240VAC single phase to 650VDC voltage doubler / boost converter. 3

[jraef]

I'm looking for a way to supply roughly 75A, 650VDC (to a VFD DC Bus) from a 240VAC single phase source, without a transformer. I'm doing it with a transformer now, the problem is the size and weight (>500 lbs) for the machine OEM, as it is for the transportation industry and everything heavy means money. I've though of a voltage doubler circuit on the rectifier side, but I can't find a resource using Google and the couple of VFD suppliers I knew of that used to do it no longer do, indicating there may have been a problem with the concept. I get thousands of hits on circuit designs, not suppliers. I also thought of rectifying the AC to DC, then using a boost converter on the DC side, still no hits on suppliers with anything that big or at that voltage. I'm no longer in a place to be able to have my own company make this, or rather if I were to convince someone to take it on, I will be retired by the time it's released.

Any ideas? I'm looking for someone who already makes it, or close enough to it who would be willing to modify.

"Will work for (the memory of) salami"

 

[SceneryDriver]

I think you will have trouble at that power level. You can derive approximately 340vdc by full-wave rectifying 240vac and filtering it. Capacitive voltage multiplier circuits will work to get you the voltage you desire, but would have to be physically HUGE to achieve your desired amperage. I've seen VFD's that will take 120v single phase line power and spin a 208v 3ph motor, but those are limited to small motors.

Perhaps a transformer-based circuit that operates on a frequency higher than 60hz could work. Higher frequency transformers can be much smaller and much more power dense than those that run at line frequency. Have you considered a 400hz power system, like are used on aircraft?


SceneryDriver

 

[Skogsgurra]

Agree. It will be difficult to find a single unit with that voltage and amperage.

But, I think that a combination of several series connected standard switchers (cheap) can produce the voltage needed. The problem may be that the outputs have to work at elevated insulation levels. But, if you find something that can do this, then your need for current can probably be met if you parallel several Groups os series connected switchers.

There may be an optimum configuration, but I doubt very much if can beat a standard transformer. A compromise could be to use an autotransformer (not very good if you need to rise voltage around 100%, but still) and then use a doubler Circuit. Either way, it looks gloomy.

I use a cascade rectifier for voltage tests, but that one can only deliver around 50 mA and is already quite HUGE.

Can't the motor be wound for a lower voltage? Tell the customer about the problem with PWM in old motor windings and that the motor probably needs rewinding anyhow. Worth a try, I think.




Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[itsmoked]

Jeff: I suggest a high power switcher is your only viable bet.

Sorenson makes DC supplies up to 130kW. They have a couple of series that exceed 50kW and several that don't but are parallel-able.

Sorenson (Now Ametek)

If you're talking a quantity instead of one or two you will likely get very good attention.



Keith Cress
kcress -

http://www.flaminsystems.com

 

[waross]

I like SceneryDriver's suggestion.
It looks like you need 50 KVA or more of capacity.
A 5 KVA, 60 Hz transformer will handle 50 KVA at 600 Hz. You may want to strip the windings and rewind the transformer with fewer turns of heavier wire.
At higher frequencies the transformer gets proportionally smaller.
If i was going to attempt to use a VFD to power a small three phase transformer bank one of the first people I would ask for help is jraef!
How about three 3 KVA transformers at 300 Hz?

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[jraef]

Would that I could change the motor voltage... The issue is, this is not a standard AC induction motor, it's a 480V IPM AC motor, being used as a de-facto servo. We can control the motor, but the drive that can do this is only available as a 480V version and it is being used in an application that only has 240VAC single phase power available from an on-board generator. I can handle the single phase input and like I said, the transformer works, it's just so damned heavy. Yes, right now we are using an isolation transformer and given the situation, I think an autotransformer design might take some significant weight out of it, I was just looking for another alternative for them given the heartache they expressed over having a transformer at all. So far, we are the only ones to propose ANYTHING that works given all the restrictions, it's just that the 500 pound gorilla (transformer) in the room refuses to be ignored.

