Power supply topology for 150kV

[BootstrapBill]

I'm going to be building a 30kW, 150kV supply. I want to be able to control it from 50kV to 150kV. This is for an x-ray tube and current will be controlled by filament current. There will be be a sense resistor to monitor current and trip circuit if their is a spike.

I've mostly seen resonant converters as they work well with multipliers. I like this topology but the control and sense circuitry are complex. I have also seen transformers with their secondaries in series. This topology would be hard to isolate HV from primary.

I had a mag company give me a ferrite core/winding suggestion for a 340 pri/15kV sec. I want to go from a half/full bridge to the step up transformer to a multiplier stage. I would like to use something simple like pwm to control voltage but I can't seem to come up with a topology because of the multiplier after the secondary.

Is there a simple hard switching converter topology that could accomplish my goal?

Thanks

[CM800]

I'm going to be building a 30kW, 150kV supply. I want to be able to control it from 50kV to 150kV. This is for an x-ray tube and current will be controlled by filament current. There will be be a sense resistor to monitor current and trip circuit if their is a spike.

I've mostly seen resonant converters as they work well with multipliers. I like this topology but the control and sense circuitry are complex. I have also seen transformers with their secondaries in series. This topology would be hard to isolate HV from primary.

I had a mag company give me a ferrite core/winding suggestion for a 340 pri/15kV sec. I want to go from a half/full bridge to the step up transformer to a multiplier stage. I would like to use something simple like pwm to control voltage but I can't seem to come up with a topology because of the multiplier after the secondary.

Is there a simple hard switching converter topology that could accomplish my goal?

Thanks

Sounds like you just want a high power full-bridge, resonant may be more efficient. You are really scaling up a small SMPS to large scale, you'd just use brick IGBTs rather then TO-220 and independent gate-drivers, possibly coupled, very likely with protection on them.

http://www.stevehv.4hv.org/ccps1.htm
http://uzzors2k.4hv.org/index.php?page=slrconverter
http://uzzors2k.4hv.org/index.php?page=multiinverter

Naturally these are hobbiest projects, you will want to add a lot of protection & Safety circuits on them, along with reliability etc.

[BootstrapBill]

I already have single brick IGBTS(300A, 1200V). I'm probably going to use 3120 optos to drive gates of mosfets that will then power gate drive transformers (I don't know of anything else that can drive the brick IGBT gates). I'm just not very familiar with control circuits for Resonant converters. All the x-ray supply scholarly articles are series-parallel-series topoligies focusing on efficiency. I like the idea of the fixed frequency rescon. Thanks for the links

[CM800]

I already have single brick IGBTS(300A, 1200V). I'm probably going to use 3120 optos to drive gates of mosfets that will then power gate drive transformers (I don't know of anything else that can drive the brick IGBT gates). I'm just not very familiar with control circuits for Resonant converters. All the x-ray supply scholarly articles are series-parallel-series topoligies focusing on efficiency. I like the idea of the fixed frequency rescon. Thanks for the links

You're going to want to work out what frequency your IGBTs can handle and then design the transformer to operate at a suitable frequency. As a hint, most XRTs run at 20kHz.
I would suggest reading all the application notes you can find, expecially from AVAGO, Fairchild, IXYS. IXYS and AVAGO expecially.

Honestly I'd recommend just buying an x-ray transformer and driver, 50Hz is best, then set up a rectifier with vacuum tubes, the xrt may or may not already have one.

Contact x-ray machine installers, medical equipment servicemen etc. It's best to visit in person, offer beer and PAY BY CASH (off the books, everyone likes tax - free beer money)

[BootstrapBill]

Thanks for advise. I'm a medical equipment serviceman haha. There is no x-ray equipment I know of that delivers that kind of power except CT scanner generators. They are all so involved with the rest of the system to ensure patient dose safety that it would be easier to find the space for a whole CT scanner than to trick the control circuitry into working.

[Lightages]

Please be careful with this. This WILL KILL YOU if you do the wrong thing. I always get afraid when people ask questions online about VERY dangerous things. If you need to ask on a forum how to do something, then it makes me think you have never had experience or had someone show you the correct safety procedures.

[Alex Eisenhut]

This is even beyond the machines they use to X-ray welds and NDT stuff.

http://www.balteau.com/products.php?type=56

Are you trying to create a Therac-25 for the 21st century?

[BootstrapBill]

I think that would be a Probeam.. slightly more voltage required

[T3sl4co1l]

Or a particle collider Marx generator...
 

Tim

[tggzzz]

Please be careful with this. This WILL KILL YOU if you do the wrong thing. I always get afraid when people ask questions online about VERY dangerous things. If you need to ask on a forum how to do something, then it makes me think you have never had experience or had someone show you the correct safety procedures.

Just to amplify that. There are several ways that it could kill/maim you, some instantaneous, some much slower.

[CM800]

Please be careful with this. This WILL KILL YOU if you do the wrong thing. I always get afraid when people ask questions online about VERY dangerous things. If you need to ask on a forum how to do something, then it makes me think you have never had experience or had someone show you the correct safety procedures.

Just to amplify that. There are several ways that it could kill/maim you, some instantaneous, some much slower.

A few points on keeping safe:

-Duel deadmans switch (you find these on robot arms, if you release it goes into emergency stop, it also goes into e-stop if you push too hard) This is important as the currents involved could equally force you to press down very hard on it, or release it.

-Fiber optic / air switch everything. Keep your distance, use a non-conductive method of control, fiber is best IMHO.

-Cover your Powered components, detonating IGBTs can shoot out bits of copper and shards of fiber re-enforced plastic / epoxy, this will happily take your eye out, it will also happily cut into your neck...

-Test / develop with a low voltage transformer. Get your transformer manufacturer to wind you a unit with less turns, say 100V output, you can then dump this into a massive resistor bank or salt pool as a constant load, this will allow you to test and measure much safely.

-Equipment. You NEED differential probes and current probes, if you don't have these, just stop right there.

[mk_]

I'm going to be building a 30kW, 150kV supply. I want to be able to control it from 50kV to 150kV. This is for an x-ray tube and current will be controlled by filament current. There will be be a sense resistor to monitor current and trip circuit if their is a spike.

I had a mag company give me a ferrite core/winding suggestion for a 340 pri/15kV sec. I want to go from a half/full bridge to the step up transformer to a multiplier stage. I would like to use something simple like pwm to control voltage but I can't seem to come up with a topology because of the multiplier after the secondary.

Is there a simple hard switching converter topology that could accomplish my goal?

Thanks

Take a look at phaseshift-Controllers (full bridge on primary side is more or less easy, even if you have to handle more then 100A at full power)....  a well dimensoned inductor on secondary side reduceds current peaks in the Rectifier/Multiplier.


Feedback from the high-Voltage side could be funny, specialy if you need high bandwith for your regulation... (you must do pulse per pulse regulation on your primary side, so oscillation is not far away)

And now something you din`t ask but should be said:

Anyway - the most critical and expensive investment is all that equipment you need for your security (Variable transformer for variable voltage @ primary power in, you will not power up your first testsystem with full voltage) including some emerengy shutdownbuttons, a partner in the lab who knows how to react if something blows up (never, never, never be alone in the lab when running with power), cooling-equipment, a load were you can dump your 150kV@30kW.... Protectionglasses for your eyes which don`t break if something blows violently up, ear protection you can wear a long time comfortly (it is really loud if something goes booom)

Not to forget a scope with a lot of channels (4 are not very much for such a task when studying feedbackloops or Vce_sat on an IGBT ), reliable Currentprobes, Differential probes for low voltage (you can`t measure low voltage stuff if 100A and 500V rising within 50ns from 0 to 500V are roaring some 3cm away) and also some very high voltage differential  probes with specified voltage at required bandwith for secondary side.

But beside that: it is great fun and I like my work doing this on a regular base...

btw: at the eth zürich you can find a lot of great papers about that stuff... PM if urls required

Good luck

mk_

[BootstrapBill]

I really do appreciate all the safety recommendations.. can never be too safe. To make you feel a little better I have been working with high voltage equipment for over 10 years and know the industry safety standards.

