3 Phase power

Hi group:

I have a question about 3 phase rotary converters, I currently have a 30hp synchronous 220V system, started by idler at 1750, fed through a bank of GE run caps and it produces 217 +- 2% at load target (about 50 amps). After a tiff with the local power company ( they dropped their requirement to provide 208/120 wye down to $9600! that is no help) I must add some capacity through an addition to my current system. I have 200 amp, 240v single phase to the shop, but I do not know what is the best way to get the 3ph power I require. I will need about 160 amps of 220 3ph (that is everything running x 1.3), will the losses make such a system impossible? should I have a "bank" of small induction idling motors(3 x 10hp) or one @ 25hp? Do I want a parallel hookup thru a switched bus of single phase to all motors or start the biggest one and go from there? simply adding capacity through the other idlers.

TIA Dave G.

Reply to
DAVE GEE
Loading thread data ...

That's about 40 horsepower going there, Dave. That's a whole lot of power! Are you sure you're going to run everything full blast all at once? My shop has a 30A 220V 3 phase setup which can easily start/run a 3hp motor. My design assumption is that I only use one machine at once, thus my phase converter need only power my biggest machine.

Grant

DAVE GEE wrote:

Reply to
Grant Erwin

Yes, that is correct.

Again that is correct, I think the current setup will produce about 80 amps of 3 phase power, but the machine it runs only "usually" draws about 40 amps, you can sometimes see it jump a little higher but I have never seen it go above

60 amps and then just for an instant (events like a spindle start with colant on and hydraulics operating to bring a tool into position, I try to avoid this thru programing)

snip

Reply to
DAVE GEE

Sorry about that, I can see why you would think that, let me clarify. The 160 amps that I am speaking about is the draw on the single phase supply side, that is where I measure the "draw" on the machines because from there it goes thru a buck/boost transformer into the 30hp idler and then thru the capacitors. Since I would be dealing with many unknowns if measured on the 3ph side I went with the single phase side for kiss reasons. also included in the 160 fig. are all the lighting cooling and recptacles that could be in use, this came to 120 amps x 1.3 or 160amps. These things that I am powering are medium sized CNC's that have pretty small requirements for powering actual 3 phase electric motors, most of the 3 phase is converted to DC 24v or 5v control voltage, or back to single phase ac 110v for the internal computers. The total for the actual "rotating" or motor requirements is about 12hp for both machines, and consists of hydraulic, lubrication, and coolant systems. None of these "idle" (with 2 exceptions) but come on when the PLC demmands them, so there is no, or negligable gain to the system. My question is, are several small idlers or one large idler better as far as being easy to "tune" or will friction and electrical losses make an array a non starter?

I think 1.7 is right IIRC 1.73 is the very number

Reply to
DAVE GEE

Why not buy a diesel 3 phase generator... Make you'r own electric...

Reply to
Kevin Beitz

I don't do this every day, but I came up with 67 KVA.

But, a later clarification says this will be 160 A on the single-phase feed. Do you need to run any other equipment such as air conditioners, lights, etc. from the same single-phase 240 V feed? If so, you are in big trouble. Even without, you are most likely going to need a power factor correcting cap bank to keep the total line current down.

For your question of multiple idler motors, it sounds like a nightmare. I would try for two motors at the max, just to try to make it simpler.

(Actually, the whole idea of rotary phase converters revolts me. I much prefer VFDs wherever possible.) Synchronous AC machines have one interesting feature, though. By varying the rotor field supply, you can vary the phase relationship between voltage and current. There used to be a machine called a synchronous capacitor, which was an over-excited synchronous AC machine, and a control device would vary the exciter to adjust the plant's power factor.

Jon

Reply to
Jon Elson

Hmm. That would be 1.4 or so. Which sounds like the square root of two, one of those magic numbers that keeps popping up all the time.

Where's Gary when you need him! Maybe if I insisted that the gent's 240 volt feed was two-phase then he would jump in here.

Then I could offer to bow down and say it really *is* single phase after all, if only he would explain how to calculate how much power would be delivered by a balanced three phase line with line-to-line voltage the same as a given single phase line.

I think it involves an integral over an entire cycle.

Whoops, looked it up, the book says that the power delivered is (sqrt 3) times that of a given single phase source.

Never mind.

Jim

================================================== please reply to: JRR(zero) at yktvmv (dot) vnet (dot) ibm (dot) com ==================================================

Reply to
jim rozen

What are you planning on running? I did a survey last week on a grinding shop, with 14 cylindrical grinders..big suckers, each with its own coolant system and waste conveyers...and shop lighting and whatnot all came to less than 140 amps. Start up came to something like 350 amps if every machine and every light was turned on simultainously.