Keith, Sorenson/Ametek is looking like almost exactly what I wanted! Nothing they have exactly fits the bill, but they appear to be someone I can talk to about it, and yes, it's an OEM, so there are quantities involved. But the one thing they have that's close weighs in at 2150lbs, so obviously they are heavily magnetic in that design. Still, like I said, someone to talk to tomorrow. PLS (as Gunnar refers to them) for you!

"Will work for (the memory of) salami"

 

[ScottyUK]

Schaefer Power in Germany and - as I just found out from their website - the US do a good range of switchmode units. Their C-Series is borderline indestructible in the smaller sizes (we've tried) and they can parallel units to get the power level you want. Their catalogue doesn't list anything quite as big as you're looking for but they do custom work.

http://www.schaeferpower.de/cms/en/produkte/ac-dc-stromversorgungen-a-batterieladegeraete.html

 

[jraef]

Thanks Scotty,
Ironically, we use Schaefer for EMC filters and sine wave filters all the time, but I'm embarrassed to say I've never looked at their entire product line. I tend to pigeon-hole suppliers into niches that fit what I know of them already. I suppose I'm a marketing department's worst nightmare in that regard.

"Will work for (the memory of) salami"

 

[jraef]

Lol, two of those Shaefer units in parallel... 506lbs.

Something is nagging at me here... Something about "free lunch"...

"Will work for (the memory of) salami"

 

[waross]

Compared to this;
" But the one thing they have that's close weighs in at 2150lbs,"
It may just be a light lunch. Grin.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[mikekilroy]

only problem with using a higher freq thru the transformer is you will need an ac to dc bus then vfd type gadget to get back to 400+hz ac so the savings probably goes away.

why 500#? Is that just a round number thrown out? My guess is a 230->460 iso xfmr should weigh in around 350#?

the same as auto should weigh same as a 25kva iso or maybe 250#? alum might shave another 10-15% Cu windings (a bit more steel but net weight loss)... alum encl to reduce nema 1 weight or no encl and put in your customer's oem cabinet? optimize for 150C or higher temp rise (can use 220C insulation)? I would think a good custom xfmr mfgr could shave weight down to about 200# as auto? since you say there are more than 1 pc a lot of custom xfmr designers would put an hour into seeing how light they could make it....

anyone make a dc bus pump up unit like Siemens that does not necessarily use a xfmr (just a choke that weighs a lot less)? I can't think of a mfgr off top of my head but maybe some googling on whatever words siemens uses for their dc bus pumping scheme? I hear there is no real limit to how high you can pump the dc and with high currents....

http://www.KilroyWasHere<dot>com

 

[Sparweb]

Jraef,
Not sure how useful this will be, but...
Is there any truth in the assumption that an electronic device that can do one thing will weigh approximately the same as an electronic device that can do the opposite?
Transformerless grid-tie inverters are on the market that can regulate input voltage up to 600VDC and "sell" it to the grid at 240VAC single-phase (but at your power level more likely 3-phase at higher voltage). I quickly looked up one from ABB sized about 100 kW (I think that's the range you are looking for) then found its weight: 550 pounds.

You may be pushing a rope.

STF

 

[sibeen]

Maybe talk to some of the UPS suppliers. They've all basically moved to some form of active front end and a DC boost topology. Granted, they're not going to have anything off the shelf at 50 kW and single phase; but if there is a quantity of units involved it may perk their interest.

 

[iop995]

80kW, wide range input and output voltage / 28kg water cooled.

http://www.brusa.eu/en/products/converter/power-dcdc-750-v/gic246.html

Don't ask price
Best solution is DC/DC boost converter. I'm sure there are many other much cheaper solution than that (ok, not so small weight).

 

[mikekilroy]

WOW!!!!!!!!!!!! and only 61#!!!!!!!!!!!!!! Being for EV use, it probably is reasonably priced too! Great find! lps!

http://www.KilroyWasHere<dot>com

 

[racookpe1978]

Lol, two of those Shaefer units in parallel... 506lbs.

Something is nagging at me here... Something about "free lunch"...

OK ... But would they not create the right voltage (plus a margin) if they were in series, not parallel?

True, it would be more weight for the power supplies, but that eliminates the transformer weight, doesn't it?