I have a 4kW 60kV generator already in use. The room is lead shielded and x-ray is wirelessly controlled.

I had planned on starting out low rail voltage, a step down transformer, and a resistive load. I would then work up to adding a few multiplier stages, then upping the turns ratio, etc.

This is also going to be properly built. It will appear to be a commercial product. Everything will be properly isolated and in dielectric oil. 

[Fungus]

Please be careful with this. This WILL KILL YOU if you do the wrong thing. I always get afraid when people ask questions online about VERY dangerous things. If you need to ask on a forum how to do something, then it makes me think you have never had experience or had someone show you the correct safety procedures.

Just to amplify that. There are several ways that it could kill/maim you, some instantaneous, some much slower.

And that's just the voltage. Never mind the X-rays. 

A few points on keeping safe:
-Duel deadmans switch
-Fiber optic / air switch everything.
-Cover your Powered components, detonating IGBTs cank...
-Test / develop with a low voltage transformer.
-Equipment. You NEED differential probes and current probes

Most important: Record everything on hi-speed video for the coroners report and/or amusement of electrical engineers everywhere. Pay particular attention to lighting and microphone placement. Multiple camera angles would be appreciated.

[BootstrapBill]

Please be careful with this. This WILL KILL YOU if you do the wrong thing. I always get afraid when people ask questions online about VERY dangerous things. If you need to ask on a forum how to do something, then it makes me think you have never had experience or had someone show you the correct safety procedures.

Just to amplify that. There are several ways that it could kill/maim you, some instantaneous, some much slower.

And that's just the voltage. Never mind the X-rays. 

A few points on keeping safe:
-Duel deadmans switch
-Fiber optic / air switch everything.
-Cover your Powered components, detonating IGBTs cank...
-Test / develop with a low voltage transformer.
-Equipment. You NEED differential probes and current probes

Most important: Record everything on hi-speed video for the coroners report and/or amusement of electrical engineers everywhere. Pay particular attention to lighting and microphone placement. Multiple camera angles would be appreciated.

Will do. Video won't be too exciting.. there won't be anyone in the room.

[jbb]

Please be careful of your capacitor banks.  Someone died in my last workplace because of a still-charged capacitor.

I don't know a lot about high voltage (in the sense of > 30kV) supplies, but a lot of the papers, literature, marketing material and teardown videos I have seen tend to use a transformer plus voltage multiplier.  So they're probably right.  (Especially as this will reduce the insulation requirements of the transformer.)   Considering something like a classical Cockroft-Walton multiplier, it's clear that a classic constant frequency variable duty cycle control isn't going to work well.  You probably want to drive that with a resonant converter in the interests of achieving a high switching frequency (smaller equipment) without spewing EMI all over the place.  (Although if the HV side is in a metallic oil enclosure you'd probably be OK.)

However, achieving a 3:1 adjustment range of output voltage is asking a lot of a resonant design (also think about what beam current range you'll need).  I therefore suggest you deploy a variable-ratio DC/DC buck converter to feed power into a fixed-ratio resonant stage.

Also, you should think carefully about which end of the supply you're going to float as this will impact your entire insulation design.  I assume you also know where to look for the relevant oil-compatible components.

[Zero999]

What duty cycle do you need?

Producing 30kW pulses at 150kV is not that difficult but doing it continuously is.

Where are you getting the power from? Just a big transformer connected to 3 phase mains is probably the easiest solution. The AC output can be converted to DC with an old fashioned rectifier. To minimise ripple and harmonics, use a 12  or even an 18 pulse rectifier.

[Marco]

Why resonant for something boutique nowadays with stuff like TPH3207WS out there? Not even that expensive.

[varghese]

30KW is huge ..  I have handled few watts but upto 120KV ,  Ussual procedure is convert it to dc & then make it to PWM pulses with frequncy around 20kHz - 30KHz (will give you good control & reduce the size of transformer/capacitor ).. Then step up the volatge to 10KV /20KV using transformer .. then introduce it to 10 to 15 stage CW multiplier .. For  Testing Have a good HV probe ..   Have a good earthing connection ..  Hold a ground wire every time you come close to the setup ..  Always ground the setup before touching it .. Use Transformer oil & High voltage Epoxys for insulation/ heat ttransmission..  For 150KV , I think you need keep atleast 1.5 meter distance (incase you are not holding earth wire ).. No idea on X ray side

[BootstrapBill]

Duty cycle- I need 2 sec bursts with 5 seconds between or continuously for 60 sec at 10-15kW

Not sure on the power but it will be rectified and and be going to a huge cap with soft start.

[CM800]

Duty cycle- I need 2 sec bursts with 5 seconds between or continuously for 60 sec at 10-15kW

Not sure on the power but it will be rectified and and be going to a huge cap with soft start.

Starting to sound like you're making a death-ray here.

[Jay_Diddy_B]

Duty cycle- I need 2 sec bursts with 5 seconds between or continuously for 60 sec at 10-15kW

Not sure on the power but it will be rectified and be going to a huge cap with soft start.

What kind of X-ray tube are you planning on using? A typical medical X-ray tube is limited to around 1kW dissipation.

Here is a link to a reasonable X-tube:

https://www.varian.com/media/107301/browser

It has a maximum anode dissipation of 1.5kW, but this may be limited by the oil-filled housing. a B130 Housing is limited to about 250W average without a fan and 500W with a fan.

Link: https://www.varian.com/sites/default/files/resource_attachments/B-130HPDS.pdf


Tube are sized by the anode diameter. This is a 4" tube, there also 3" tubes and the odd 5" heavy tube.

Regards,

Jay_Diddy_B

[CM800]

Duty cycle- I need 2 sec bursts with 5 seconds between or continuously for 60 sec at 10-15kW

Not sure on the power but it will be rectified and be going to a huge cap with soft start.

What kind of X-ray tube are you planning on using? A typical medical X-ray tube is limited to around 1kW dissipation.

Here is a link to a reasonable X-tube:

https://www.varian.com/media/107301/browser

It has a maximum anode dissipation of 1.5kW, but this may be limited by the oil-filled housing. a B130 Housing is limited to about 250W average without a fan and 500W with a fan.

Link: https://www.varian.com/sites/default/files/resource_attachments/B-130HPDS.pdf


Tube are sized by the anode diameter. This is a 4" tube, there also 3" tubes and the odd 5" heavy tube.

Regards,

Jay_Diddy_B

I'm pretty curious about this too, in even the smaller machines, tubes have spinning anodes as the focal point of the anode can reach in excess of 2500*C, Is it going to be made of unobtanium?

[BootstrapBill]

http://www.dunlee.com/page/reevo240

Yeah its the size of a small engine.. that's just the tube in a dielectric housing. 100kW for this guy.. ever hear of a CT scanner?

[MagicSmoker]

This is the coolest project I've seen in awhile... then again, I don't get out much.

So, a CT scanner x-ray source that needs 100kW at 120kV. I don't have any direct experience with this sort of thing, but my understanding is that x-ray tubes are fairly benign loads, with the caveat that when used in a CT scanner they need pretty good regulation of their anode voltage and current to prevent density artifacts in the resulting image slices. That argues against using a 5+ stage multiplier, as regulation tends to decline with every stage and you need close to 1A of current (Cockcroft-Walton multipliers are usually used for less than 10mA of current). Thus, I would use appropriate construction techniques to allow for a 30-40kV secondary so that only a 3-4 stage multiplier is needed. This page has some useful information on selecting/sizing a multiplier: http://www.voltagemultipliers.com/html/multdesign.html.

One other thing to consider is that you really don't want a choke on the secondary side, as it will need to be designed to withstand the full secondary voltage as well, so a good topology to consider that isn't too hard to get working is a PWM'ed buck converter feeding a full bridge with the bridge switches running at a fixed 50% duty cycle. The late Abraham Pressman covers this topology at length in his book on switching power supply design.

[BootstrapBill]

This is the coolest project I've seen in awhile... then again, I don't get out much.