Gunner

"What do you call someone in possesion of all the facts? Paranoid.-William Burroughs

Reply to
Gunner

Isnt there some sort of 'Boost Effect" that occurs when you have more than one 3 phase device running on the same circuit?

Gunner

"What do you call someone in possesion of all the facts? Paranoid.-William Burroughs

Reply to
Gunner

The friction losses are probably going to be pretty small. If I were doing this, I would use two motors, one smaller than the other.

The smaller one being 5 or 10 hp, depending on what I could find. The smaller one could be depended on to start the larger one, if you are not running the entire shop then there is no need to start the larger one.

You will need to properly balance these because your load will be near the limit of what they can do. The usual rule is the idler motor has to be 1.5 times the hp rating of what you want to run.

The reactive power draw from this will be large so you also will need to include power factor correction capacitors.

Jim

================================================== please reply to: JRR(zero) at yktvmv (dot) vnet (dot) ibm (dot) com ==================================================

Reply to
jim rozen

So basically, you want to convert the entire 160 amps into three phase power, yes? At least that is what the post above is saying. That would be about 92 amps of 'real' balanced three phase service, using the sqrt3 rule.

But if your three phase load is 160 amps at 240 volts that would be 38.4 kW which is about 51 hp. Using the rule that you need to have an idler motor capacity of about 1.5 times the rating of the machines you want to run means a total of 75 hp of idler motor. That's a lot.

Ah. If you really only need to get 12 hp of three phase, that makes your requirements a lot less severe. But I think that would mean you would have to somehow separate out the 'real' three phase loads and supply them separately, thru your converter setup. And power the other loads in the machine that are single phase, directly from the single phase service. Is that what you are doing now?

Right now you say you have a 30 hp idler. How much machinery hp is that running at the moment?

As I mentioned before in another post (sorry this thread is getting somewhat fragmented) the frictional losses will be pretty small. You might want to run a smaller converter that can be either capacitor or pony-motor started, with that idler being 5 or 10 hp. Then that one can start the other(s). Finding a 75 or so hp motor will be tough and expensive I think, and the inrush current will be quite large. So breaking the 'larger' one into two or three is not a bad idea. (You will probably exceed the 200 amp capacity by starting a single 75 hp idler, even just the transient - by enough to pop the main.)

But you will need to tune them and also correct for the power factor. It sounds like you are already doing that with your present setup, though.

Most of the folks here don't have installations nearly that large so one can get away with a lot. BUt when you are running close to the total allowed by your service entrance (160/200, though you did add in the 1.3 factor) one has to watch all the details pretty closely.

I am by no means a rotary guru, there are others here that know a *lot* more than I do.

Jim

================================================== please reply to: JRR(zero) at yktvmv (dot) vnet (dot) ibm (dot) com ==================================================

Reply to
jim rozen

Hi there: Actually no, only about 100 amps will be dedicated to 3 phase and 20 amps for 120/240 single phase with a fudge factor (x 1.3). as I stated above.

(from your earlier post)

Hi again: That is sort of what I am thinking, I would use the current 30hp to start a 25 (I know where to get one of these free) but am a little worried because the motors are so close in horsepower, do you think that would be a problem? (slow starting of the 25). I could always use a pony.

Again no, the machines are full of transformers, there is some latitude to tune these from 190v to 220v but the transformers and spindle thyristors (sp? these are in effect vfd's) all require 3ph power and they will alarm if the voltage varies more than about

5%. So while there is not much rotating 3ph requirement they do require 3 phase and I would not like to try to change the machine internals since the machines have no problem running on my current 30hp 3ph system (mind you, only one at a time). The thing is I want to be able to run both machines at once and I am VERY uncomfortable trying that on the current system!

Nor am I Jim, but here is my current thinking,

OK the 3 ph "load" on the supply side should = about 100 amps of 240 (really 248-253v where I am) or about 25kw or 34hp this gets transformed down to about 220v there is some loss here but I don't know how to quantify it. The two idling motors should be capable of 55hp / 1.5 for 36.6 hp so I think I am in the ballpark. Much of this is "guesstimates" because while I know the spindle hp (5 & 7.5) and the rotating

3 phase ancillary stuff in the machines 12 (for both) that is only 24.5 hp for both and they require more in order to run the DC servo motors and internal computers & plc's. There are currently 850 uf (370v run caps) in the system, I think I will be looking at about 1550 uf before I am done. Does that sound about right?

