Operating three phase high voltage power supply on single phase power

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.
Dan Knapp
Posts: 263
Joined: Wed Aug 06, 2008 12:34 pm
Real name: Dan Knapp

Operating three phase high voltage power supply on single phase power

Post by Dan Knapp » Sun Sep 23, 2018 7:00 pm

From time to time, a bargain appears on eBay in the form of a high power high voltage power supply that uses three phase power input. Since most fusor enthusiasts don't have three phase power available, they are not generally able to take advantage of what could otherwise be a very good deal. I offer here an approach to operating such supplies on single phase power (at reduced output current of course).

Our lab recently received a loan of a large Matsusada AKP40P320 40 kV, 320 mA, 13 kW! supply (this is a positive HV supply not for IEC work, but the plan is equally applicable to a negative supply). It uses 208 VAC, 3 phase, 44A (for full output) power input; and I have been trying to decide how to power it since we don't have three phase power in the lab. We don't need anywhere near the 13 kW capability and, indeed, this amount of power is a bit scary. One can operate such supplies at lower output current drawing proportionately lower input current, but the three phase input was still an obstacle. Single to three phase converters are readily available for powering three phase motors on shop machines. The "rotary" type converter uses a three phase motor fed with single phase, and the turning motor generates the third leg. These work OK for machines, and one can cobble one together at low cost from a surplus three phase motor; but they have the disadvantage of giving asymmetric output, consume wasted power turning the idler motor, are noisy, and present a safety hazard with the turning shaft (sealed idlers with no shaft are available, but they cost more than a surplus motor). Solid state phase converters give more efficient conversion and symmetric power, but they are quite expensive. I decided to look into whether one could just use single phase power. Three phase supplies use three phase because they are high power supplies that need more input power. If one doesn't need the full rated power output, one could feed them with less power and still get a usable high voltage output power.

The Matsusada manual gives no description of the circuitry, but in reading the circuit overview in the manual of a comparable Spellman supply, they state that the AC input power is immediately rectified with a three phase bridge. If there isn't something on the input side of the bridge that would create a problem, then one could use single phase power on two legs of the bridge, and the only thing the supply would see would be a lower DC current capability for the rail that feeds the inverter. I looked inside the Matsusada supply and traced the AC input to the three phase bridge as shown on the attached drawing. The only other connections between the input and the rectifier are leads tapped off of L2 and L3 to feed two low voltage power supplies for the control circuits. There are no other connections to the three phase AC lines that would tell the supply a leg is missing and there is nothing connected to L1 ahead of the rectifier. So it looks like one can just leave L1 unconnected and feed single phase power to L2 and L3 with the only result being a lower overall power capability.

This Matsusada supply is spec'd. at 208VAC ±10% input for full power output. It's not clear whether feeding it with my 240VAC (+15%) would be any problem, but I can get within the spec. using a boost/buck transformer to lower the single phase AC voltage. My plan is to use a 20A 240VAC single phase line and use the 20A breaker to limit output if I should inadvertently set the output current control too high for the available input current. I estimate that feeding the supply with single phase power (2 legs instead of 3) should give 2/3 [or should this be 1/3?] the rated power output (i.e. about 200mA maximum instead of 320mA), and limiting the input current to a 20A circuit instead of 44A should further halve the maximum output capability to about 100mA. This is still far more than we need for the application, but much less likely to vaporize our experiment in a malfunction than the full output. I haven't actually fired this up yet because I'm awaiting delivery of a receptacle to match the supplied power cord plug (it needs an expensive Leviton CS8369 receptacle, but a used one on Amazon is cheaper than replacing the cord, plug, and receptacle). I'll post results when I have this running (or smoke the supply). I welcome input if anyone sees a reason why this should not work.
Attachments
AKP40P320inputpwr.pdf
(68.54 KiB) Downloaded 171 times

User avatar
Rich Feldman
Posts: 1108
Joined: Mon Dec 21, 2009 11:59 pm
Real name: Rich Feldman
Location: Santa Clara County, CA, USA

Re: Operating three phase high voltage power supply on single phase power

Post by Rich Feldman » Sun Sep 23, 2018 10:51 pm

Sounds very similar to using VFD's (variable frequency motor drivers) with 3-phase inputs on single phase wall power.
Jerry Biehler spoke up authoritatively when that subject was discussed here, maybe a year ago, for a vacuum pump application.

I know that for some units, the manual identifies which one of the three power input connectors to skip. Maybe the other two feed an internal control transformer. Duh, just like In the diagram you posted, where you need to skip the red wire when using 1-phase.

When I speculated that the manuals would also specify power derating for operation on 1-phase,
Jerry said that newer VFD's didn't need derating like the old ones.
It's a pretty broad subject. I bet there are VFDs (and HV power supplies) with ratings ranging from below 1 HP (or kW) to over 100.