So, a CT scanner x-ray source that needs 100kW at 120kV. I don't have any direct experience with this sort of thing, but my understanding is that x-ray tubes are fairly benign loads, with the caveat that when used in a CT scanner they need pretty good regulation of their anode voltage and current to prevent density artifacts in the resulting image slices. That argues against using a 5+ stage multiplier, as regulation tends to decline with every stage and you need close to 1A of current (Cockcroft-Walton multipliers are usually used for less than 10mA of current). Thus, I would use appropriate construction techniques to allow for a 30-40kV secondary so that only a 3-4 stage multiplier is needed. This page has some useful information on selecting/sizing a multiplier: http://www.voltagemultipliers.com/html/multdesign.html.

One other thing to consider is that you really don't want a choke on the secondary side, as it will need to be designed to withstand the full secondary voltage as well, so a good topology to consider that isn't too hard to get working is a PWM'ed buck converter feeding a full bridge with the bridge switches running at a fixed 50% duty cycle. The late Abraham Pressman covers this topology at length in his book on switching power supply design.

I was thinking exactly what you are thinking with the buck feeding the full bridge. The only thing of concern would not being using an inductor on either side of the HV transformer. I've never seen just a bridge and a transformer. Can this be done at this power level? High frequency noise? It would have a lot of losses but I would not need to worry about the voltage control with varying load like with a resonant design.

I'm tempted to use several transformers with secondaries in series as some x-ray heads do. This would make isolation more reasonable.

[IanB]

Thinking crazy thoughts here, but wouldn't it be easier just to take a direct feed from the power company at their distribution voltage (~ 10 kV?), step it up and rectify it? For the different voltages you could use different output taps on the transformer.

Yeah, OK, it's gonna be wildly expensive, but so surely is building this thing from the ground up...

I know, not really practical 

[Marco]

I'm tempted to use several transformers with secondaries in series as some x-ray heads do. This would make isolation more reasonable.

How does that make life easier for the transformer maker compared to a segmented secondary?

The MOSFETs I linked simply allow you to care a little less about switching losses, fast and very low reverse recovery, so if you really wanted to use hard switching you could. Although in retrospect I guess even with those switches you wouldn't want to at that power level. Still, they just switch just a tiny bit faster than your IGBT bricks and you don't have to worry about reverse recovery or dV/dt so much.

[BootstrapBill]

Segmented secondaries with doublers on each winding is how some x ray generators work. I haven't seen an off-the-shelf core that can handle this power.

[LaserSteve]

Former CT instructor here. CT tubes get babied. I would not call them benign.

The cathode to anode distance is very short, making them vulnerable to small amounts of gas being liberated from the anode and tube wall. Thus arc diagnostics need to  decide in a few milliseconds either to keep scanning or abort based on the severity and duration of the arc.

What I worked on had a scan electrode as well,  which gave the system a beam shift on the anode for interpolation.

The power supplies were beefed up much more then you might expect, for "conditioning" the tube prior to use on a PT. CONDITIONING was a 10 to 20 minute sequence to getter any outgassing, evenly expand the rotating anode, prevent thermal shock from cracking the vacuum seals, and store a known amount of heat  units  in the anode. A rather sophisticated HU counter kept track of all this. If PT treatment did not occur soon enough, a cooldown was mandated, followed by more CONDITIONING as needed.

It's been a while, but the PSU module was two sixty plus KV units with independent current and voltage regulation for 120 KV differential, and 850  Ma total. All in an amazingly small box powered by 440/3PH on the sliprings. Split supplies allowed for some tube management  tricks that I was never briefed on.

Said supply could adjust voltage on the order of 500-1000 times per revolution around the PT. This was used to reduce PT exposure by "creative" modulation patterns.

Full voltage full current[~~450 mA was only for warmups and emergency imaging of certain kinds of accident trauma when the radiologist decided saving an adult  life outweighed the consequences in terms of probability of tissue damage.

There was a set of conditions where you could skip CONDITIONING, if total HU to be used was short and use was immediate.
 

Steve

[BootstrapBill]

Nice to hear from another FSE. They pretty much are the same these days. Some use grounded-anode single supplies. It couldn't have been that long ago if adaptive dose modulation was in play.. They spend a lot of effort on dose modulation. They should have gone with the math guys.. the new iterative reconstruction algorithm that is replacing single back projection has cut dose in half.

[jbb]

Hi all

On 11kV utility feed: just no.  You'd then be exposed to prospective faults on the order of 11kV * 10,000A = approx 200MVA.  You don't want that. Huge expensive switchgear, and you'd need a special license just to turn it on or off.

On available cores: it's easy to just stack multiple coils next to each other to increase the core area (but not the winding area).  Leave a small gap (e.g. a sheet of mica paper) between each cores to allow some wiggle room as large cores can have poor mechanical tolerances.

On using multiple transformers:

• Reduces secondary output voltage, and therefore turns ratio +
• Allows use of more, lower-voltage, rectifier stacks.  Might also avoid the direct series connection of diodes ++
• You need primary-secondary isolation that can handle the full stand-off voltage --
• If you already have oil in the system somewhere, oil-insulated transformers are well understood and give good lifetime.
• High primary-secondary isolation means air/oil/whatever gaps.  This leads to leakage inductance which some topologies can't handle.

Like I said, you'll need to work out your insulation scheme.  Will you run either the anode or cathode at/near ground?  Grounded cathode makes heater control easier.  Grounded anode makes ... err ... anode-stuff easier?  Or you could run a split-rail system, with ground in the middle (easier transformer isolation).

[BootstrapBill]

Thanks for quantifying some of the things I was thinking in regards to the multiple transformers. My high voltage "tank" as they say in the industry will definitely be an oil bath. The filament current (heater) may be a thread in itself haha.. it would be floating on that -120kv

[jbb]

It's probably good that you've already crossed the oil Rubicon.  Once you've accepted the (small!) risk of an oil spill / fire for the "tank," adding a bit more oil probably isn't going to violate customer specs.  Oil cooling could also help with your peaking requirements (it adds specific heat to your system).

I'm not sure what your physical layout will be, but even if the transformer and "tank" (rectifiers, capacitors, bleed resistors, sense resistors??) are right next to each other, I suggest you take the trouble to separate their oil baths with a bulkhead & proper bushings.  That way when you cook something the damage will be limited :-)

Sounds like your cathode end will be on potential.  Isolating the filament to -120kV will be a challenge.  It is definitely possible to do so.  You might want to look at a classic AC transformer connected directly to the filament, with filament measurements (V, I etc.) on the primary side; that approach should be quite reliable.  This assumes that you don't need super-accurate sensing, or true DC excitation.

[MagicSmoker]

I was thinking exactly what you are thinking with the buck feeding the full bridge. The only thing of concern would not being using an inductor on either side of the HV transformer.
...
I'm tempted to use several transformers with secondaries in series as some x-ray heads do. This would make isolation more reasonable.

The inductor required by the buck converter takes over the (current) averaging function of the secondary side inductor in a conventional forward-type converter (of which the full bridge is one variant). You can also delete the buck output capacitor (aka - bridge input capacitor) if you run the bridge legs with a slight overlap; this makes the bridge current-fed, instead of voltage fed, and has some significant benefits as far as switching losses and robustness to short circuits. Some overlap of the bridge leg conduction time is critical for the current-fed variation because you don't want to interrupt current through an inductor, but since the bridge isn't width-modulated you can usually rely on the IGBTs turn-off delay to ensure a bit of overlap when driven at 50% duty cycle.

As for secondaries in series, yes, that is one of the many "appropriate construction techniques" I was alluding to (along with vacuum varnish impregnation, quad insulated wire, etc.). Each secondary only needs to withstand the voltage difference across it, then. And as others have mentioned, you won't be able to process this amount of power with a single core, anyway. Which reminds me, you probably want to use UU cores to achieve maximum separation of the windings (at the expense of increased leakage inductance).

[Marco]

vacuum varnish impregnation

For an oil submerged transformer?