Thanks Dave G

Reply to
DAVE GEE

Honestly I have never tried doing something like this, granted the 25 will be completely unloaded but the

30 might still complain. Unless somebody here has done something like this and can chime in, I would have to say "I honestly don't know." It does violate the 1.5 rule but I've seen folks do all kinds of stuff that does work. I do know that older, larger frame motors seem to do a better job at this (starting large loads) than smaller, newer ones.

OK, that settles that issue.

If you believe your 100 amp number then that gives me

32 hp which is the same as your 34 hp number. 1.5 times that is 50 hp so your total of 55 would work out well, and be a tad oversized which is good.

I can give you no suggestions about how much tune and power factor correction capacitance you will need, the published 'uf per hp' numbers should get you close if that's what you are using.

Can you start the 25 off the 30, that's a good question. YOu might consider doing it the other way around, starting the smaller one. If you are right on the edge that will help a bit. Of course the pony motor idea for both of them could work. Mount both idlers on the frame, one on each side, and the pony motor in the middle?

If you had the 25 hp motor there, I would say to just try it, but obviously you don't.

Jim

================================================== please reply to: JRR(zero) at yktvmv (dot) vnet (dot) ibm (dot) com ==================================================

Reply to
jim rozen

Hi again: Yeah, I like that idea, or maybe start the 30 use the 30 to "spin up" the 25 with an idler wheel that would be movable to connect the two shafts then be removable. Picture a 6 or 8" rubber lawn mower type wheel on a pivot with a spring to move it out of contact with the 2 shafts once the motor is up to speed, this ought to minimize the "inrush" of starting current. then "tune" the 25 to get the best "blend" of phase current. I think that is what I will start with because I am pretty sure that it would work and it is certainly the most cost effective thing I can think of. Initial cost will be low and much of the time I will only be running one machine.

I think I may run down and get it tomorrow and try some experiments with it .

thanks much to you, and all who shared their thoughts on this. Rotary 3ph seems kinda arcane to me, but it does work, and work well. I built the 30hp setup from mining old google archives mainly from this group and the woodworking group. My experience is that large systems (above 10hp) are less fussy and produce better, cleaner power, with only a little trial and error involved in the initial tuning (with run caps) for balance. The system has been running a CNC for over a year and has never given me any problems.

again thanks Dave G.

Reply to
DAVE GEE

Since you're running spam-proofed, I won't bother trying to clean it up, but will post this to the newsgroup instead.

What is this number? the 7/10 make sense as month and day, but the final number might be a typo for 2003 (both '2' and '3' being typed as '1' for whatever reason) (except for the 'p'), or it might be some form of article number as seen on *your* system. If the latter, I though that you should be told that the numbers presented by a server are unique to that server, and there are tens of thousands of servers carrying each newsgroup. The number is a function of how long that particular server has been carrying the newsgroup, how many articles have been missed for whatever reason, and (perhaps) how many crashes have partially reset the numbers, since most sites do not fully back up usenet news -- it is too transient to make it worth while.

Or -- it *might* be some kind of threading number applied by supernews -- if you are reading this with a web-based interface. In that case, the same number would only apply to others reading it on supernews as well.

FWIW -- on my news server (I run my own), I see the article to which I'm replying as: article number 464032 -- and I've carried the newsgroup since it started, but have switched to a new server machine at one point (running two in parallel for a while to make sure that it all worked), plus lost my newsfeed for a few days a couple of times, so the number is as useless as all the others on usenet -- just good for identifying the article on this one server for the (relatively) short time until it expires. I run a 30-day expiration on this newsgroup, many sites run three days or less.

If you want a unique identifier for an article, use the "Message-ID: ". Here is the one for your article to which I'm replying.

I won't know what my reply has as a "Message-ID: " until it is sent and the news server assigns the "Message-ID: " to the article.

The disadvantage of using the "Message-ID: " is that most newsreaders don't allow you to easily search on that basis.

Enjoy, DoN.

Reply to
DoN. Nichols

Argh! Missed my chance by *that* much. :-)

Gary

Reply to
Gary Coffman

Hee hee. Pay no mind to it. It wouldn't be the same otherwise, anyway.

"you say to-ma-to, I say to-may-to...."

Jim

================================================== please reply to: JRR(zero) at yktvmv (dot) vnet (dot) ibm (dot) com ==================================================

Reply to
jim rozen

Oh, and btw, *can* that gent start his

25 hp idler, off of a 30 hp one? Best guess and all?

Jim

================================================== please reply to: JRR(zero) at yktvmv (dot) vnet (dot) ibm (dot) com ==================================================

Reply to
jim rozen

Yes, but the starting surge is going to be horrific. I'd recommend a pony motor to spin it up first.

Gary

Reply to
Gary Coffman

PolyTech Forum website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.