun 28, 2006 7:17 pm
Real name: Silviu Tamasdan
Location: Connecticut

Re: Criticize my HV supply design

Post by Silviu Tamasdan »

What better thing to do on a Christmas day than sit down and make a present for myself. I made the secondary winding for my HV transformer in 2 sittings today, 2h each with a break in between. 450 turns in 18 layers of 25 turns each, 28AWG wire with 0.5mm silicone insulation; 0.1mm of mylar between layers, doubled mylar above/below the connecting loops between layers.
Attachments
About halfway done.
About halfway done.
Finished all layers, fit test on the core.
Finished all layers, fit test on the core.
And as an icing on the cake, just received some nice BIG power resistors to be used as ballasts for the fusor.
And as an icing on the cake, just received some nice BIG power resistors to be used as ballasts for the fusor.
There _is_ madness to my method.
Silviu Tamasdan
Posts: 147
Joined: Wed Jun 28, 2006 7:17 pm
Real name: Silviu Tamasdan
Location: Connecticut

Re: Criticize my HV supply design

Post by Silviu Tamasdan »

Finn Hammer wrote: Fri Nov 10, 2017 3:31 pm
Another thing you want to have at your disposal, is a high voltage power load. I am in the fortunate position to be payed for building one for a project and to keep it afterwards. I built it just today.

IMG_20171110_155852.jpg

Cheers, Finn Hammer
Quick question. What material do you use for the posts supporting the resistors? I assume some sort of phenolic board. Plastic of any kind wouldn't be thermally stable enough. I'm having some difficulty sourcing it though; been thinking of wood covered with epoxy lacquer.
There _is_ madness to my method.
ian_krase
Posts: 636
Joined: Mon Nov 28, 2016 2:48 am
Real name: Ian Krase

Re: Criticize my HV supply design

Post by ian_krase »

Wood is at least a little conductive, might be problem.
User avatar
Dennis P Brown
Posts: 3159
Joined: Sun May 20, 2012 10:46 am
Real name: Dennis Brown

Re: Criticize my HV supply design

Post by Dennis P Brown »

I use ceramic that are pre-drilled and tapped. These are available from electrical supply sites that handle high voltage. Also, plastic posts (easy to drill and tap for screws) or even plastic bolts work when placed under synthetic oil; then the heat load can be handled. One could use short glass tubes with epoxy to secure the screws in the tube; then place this and the resistor under oil.

Impressive work on assembling the x-former!
Post Reply

Return to “High Voltage - Fusor Input Power (& FAQs)”