[BootstrapBill]

Thanks for the Abraham Pressman recommendation.  I just looked over the 3rd edition- looks like it's all there.

So for this transformer- I would ideally like to build it myself under the supervision of a consultant for learning/cost purposes but understand most companies would want to do the building themselves. It's funny you can contact Mag-inc and tell them what you need and they give you core/turn/gauge. I'm pretty sure it's just a goon plugging a few numbers into an online calculator. They won't let you talk to the engineer on the phone so I will never know.

Anyone have any transformer specialist to recommend?

[Marco]

Here's a transformer in the right ballpark. Indeed a remarkably small box and only 23 kg ... good luck getting near that.

[MagicSmoker]

vacuum varnish impregnation

For an oil submerged transformer?

No, vacuum or vacuum/pressure varnish impregnation are alternatives to oil immersion, and possibly superior for high frequency/high voltage transformers. The OP is encouraged to do his own research into the matter.

[BootstrapBill]

So I am looking for a good buck converter IC. Haven't been able to find any info on converters in the kW range. I'm looking to go voltage controlled. I suppose voltage control for my adjustable "rail voltage" could be done using a voltage divider from output with trim pot to send feedback to error amplifier in the pwm chip? Would a TL494 be ok or is there a more modern go-to these days?

[Vtile]

What about 12500 lead acid batteries in series to get on the right voltage.

Jokes aside, cool project.

PS. Don't blow up your siliscope.

[BrianHG]

So I am looking for a good buck converter IC. Haven't been able to find any info on converters in the kW range. I'm looking to go voltage controlled. I suppose voltage control for my adjustable "rail voltage" could be done using a voltage divider from output with trim pot to send feedback to error amplifier in the pwm chip? Would a TL494 be ok or is there a more modern go-to these days?

I know MCUs might not be your territory, but I would probably use a dsPIC with it's PWM output and flash ADC inputs.  You would be able to software control frequency, using 2 x PWM channels, one for the high side and another for the low to software control your gate drive overlap & with the multiple ACD inputs, you can both resistor divide input and monitor you transformer drive & monitor current consumption for diagnostic & emergency shutdown.

The last project my friend did was a 50kva 600v, 3 phase BLDC PWM driven motor designed to test snow-mobile transmission units, torture-testing them to their breaking point to make sure the mechanics don't explode, with the IGBT driven by one dsPIC, with monitor, control and torque readouts through the RS-232 port.

[jbb]

I agree with BrianHG that using a microcontroller is a good step.

I know MCUs might not be your territory, but I would probably use a dsPIC with it's PWM output and flash ADC inputs.  You would be able to software control frequency, using 2 x PWM channels, one for the high side and another for the low to software control your gate drive overlap & with the multiple ACD inputs, you can both resistor divide input and monitor you transformer drive & monitor current consumption for diagnostic & emergency shutdown.

Let's look at the functions you'll need:

• Incoming supply monitoring
• Soft charge control & monitoring
• "Adjustable rail" control
• System monitoring (e.g. temperature)
• Communications with interlocks / logging systems.

For your reference, the Magna-Power brand is generally considered quite robust and capable of handling challenging loads.  They use a current-fed approach http://www.magna-power.com/support/technical-notes/overview-current-fed-power-processing.  If you want to use a topology like this, the buck stage and H bridge stage need to be PWM'd synchronously.  It's much easier to make this happen with a suitable micro.

There are many options for the micro.  Don't stress about the cost - it will be dwarfed by the power components.  Here are some considerations:

• High clock frequency (>10MHz) to yield fine PWM resolution.
• Good PWM peripherals - you should be able to synchronise many PWM units to run in lock-step.
• Good ADC - you should be able to a) set up a list of channels to sample in hardware and b) trigger this sample from the PWM units.  Minimum 12b resolution.
• Lots of memory.  I suggest min 64kB program and 16kB RAM.
• >= 16 bit core.  8 bit micros will require more programming effort to make your arythmetic happen.
• Floating point is not required. But you can use it if you want.  If using fixed-point, the general suggestion is to do everything in per-unit scaling.

Some possbilibites:

• dsPIC
• Texas Instruments C2000
• ARM Cortex M
• Maybe MSP430

On switching frequency & materials: if using IGBTs, you'll probably end up in the 1 - 10kHz range and may produce a lot of accoustic noise.  Amorphous/nanocrystalline iron cores are particularly susceptible to magnetostriction and have been known to make unacceptably loud products.

On connecting the MCU to the power stage:

• Isolate the controls from your PC! You will blow the converter up at least once and don't want to send your PC with it.
• I suggest you isolate the controls from the power stage as well.  This will reduce the impact of electrical noise and improve safety.
• At least in the initial development phase, I suggest you do hardware (opamps, comparators, flip-flop) protection of the hardware stage as well as software protection.  This  means that your converter doesn't blow when your software hangs.
• LEM current sensors are very well regarded. http://www.lem.com/ They provide isolation for free.
• LEM voltage sensors are not great (actually current sensor + shut resistor) but do you really want to mess about?
• Typically voltage sensors consist of a resistor divider + isolating amplifier + isolated PSU
• Use isolated gate drivers.  You can usually just but an industrial module (but they're expensive!) is you don't want to have trouble.  Rolling your own for such a high power job will be hard if you haven't done it before.
• Use gate drivers with desaturation protection.  This is a last-ditch short-circuit protection method and it will help.

[T3sl4co1l]

If the OP is struggling to complete a design with such an antiquated and basic controller as a TL494, I don't see MCUs being any help here.

The OP needs professional help.  And I mean that in the best way possible, of course -- he should hire an experienced designer, before he begins work on something powerful enough to be quite dangerous!

Tim

[BrianHG]

If the OP is struggling to complete a design with such an antiquated and basic controller as a TL494, I don't see MCUs being any help here.

The OP needs professional help.  And I mean that in the best way possible, of course -- he should hire an experienced designer, before he begins work on something powerful enough to be quite dangerous!

Tim

Yes, I agree.  My friend who worked on the snow-mobile transmission test jig is a serious electrical and Newtonian engineer who did all the math / research beforehand with experience of many years in motor control and robotics, working his way up in power through the years and with the power factor I mentioned above, the power source could have easily killed through both the voltages involved, or created a deadly explosion as I saw all the IGBTs and all the other high power components were in a thick steel explosion proof housing with gigantic fuses.

[razberik]

What about 12500 lead acid batteries in series to get on the right voltage.
Jokes aside, cool project.

Actually not a joke. My colleague told me about some ionizing detector for SEM microscope they bought 25-30yrs ago from some Japanese company. They supplied the +1kV potential and floating electronics by a lot of batteries in series. It was a few plastic tubes filled with chains of batteries. Replacement of batteries was "service personal only", so these Jap guys had to travel around the world.
This colleague got actually inspired for another type of detector. Kilovolt supply was SMPS and floating electronics was supplied by NiMH battery. Communication through some Avago optocouplers available that time. Floating electronic had ON/OFF mechanical switch to save battery life.
Never got to serial production.

[BootstrapBill]

I actually like the MCU approach and thank you for the links/suggestions. I was planning on controlling soft start, contactors, interlocks etc with the arduino since I'm familiar with it and use it regularly. I starting thinking I would go the analog chip route because the arduino only does 490Hz pwm. I did not explore more professional MCU options.

Regarding switching frequency I am going for 40kHz. Reason is to be out of the audio range and make it more reasonable for my transformer designer.

I have hired professional help for the magnetics and multiplier and may be taking the next step soon and hire a designer. Problem is no one wants to get involved with 100+kV. Anyway after consulting with a company that does large mulitipliers we decided on a 10kV dry transformer to feed the multiplier. They claim it should be stable.

[james_s]

What about 12500 lead acid batteries in series to get on the right voltage.
Jokes aside, cool project.

Actually not a joke. My colleague told me about some ionizing detector for SEM microscope they bought 25-30yrs ago from some Japanese company. They supplied the +1kV potential and floating electronics by a lot of batteries in series. It was a few plastic tubes filled with chains of batteries. Replacement of batteries was "service personal only", so these Jap guys had to travel around the world.
This colleague got actually inspired for another type of detector. Kilovolt supply was SMPS and floating electronics was supplied by NiMH battery. Communication through some Avago optocouplers available that time. Floating electronic had ON/OFF mechanical switch to save battery life.
Never got to serial production.