Dan, your plan to use a mains-frequency buck transformer is within my personal experience. Sounds like a fine idea, if the 208-V 3-phase unit doesn't say it will work on 240 (which runs close to 250 at my house). Transformers sold as "buck boost" units, say from Grainger, cost a lot more than ordinary power transformers with the same volt and amp ratings. The electrical code is picky about one thing: no overcurrent protection device may interrupt the connection between the boost secondary winding and the hots side of primary winding. The safety reason is obvious on inspection of a schematic, IIRC. Beware of free advice from the Internet.

Hey, how about doing what amateur machinists have done in garages for decades. Generate real 3-phase kilowatts with a Rotary Phase Converter, otherwise known as an idling 3-phase industrial motor. Maybe cheaper than a VFD when you need to run motors, but probably unnecessary when the load is a power electronics unit.
Mike echo oscar whisky! I repeat! Mike echo oscar whisky, how do you copy? Over.

User avatar
Rich Feldman
Posts: 1108
Joined: Mon Dec 21, 2009 11:59 pm
Real name: Rich Feldman
Location: Santa Clara County, CA, USA

Re: Operating three phase high voltage power supply on single phase power

Post by Rich Feldman » Tue Sep 25, 2018 11:51 pm

Too late to edit prev post.
Dan, sorry about me being more eager to write than to read. Didn't need to repeat details in your OP
about rotary converters, knowing which two of the 3 wires go to the controls, etc.


The "Input 208 VAC 3 phase 44A" note in your power section diagram is in same ballpark as some large options of instruments that I help to design. On factory floor they use the thickest cords and largest wallplugs I've ever seen.
DSCN0203.JPG
.
You mention wiring up a used receptacle. Sorry, no picture available of recently destroyed 3-phase switch/breaker like the one in your diagram. Some mechanical assembly tech hadn't properly tightened at least one setscrew where the thick wires connect. By the time somebody smelled burning electric stuff, the wire insulation and breaker enclosure plastic were melted and charred within about a 2 inch radius.
Mike echo oscar whisky! I repeat! Mike echo oscar whisky, how do you copy? Over.

Dan Knapp
Posts: 263
Joined: Wed Aug 06, 2008 12:34 pm
Real name: Dan Knapp

Re: Operating three phase high voltage power supply on single phase power

Post by Dan Knapp » Wed Sep 26, 2018 2:50 am

Rich
I didn’t follow your comment on the code prohibiting over current interruption of the connection of the secondary winding. Why would anyone do that? Can you elaborate?

User avatar
Rich Feldman
Posts: 1108
Joined: Mon Dec 21, 2009 11:59 pm
Real name: Rich Feldman
Location: Santa Clara County, CA, USA

Re: Operating three phase high voltage power supply on single phase power

Post by Rich Feldman » Wed Sep 26, 2018 4:09 am

Stumbled on that a year ago, holding forth on another forum where someone actually rewound a MOT secondary and used it as a boost on 240 V.
buck-boost_xformer_208-240v.jpg
.
Even with snipped image in front of me, it took a while to re-find the source (or one like it).
Page 2 of https://www.emerson.com/documents/autom ... 164082.pdf

In our thread here I might have mis-rembered the rule. Now looking at the schematic, and since NEC told us what not to do, here's one hazard that jumps out.

Suppose the boost transformer "shunt" winding were interrupted between points A and B. E.g. after someone had mistakenly wired a switch or fuse there, for turning off or protecting the transformer. How could an extra fuse hurt anything?

That would leave unboosted mains voltage applied to the series connection of 32V winding and the load.
If the resulting division puts more than 32 volts across that winding, it could be a problem.

Overvolting the "series" winding might make the transformer core saturate, with current limited by the load. Probably OK, unless the transformer eventually overheats from ordinary copper loss + extraordinary core loss.

With or without saturation, an overvolted secondary will cause an overvolted primary. NEC doesn't want that to happen downstream of a circuit breaker in the OFF or TRIPPED position, even if it's a two or three pole breaker. I bet they thought of the rule after some accidents.
Mike echo oscar whisky! I repeat! Mike echo oscar whisky, how do you copy? Over.

Dan Knapp
Posts: 263
Joined: Wed Aug 06, 2008 12:34 pm
Real name: Dan Knapp

Re: Operating three phase high voltage power supply on single phase power

Post by Dan Knapp » Wed Sep 26, 2018 3:32 pm

Thanks for the clarification, Rich. Another way of saying the same thing is that the overcurrent protection must be on the input side of the buck/boost transformer.