Several years ago I was given a box of roughly 400 lightly used 9V batteries that came from the required annual replacement in the primarily line powered smoke alarms in a retirement home. I was real tempted to chain up a couple hundred of them in series but after seeing the arc I could draw from 15 of them I was always too nervous. They can deliver a few Amps into a short circuit for a brief period.

[oldway]

http://www.fug-elektronik.de/en/products/high-voltage/hch.html
HCH 50000-150000    150 kV    300 mA    50 kW    600 mm    2000 mm    800 mm    930 kg

https://www.alibaba.com/product-detail/50kv-100kv-150kv-200kv-high-voltage_60398045736.html

https://www.google.com/patents/US6967559

[BootstrapBill]

Picked up an arduino DUE which has the arm cortex M3.

Much to the dismay of T3sl4co1l I now have 10 bit variable duty PWM at 40kHz   

int Feedback = A0;   // feedback connected to pin 0
int val = 0;         // variable to store the read value

void setup() {
  // PWM Set-up on pin: DAC1
   
  REG_PMC_PCER1 |= PMC_PCER1_PID36;                     // Enable PWM
  REG_PIOB_ABSR |= PIO_ABSR_P16;                        // Set PWM pin perhipheral type A or B, in this case B
  REG_PIOB_PDR |= PIO_PDR_P16;                          // Set PWM pin to an output
  REG_PWM_CLK = PWM_CLK_PREA(0) | PWM_CLK_DIVA(1);      // Set the PWM clock rate to 84MHz (84MHz/1)
  REG_PWM_CMR0 = PWM_CMR_CPRE_CLKA;                     // Enable single slope PWM and set the clock source as CLKA
  REG_PWM_CPRD0 = 2100;                                  // Set the PWM frequency 84MHz/40kHz = 2100                                   // Set the PWM duty cycle 50% (2100/2=1050)
  REG_PWM_ENA = PWM_ENA_CHID0;                          // Enable the PWM channel     
}

void loop()
{
val = analogRead(Feedback);
REG_PWM_CDTY0 = val * 2;
}

Just need to dig deeper into the atmel ADC so I can get past the arduino IDE limited 10bit and get the 11bit to match my output.

I figure 11 bit should be good enough?

[MagicSmoker]

...
Just need to dig deeper into the atmel ADC so I can get past the arduino IDE limited 10bit and get the 11bit to match my output.

I figure 11 bit should be good enough?

10b = 1024 steps, or <150V per step if you want to adjust the output from 0-150kV; over a more restricted range of 50-150kV that's <100V per step. Generally speaking, a step size of 500V would not be unusual for this high a voltage, so I'd say you have more than enough granularity with 10b.

But, you should know what your x-ray tube needs better than us. Or me, anyway.

[james_s]

Generally a step size of 5kVp is adequate for most xray imaging, and 150V would be ridiculously fine. I'm sure there are specific applications where greater precision may be needed though.

[BootstrapBill]

The inductor required by the buck converter takes over the (current) averaging function of the secondary side inductor in a conventional forward-type converter (of which the full bridge is one variant). You can also delete the buck output capacitor (aka - bridge input capacitor) if you run the bridge legs with a slight overlap; this makes the bridge current-fed, instead of voltage fed, and has some significant benefits as far as switching losses and robustness to short circuits. Some overlap of the bridge leg conduction time is critical for the current-fed variation because you don't want to interrupt current through an inductor, but since the bridge isn't width-modulated you can usually rely on the IGBTs turn-off delay to ensure a bit of overlap when driven at 50% duty cycle.

As for secondaries in series, yes, that is one of the many "appropriate construction techniques" I was alluding to (along with vacuum varnish impregnation, quad insulated wire, etc.). Each secondary only needs to withstand the voltage difference across it, then. And as others have mentioned, you won't be able to process this amount of power with a single core, anyway. Which reminds me, you probably want to use UU cores to achieve maximum separation of the windings (at the expense of increased leakage inductance).
[/quote]

I am at the point where I am actually going to construct the Buck portion. I have access to many caps of both the ceramic and electrolytic variety. I was going to have an output capacitor stage so I can test one stage at a time. I am aware that I will need a many caps in parallel to withstand the current ripple of the inductor.

Would there be any negative effects of leaving this cap stage in there?

Would there be any positive effects?

A 10+ stage multiplier on the other side of the transformer isn't exactly a large value output cap..

And most importantly can you be more specific with the overlap?

[MagicSmoker]

I am at the point where I am actually going to construct the Buck portion.
...
Would there be any negative effects of leaving this cap stage in there?

Would there be any positive effects?

In this particular application the main downside of leaving the buck-output/bridge-input capacitor in is the cost/volume of such, though it won't be anywhere near as costly/big as the input capacitor for a width-modulated full bridge (because each bridge leg will be operated very close to 50% duty so it will draw a nearly constant current from the capacitor). Otherwise, there are two other major downsides to the buck voltage-fed bridge, of which only one is relevant here: 1) no intrinsic tolerance of a short on the secondary (irrelevant because the C-W multiplier presents a high impedance load even with its output shorted);  much higher turn-on loss in the bridge switches due to bridge input capacitor ensuring voltage remains nearly constant across the switches during the entire turn-on transition, exacerbated by reverse recovery of the anti-parallel diodes because during the dead-time some or all of the primary current will have transitioned to them.

A 10+ stage multiplier on the other side of the transformer isn't exactly a large value output cap..

And most importantly can you be more specific with the overlap?

Hmmm... a 10 stage C-W multiplier is going to have terrible voltage sag and very high ripple under load. Dave, our fearless leader here, did a very good video on C-W multipliers and I encourage you to pay particular attention to the last few minutes where he tests a 4 stage (x8) multiplier under load:

(I guess youtube embedding tags don't work here?) [edit - I guess it does; didn't show up as embedded when I previewed my post...]

Remember, there's no such thing as a free lunch so if you want, say, 150kV at 1A from a 10 stage multiplier you will need to theoretically feed 20A at 7.5kV into the first stage (realistically you will have to feed a lot more that 7.5kV because of voltage sag). You can reduce voltage sag by making the capacitors in the first stage n times bigger than the capacitors in the final stage, where n is the number of stages (edit: and making the capacitors in the 2nd stage n-1 times bigger, etc.), and by using a center-tapped secondary and mirroring a second C-W multiplier on the other leg, which turns this into a full-wave multiplier with the center-tap as the negative output reference. Actually, I should emphasize this much more clearly: you pretty much have to use a full wave multiplier at this power level.

The overlap time for the bridge switches in the current-fed version is not too critical. A somewhat fast-n-loose approach which nevertheless tends to work well with IGBTs (not MOSFETs!) is to simply drive each bridge leg with a straight 50% duty complementary square wave, relying on the turn-off delay time of the IGBTs to ensure there is some overlap. There is a dedicated PWM controller IC for buck voltage- or current-fed push-pull converters, the LM5041, which lets you select between overlap or dead-time of the push-pull switches; it is very easy to adapt this IC to a full-bridge converter.


edits - additional info; removed erroneous reference to poor transformer utilization (due to my misinterpretation of the half-wave ripple from a C-W multiplier automatically meaning it was a half-wave rectifier, though it is actually full-wave).

[T3sl4co1l]

FYI, the only thing that's half-wave about a CWM is the output ripple; the input is always fully utilized.

7.5kV at 20A sounds quite nice for winding the transformer: it's only 375 ohms, so it will be quite reasonable to achieve low leakage inductance.

Tim

[james_s]

Some years ago I reverse engineered a Kodak dental xray head and it used a ferrite transformer with a C-W multiplier and closed loop feedback for the HV. It was only about 600W but the principal should be scalable. https://www.repairfaq.org/sam/xraysys.htm#xraytro

[MagicSmoker]

FYI, the only thing that's half-wave about a CWM is the output ripple; the input is always fully utilized.

Huh, you're right. I had to simulate it in LTSpice before I believed you because no matter how I look at it I can't see how it wouldn't act as a half-wave rectifier and, indeed, the half-wave ripple only corroborated that analysis. So, learn something new every day...

[T3sl4co1l]

It's like that thing Jesus said, but opposite.  In this case, if you take it on faith, that's fine, but actually testing it to prove it's true, that way lies the path to righteousness.

Tim

[BootstrapBill]

Dude at VMI told me to stay away from full wave. He said it would be too hard to control. I didn't ask but now I'm wondering what he meant by that. Also told me only way to get output reference was to string a voltage divider from output to ground.

I'm basically going to be spending a lot on HV capacitors to maintain smooth output. They get quite expensive outside of the nF range

[MagicSmoker]

Generally a step size of 5kVp is adequate for most xray imaging, and 150V would be ridiculously fine. I'm sure there are specific applications where greater precision may be needed though.

I didn't explicitly state this in my previous post - and then got sidetracked by Tim with the C-W stuff - but the main reason for wanting the ADC monitoring the output voltage to have a finer step size than is strictly necessary is to avoid/minimize limit cycle oscillations, a phenomenon in sampled data systems where the output value oscillates between two steps because the reference value has been set to something between them. This applies to the digital pulse width modulator (dPWM) as well - it needs to have an even finer step size than the ADC.

You can employ some fairly heavy math to determine just how much finer the ADC and dPWM step sizes need to be, but I tend to categorize such as "getting a really precise answer to a hopelessly imprecise question"; ie - of dubious utility. In my experience, simply going with +2b higher resolution for the ADC and another +1b for the dPWM is sufficient. E.g. if there are 256 (8b) output voltage steps then a 10b ADC and an 11b dPWM will usually suffice.

Dude at VMI told me to stay away from full wave. He said it would be too hard to control. I didn't ask but now I'm wondering what he meant by that. Also told me only way to get output reference was to string a voltage divider from output to ground.

I'm basically going to be spending a lot on HV capacitors to maintain smooth output. They get quite expensive outside of the nF range

Yeah, I'd ask for clarification on that full-wave comment; it doesn't make any sense as-is.

There are non-contact means of measuring DC voltages, but they are quite coarse/imprecise and subject to influence from static, high dV/dt waveforms (you know, like the kind produced by switchmode power supplies... ahem), so, yes, a resistor voltage divider on the output will be necessary.

But if you are concerned about the cost of capacitors you might be surprised at just how expensive this simple voltage divider will be... For example, to scale 150kV down to 5V with a 10k output impedance the upper resistor in the divider needs to be 300G (that's 300 gigaohms) and it will dissipate a maximum of 75W. Practically speaking, that means a whole lot of resistors in series to achieve the necessary resistance and power rating, while also avoiding errors due to voltage coefficient (ie - resistance value change with voltage). One possible solution would be 60 5G/2.5W Ohmite Mini-Mox resistors in series, at a cost of $5 each.

So, yeah, ain't nothin' cheap 'bout a 150kV regulated power supply.

[BootstrapBill]

I didn't explicitly state this in my previous post - and then got sidetracked by Tim with the C-W stuff - but the main reason for wanting the ADC monitoring the output voltage to have a finer step size than is strictly necessary is to avoid/minimize limit cycle oscillations, a phenomenon in sampled data systems where the output value oscillates between two steps because the reference value has been set to something between them. This applies to the digital pulse width modulator (dPWM) as well - it needs to have an even finer step size than the ADC.

You can employ some fairly heavy math to determine just how much finer the ADC and dPWM step sizes need to be, but I tend to categorize such as "getting a really precise answer to a hopelessly imprecise question"; ie - of dubious utility. In my experience, simply going with +2b higher resolution for the ADC and another +1b for the dPWM is sufficient. E.g. if there are 256 (8b) output voltage steps then a 10b ADC and an 11b dPWM will usually suffice.

Thanks, function of the feedback was what I was getting at with resolution, not achievable voltage steps.

So after reading over the Pressman buck current-fed bridge portion it seems they are well suited for multipliers. I now understand the overlap and like the idea of not blowing $100+ from possible shoot through. It states the frequency must be 2x for the buck portion. I wanted something well out of the audio range and the magnetics company was pushing me towards 40kHz. Not sure you can do 80kHz with an IGBT brick.

Also any thoughts of the voltage-fed buck capacitor? Could I get away with a smaller value or do I need to maintain less than 1% ripple.

Thanks

[MagicSmoker]

...
So after reading over the Pressman buck current-fed bridge portion it seems they are well suited for multipliers. I now understand the overlap and like the idea of not blowing $100+ from possible shoot through. It states the frequency must be 2x for the buck portion. I wanted something well out of the audio range and the magnetics company was pushing me towards 40kHz. Not sure you can do 80kHz with an IGBT brick.

Also any thoughts of the voltage-fed buck capacitor? Could I get away with a smaller value or do I need to maintain less than 1% ripple.

If there is just one buck switch then, yes, it is best to run it at twice the switching frequency of the bridge switches (so that each bridge leg gets a pulse from the buck). However, it is much more common to split the buck stage into two interleaved (ie - phase shifted) channels running at the same switching frequency as the bridge.

Your switching frequency target of 40kHz for the bridge is totally unrealistic. Maybe if you use SiC MOSFET modules ($$$) or go with a soft-switched or series resonant converter (both of which are more complicated and therefore more difficult to get working reliably) you can run a 100kW offline converter at 40kHz, but in a typical hard-switched converter the practical upper limit on switching frequency is when switching losses reach parity with conduction losses, and that will be around 10kHz with current generation 600V IGBT modules.

As for sizing the buck output/bridge input capacitor, the conservative approach is to design it for either application; ie - assume no ripple current cancellation from the bridge sucking charge out of it at the same time the buck is dumping charge into it.

[T3sl4co1l]

FYI, at a PPoE, we were doing >50kW per IGBT brick, with derating above 30kHz or so.  This was 5 years ago, but the transistors were part of the then-20-year-old legacy design.  Our replacement design was expected to deliver slightly less power per module, with no derating to about double the frequency (i.e., ~40kW to 60kHz).

The protection circuit I designed (at first for the old transistors, but aimed towards easy integration of the new modules) was capable of protecting the transistors from complete short circuit events.  We measured a short circuit peak of 6kA for 3us, with the transistor surviving the event.

The legacy generation had no protection circuitry whatsoever, and IGBT bricks grenaded on a regular basis.

I do not use "grenaded" lightly.  Arc flash inside an IGBT module propels shrapnel.

Again, considerations like these are myriad, and the OP has shown absolutely no competence towards necessary protective features like this.

Certainly not with an Arduino.

We used an FPGA to implement a centralized controller architecture.  I would've preferred solving it with discrete logic, distributed among the gate drive modules, and an analog controller, but an FPGA is tolerable.

We burned very few transistors due to controller errors, even throughout early development.

An ATmega (the heart of the most common Arduino) would be suitable for setting the analog controller's setpoints, but obviously not Arduino code, which is terrifying just for making LEDs blink.  Arduino is utterly and appallingly unsuitable for any serious application like this.

And if one does not understand why this is the case, one should most definitely not be attempting anything more than blinking LEDs, at this point in their career.

Tim

[BootstrapBill]

So I was hoping to use these..

http://www.infineon.com/dgdl/Infineon-FZ400R12KS4-DS-v03_04-en_de.pdf?fileId=db3a304412b407950112b4336f045caa

After switching these at 40kHz (have a bunch laying around) I can see that it is in fact unrealistic. Very square at 20kHz however.

http://www.pwrx.com/pwrx/docs/cm300ha24h.pdf

Yes I would not attempt to use an Atmega. The DUE has a 32bit atmel processor.

[MagicSmoker]

FYI, at a PPoE, we were doing >50kW per IGBT brick, with derating above 30kHz or so....