We should probably add an admonition here to anyone inspired to play with homebrewing a buck/boost transformer from a regular “step down” low voltage transformer. Be sure you understand how they work before powering up your creation. Commercial buck/boost transformers are expensive (a one KVA unit can cost $300-400 depending upon brand), but not all that much more than a comparable 24VAC control transformer with the same current rating. One reason for the cost is that the low voltage winding(s) must be sized to carry the full load current. Commercial buck/boost transformers typically contain two HV windings and two LV windings that can be used in a large number of different combinations. It is used as an autotransformer configurable for various line voltages and different amounts of buck or boost. The home brewer would be well advised to study the data sheet for a commercial unit. The numerous different ways to wire them can be daunting at first until you get your head around the basic concept. The nominal KVA ratings for commercial units appear to be derived from the current capability of the LV windings as opposed to the actual load capability of the unit (which can vary widely depending upon the wiring). The actual load capability can be much higher than this rating. For example, I’m using a 0.5KVA rated unit that has #10 wire on the two low voltage windings (the HV windings use #14). This 0.5KVA rated transformer, per the data sheet, can be used for as much as a 10KVA load with the LV windings in parallel and 5KVA with the LV windings in series. Just as a buck/boost transformer can be used as an (albeit expensive) low voltage transformer, a step down transformer can be used as a buck/boost if you understand how to wire it. One also has to get the phasing correct in connecting the windings. The two winding ends are labeled on a commercial unit, but you would have to find the correct ends to connect by trial and error on a home brew job.

Dan Knapp
Posts: 263
Joined: Wed Aug 06, 2008 12:34 pm
Real name: Dan Knapp

Re: Operating three phase high voltage power supply on single phase power

Post by Dan Knapp » Sun Sep 30, 2018 3:31 pm

I just verified that the supply will indeed run on single phase power. The attached photos show the powered up supply (only cranked up to 2 kV because the probe connection to the HV was dangling in mid air) and the power adapter. I used a buck transformer to drop my 240VAC to within the 208VAC +/-10% input voltage specification for the supply, but I doubt if this was necessary (circuit attached). I suspect this approach will work for most three phase high voltage supplies, but one must look inside to see which two legs are used for the control power, and to verify that there are not other connections to the unused third leg that might tell the supply to shut down if that leg is lost. If your supply lights up but gives no power output, look for a remote enable (which this one has) or interlock connection on an interface connector.
Good luck in finding a supply bargain on eBay that others have avoided because of the three phase power requirement.
Attachments
AKP.1ph.poweradapter.jpg
AKP.1ph.powered.jpg
AKP.1ph.powerAdapter.pdf
(96.46 KiB) Downloaded 158 times

User avatar
Richard Hull
Moderator
Posts: 11335
Joined: Fri Jun 15, 2001 1:44 pm
Real name: Richard Hull

Re: Operating three phase high voltage power supply on single phase power

Post by Richard Hull » Sun Sep 30, 2018 9:48 pm

Great report on tips on how to possibly reverse engineer 3 phase supplies to single phase operation. Sometimes an item comes to us so well suited to our needs, but requires inputs we don't have or thwart us in a number of other ways that drives us to "make it work for us". Manufacturers do what they do, and in return, we do what we have to do. Good job.

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
Retired now...Doing only what I want and not what I should...every day is a saturday.

Jerry Biehler
Posts: 856
Joined: Tue Nov 24, 2009 8:08 am
Real name:
Location: Beaverton, OR

Re: Operating three phase high voltage power supply on single phase power

Post by Jerry Biehler » Mon Oct 01, 2018 2:37 am

As long as you the unit does not have phase loss detection or an input where it uses multiple internal power supplies off the different phases to run the logic you should be ok with running it at reduced capacity on single phase with a bucking transformer. Probably 1/3rd capacity would be a good rule of thumb. If it does detect phase loss sometimes you can cheat it out by jumping the empty phase to one of the other phases to trick it. Rarely do they look at phase rotation so it might work.

Phase Perfect is one brand of electronic phase converter. They are very well regarded. An RPC would work but you would need a massive one. That's why I ended up buying an old military generator to run my e-beam power supply.

You should be able to find bucking transformers used pretty cheap. I have to use one on my vacuum system for the RF power supply as well and two on my laser welder. You use one for single phase and two for three phase.

User avatar
Rich Feldman
Posts: 1108
Joined: Mon Dec 21, 2009 11:59 pm
Real name: Rich Feldman
Location: Santa Clara County, CA, USA

Re: Operating three phase high voltage power supply on single phase power

Post by Rich Feldman » Mon Oct 01, 2018 4:50 pm

Lookin' good there, Dan. Nicely enclosed, labeled, and diagrammed.

Here's a question for you & Jerry about buck topologies.
Dan's schematic (pdf in prev post) shows the input connected between H1 and X4 (series connection of HV and LV windings).
We could get a slightly higher bucking ratio with the same transformer, by connecting input to H1 and H4. As one must, in a boosting application.
alt.buck.png
alt.buck.png (12.84 KiB) Viewed 2926 times
.
Anybody know if that's discouraged for some reason? I used to think it was the default, until seeing Dan's schematic, and one like it in source doc for overcurrent diagram in this thread. Here reconfigured using MS Paint program. :-) The transformers in that doc have H windings rated for full 120/240, among other things for alternate applications like LV lighting.
Mike echo oscar whisky! I repeat! Mike echo oscar whisky, how do you copy? Over.

Post Reply