Hmm... and what was the datasheet rating for these (presumably half-bridge) bricks?

So I was hoping to use these..

http://www.infineon.com/dgdl/Infineon-FZ400R12KS4-DS-v03_04-en_de.pdf?fileId=db3a304412b407950112b4336f045caa

After switching these at 40kHz (have a bunch laying around) I can see that it is in fact unrealistic. Very square at 20kHz however.

Infineon makes some pretty fast 1200V IGBT dice, arguably the best based on current vs. switching losses, though Fuji has made some impressive progress over the last few years; Powerex (Mistubishi), in contrast, has about the slowest IGBT dice, and seems to be more concerned with the medium voltage (ie - 1700V to 4500V) market.

Since you already have a bunch of the above modules you might as well use them, especially if the incoming supply will be rectified 480VAC (ie - a DC bus of ~680V). Since they are rated for a lot more current than you will need you should even be able to run them at 20kHz, though switching losses will be about 4x higher than conduction loss at that point. Note, however, that at 30kHz the ratio of switching losses to conduction will be ~6x, while at 40kHz the ratio jumps to ~10x! At some point the increasing volume/complexity of the heat removal system required to run the modules at higher and higher frequencies will outweigh the volume reduction in the transformer. And not to point out the obvious here, but higher switching losses = lower efficiency.

As for an Arduino (regular or Due), the problem with these platforms is that they do not guarantee an interrupt service latency time, which can be downright disastrous in applications which require consistent timing and fast response to changing conditions, like controlling the state of the switches in an 100kW inverter...

I also agree that you should obtain some professional help if for no other reason than to save you a lot of time/money building stuff that has no chance of working. I'm not going to give you a safety lecture - either you are a responsible adult who knows his limitations, or you'll be losing the proverbial eye, likely from exploding IGBT modules...

[Marco]

The GAN MOSFET I mentioned before will burn something around 100 mW on the gate drive at 100 kHz and switches in 10 ns ... what's the point in messing about with IGBTs?

PS. just to point out the obvious, you're also going to have to make or buy a PFC.

[Richard Head]

GaN MOSFETs are twice the price of IGBT's and their anti-parallel diode has a Vf of around 6v which is horrendous. IGBT's give good value for money at present (until they are toppled by GaN in a few years maybe).
I'm really surprised no one has seriously considered a series resonant topology such as LLC for this application. You can use IGBT's at a much higher frequency than if they are hard switched. The leakage inductance (which is huge due to the isolation requirement) can be lumped into the resonant inductor and you need no output inductor. Also EMI is almost non existent compared to a hard switched converter.
The dynamic range issue can be a problem but there are ways around that also. The transformer doesn't see any high frequency ringing currents which often plague high power switched converters causing them to run hotter than expected. It's nirvana from all angles! 

[Marco]

GaN MOSFETs are twice the price of IGBT's and their anti-parallel diode has a Vf of around 6v which is horrendous.

They say 1.9V at 32A for the one I linked earlier, the datasheet for the hugely oversized brick IGBT he wants to use suggests it would manage 1.5V ... meh. Not unimportantly the GaN MOSFET's datasheet also claims a reverse recovery charge of bugger all. Going to need a IBGT with a SIC diode to compete with that, there goes the price advantage.

For a one off I don't see why you wouldn't just go with whatever makes your life easiest.

[Richard Head]

GaN MOSFETs are twice the price of IGBT's and their anti-parallel diode has a Vf of around 6v which is horrendous.
Apologies. Re-checked my data and that's bullshit. Vf is not 6V.
SiC is also an option. See attached Semikron device.

[T3sl4co1l]

GaN and SiC are excellent ideas.

Their different drive requirements and performance specs will keep the OP blowing transistors, and burning budget, instead of burning humans.

Tim

[Marco]

It's not like he can drop in some datasheet/white-paper solution any way. It's all custom.

Having a FWD which behaves closer to an ideal diode makes life easier. Having a switch which behaves closer to an ideal switch makes life easier. IGBTs with silicon FWDs, not so much.

PS. though I think he should just ask for a quote from SK Electrics, since they already sell a high frequency high voltage transformer in the ballpark of his spec. At 23 kg it seems well designed too.

[MagicSmoker]

The GAN MOSFET I mentioned before will burn something around 100 mW on the gate drive at 100 kHz and switches in 10 ns ... what's the point in messing about with IGBTs?

PS. just to point out the obvious, you're also going to have to make or buy a PFC.

Yes, let's switch 340V in 10ns across the primary of a 10-15kV output transformer... What Could Possibly Go Wrong? Hint - stray and distributed capacitances. Also, RFI...

And the OP is in the US so PFC is not necessary, nor would it even be desirable unless he's approaching the maximum usable VA of his service drop and/or the utility applies a penalty for poor power factor (usually only done to industrial and larger commercial customers).

GaN and SiC are excellent ideas.

Their different drive requirements and performance specs will keep the OP blowing transistors, and burning budget, instead of burning humans.

 

Especially GaN switches... those things seem to break out into destructive oscillation just by looking at 'em funny.

[T3sl4co1l]

Especially GaN switches... those things seem to break out into destructive oscillation just by looking at 'em funny.

Hell, I had a Si SuperJunction type singing at up to 400MHz the other day!

Ferrite beads are a wonderful thing

Tim

[Marco]

Yes, let's switch 340V in 10ns across the primary of a 10-15kV output transformer

Hard switching at 30 kW seems a bit silly period, do you really want to burn a couple kW in the switches?

Any way, there's the switch and the diode. You'd rather the diode switch off instantly when reverse voltage is applied, with only capacitive losses, instead of having a silicon diode snap off near instantly after allowing significant reverse current to build up if you happen to drive it just right/wrong.

[SeanB]

Any way, there's the switch and the diode. You'd rather the diode switch off instantly when reverse voltage is applied, with only capacitive losses, instead of having a silicon diode snap off near instantly after allowing significant reverse current to build up if you happen to drive it just right/wrong.

Well, it will do that for at least a half second before it either turns to smoke, or smoke, flame and loud noise. Depends on the diode, and what you consider to be significant current.

[Marco]

Lets say 10 times their rated continuous current during reverse recovery. Driven just right/wrong some even supposedly soft recovery diode can get snappy and switch that current off in well under a ns, the smoke is more likely to come out of other parts when they do.

[BootstrapBill]

I'm really surprised no one has seriously considered a series resonant topology such as LLC for this application. You can use IGBT's at a much higher frequency than if they are hard switched. The leakage inductance (which is huge due to the isolation requirement) can be lumped into the resonant inductor and you need no output inductor. Also EMI is almost non existent compared to a hard switched converter.
The dynamic range issue can be a problem but there are ways around that also. The transformer doesn't see any high frequency ringing currents which often plague high power switched converters causing them to run hotter than expected. It's nirvana from all angles!

LLC is what contemporary x-ray generators use. Its pretty insane how small they are. I would love to be able to do this I just feel the control and R&D would take time and be even more dangerous. Is there a preferred off the shelf control IC you would recommend?

Regarding the buck current-fed full bridge.. is it feasible to go any higher in frequency from the voltage-fed? Are switching losses any lower?

PS I have an oem equipment solution for this particular application but am still interested in making a low voltage/ low power version for educational purposes and possibly for the future.

[Richard Head]

With such a high turns ratio in the transformer the inter-winding capacitance of the secondary will be huge. Once this is reflected through to the primary (by the square of the turns ratio!) it will be obvious that this capacitance will drive the topology of choice. I initially suggested an LLC but in order to absorb the reflected secondary capacitance an LCC topology would probably be a better option. This topology allows you to absorb the parallel capacitance into the lower resonant capacitor. It unfortunately only has lossless switching on  the turn-on edge. Turn-off edge is lossy.
At high power levels I reckon that a regenerative snubber should be considered for the turn-off transition. Don't go for a half bridge and split resonant caps as it'll require a transformer with twice the turns ratio (with 4x the capacitance also!) as a full bridge.
The LLC and LCC arrangements are strictly speaking multi-resonant topologies as their natural resonant network frequency moves with load as does the excitation frequency.
The control strategy is variable frequency working above resonance. Always work above resonance (inductive mode) or lossless switching will be lost.
Light load operation at high line can be a problem as the transfer function is unfavourable in this regime. Sometimes pulse skipping is required at no-load to keep the voltage from running away.
These things are a bitch to design but once it's working as intended and the niggley problems addressed it should beat any hard switched approach hands down.

[MagicSmoker]

...
These things are a bitch to design but once it's working as intended and the niggley problems addressed it should beat any hard switched approach hands down.

This is precisely why I did not recommend any kind of resonant or quasi-resonant topology to the OP. This will be his first smps design, after all...

In fact, anything more complicated than a low voltage buck converter is likely to be more of a learning experience than the OP expects, but rather than flat out discourage him (or insult him into hiring a pro to design it for him - a curious marketing strategy, that), I instead chose a topology which is rather forgiving to design and highly tolerant of abuse.

For example, any kind of modulated bridge converter is almost a non-starter because the layout and timing of the bridge switches is very critical which means extra attention needs to be paid to the gate driver design and layout, and this is experience that one tends to acquire from, well, experience. Ie - not only blowing stuff up, but learning from the bits of shrapnel and escaped magic smoke. Thus no type of modulated bridge is suitable for a beginner.

Resonant (frequency modulated) converters are downright nightmarish to get working correctly, and operating on the wrong side of the resonant hump, or with the wrong Q, or about a dozen other arcane to mundane restrictions, can get you into trouble fast. It's hard enough for people like me who've been designing smps for 20+ years; I certainly wouldn't recommend it to a beginner. Frankly, I only recommend resonant converters for specific and highly reactive loads like, for example, transverse RF-excited CO2 lasers or high power ultrasonic transducers.

Quasi-resonant (aka resonant transition, lossless transition, etc.) can combine the wide load range performance of hard-switched with the (sometimes, not always) lower losses of resonant, but also seem to combine the worst traits of both converter families when it comes to designing them (especially those which require an auxiliary switch network to achieve lossless switching).

Even if you possess the requisite knowledge and experience to design any kind of converter, real world constraints often steer you away from the more exotic even if theoretically ideal topologies. Unless you design LLC converters every day I can guarantee you it will take much longer to get one working than a buck fed bridge, and at some point just the sheer difference in design time outweighs any potential operational efficiencies, especially for an intermittent load as I imagine an x-ray tube to be.

[Richard Head]

Quite frankly you'd be insane to try and build a power supply like this as a first ever SMPS. There are potential pitfalls everywhere when it gets to such high powers. Issues that are a small irritant with a 250W supply can be a show stopper at 100kW. For example it's not a problem to snubb the ringing across the output diodes of a LV 1kW supply but when the ringing is at 200kV on the secondary side and the snubbing power lost perhaps 500W - 1kW the viability of a hard switched topology will be questioned unless operated at 10khz maybe. The main problem with this type of supply as I see it is that there is a very high leakage inductance due to the high isolation requirement AND there is a huge reflected capacitance from the secondary. The high leakage will resonate with the parallel capacitance causing terrible ringing across the output rectifiers which are almost impossible to snub at those voltages. My suggestion is not to fight physics but use it to your advantage whenever possible.

[MagicSmoker]

Quite frankly you'd be insane to try and build a power supply like this as a first ever SMPS. There are potential pitfalls everywhere when it gets to such high powers. Issues that are a small irritant with a 250W supply can be a show stopper at 100kW....

Again, I completely agree, but let's just say that often times experience is the best teacher, hmm? In this case, a buck fed bridge is likely to survive long enough for the OP to make some measurements and, perhaps, figure out how to address the issues such as you (me, T3sl4co1l, Marco, etc...) have raised along the way.

[Richard Head]

Even if you possess the requisite knowledge and experience to design any kind of converter, real world constraints often steer you away from the more exotic even if theoretically ideal topologies.

I agree. Very often this is the case. And as you mentioned, design time can be a huge factor if you don't have decent volumes to amortize the development cost over.

[BootstrapBill]

The MONSTERBUCK 5000..

Actually the only thing 5000 about it is what the load is rated for.. in water.

I have a ~700uH inductor I wound. Core is a 58337 HighFlux series from Mag-inc. The Load is a 12ohm water heater element.

Operated at about 100-150vdc input. Was able to go from 0 to ~80 volts out. Very smooth output.

The waveforms are across the diode and igbt. The diode is the one with the severe ringing and the igbt waveforms are of a before and after snubber.

[jbb]

Operated at about 100-150vdc input. Was able to go from 0 to ~80 volts out. Very smooth output.

The waveforms are across the diode and igbt. The diode is the one with the severe ringing and the igbt waveforms are of a before and after snubber.

OK, now we're talking :-)  It's nice to have some goodies on the bench.

First, practical safety improvements:

• Add a discharge resistor across the main DC cap.  That way it will discharge without the aid of a screwdriver.  Also, electrolytics have an interesting trick whereby they actually recharge themselves after you remove the shorting link.
• Adding a big red LED (+ resistor) onto the main DC cap will let you see if it's charged.
• Ideally, get some kind of box with a clear lid so you can put the higher voltage stuff under cover before testing. Much safer.
• I see you're using a breadboard to hold a cable down.  You need to anchor this stuff so it doesn't slide around. (A 200V capacitor sliding onto your hand will hurt quite a bit or worse...)
• If that's a shared table, a suitable sheet of plywood or similar can be helpful. You can mount your stuff to it and reduce the risk of damage when moving.  And protect the table from soot marks.
• To go further you'll need a current sensor. Something like a LEM HAS 50-S (available ex Digikey) might fit the bill.  Without knowledge of the current control schemes are generally limited to guess-and-bang methods.

I see you've got some ringing. We can help with that.  Ringing is caused by LC circuits.  In this case the C is the stray capacitance of the IGBT and diode.  You can't do much about those. The L is due to leakage inductance.  Have a look at the power circuit consisting of main DC cap, IGBT and diode.  The wires are quite far apart.  That means that the area of the loop they form is large. Loop area makes leakage inductance.  So try:

• getting some cable ties just squashing the wires together.
• shortening the wires
• getting a combo IGBT + diode module (you can also use a half bridge IGBT module with the gate of the unusued IGBT shorted to its emitter)

Then take a look at the gate-drive circuit;

• Get the gate-emitter loop nice and tight to reduce leakage inductance
• It looks like you're using an opto-isolated driver. That's good.  I suggest you add a little isolated power supply to protect your bench supply from accidents. It might not like having the mains connected to it.
• If the IGBT has a separate emitter terminal next to the gate terminal, use that.
• Mount the gate driver board close to the IGBT. Inductance is your enemy here.
• Consider using a gate driver with desaturation protection.

Next, you can improve the fundamentals of the circuit.  I see that the power supply return on your output filter cap goes back to the diode and then daisy-chains to the main input cap.  I suggest that it should go straight from the output cap to the common point on the main cap (like a star ground).

Good luck

[BootstrapBill]

Quasi-resonant (aka resonant transition, lossless transition, etc.) can combine the wide load range performance of hard-switched with the (sometimes, not always) lower losses of resonant, but also seem to combine the worst traits of both converter families when it comes to designing them (especially those which require an auxiliary switch network to achieve lossless switching).

Looks like these guys are using this..

http://www.glassmanhv.com/glassman_tech.shtml

Any passive way of doing this with the buck current-fed bridge? I know you mention it's just as big of a pain as the resonant but it's the frequency modulation and more importantly the narrow output voltage range that turns me off. The articles I've read on this all seem to be quite elaborate, sometimes more than doubling the original part count. Also haven't found too many for the full bridge.

[Siwastaja]

Re photos: is this a joke? For real? Please just stop now, you clearly are not even close to be able to even start this kind of project. I'm sorry!

"Arduino" talk indeed was a good sign of total cluelessness, once again.

[BootstrapBill]

Arduino is a due. Planning on using atmel studio 7. Yes I am aware of the delays/interupts with the arduino language.