Rotary phase coverter - 5hp - queries re capacitor sizes

[Stewey]

Hi Folks,
I made up a rotary phase converter, because the VFD we bought would not drive a star-wound or Y-wound 3phase motor on our drill press or metal lathe.
(The Ebay chinese VFD would only control a delta wound 3 phase motor)
I have a nice professional rotary phase converter, but it was only a 2.2kW version for our cnc router spindle - bought 8 years ago for a small fortune ($2300)
So I made up a 5hp one yesterday.

The 3-phase 5hp motor I picked up had the option of being wired star or delta. All the drawings I've looked at show the idler motor as a star or wye (Y) connection in the drawing, so I set the terminals like that.

The general rule of thumb was about 60 uF per hp, so I put a 300 microfarad starter capacitor in, with a momentary 'on' switch that you held down till the idler was up to speed...
I put 2 x 70 uF in parallel = 140 uF as a run capacitor across the same phase as the starter circuit. The other line had no capacitors.
I put a 100k 1/2W resister across each capacitor as a 'bleed' to discharge them when it was off.

When I went to start it yesterday, the idler motor spun up to speed very quickly, and was actually slower and worse when I used the starter capacitor.
I'm wondering why...
The motor got rather warm also...
and one bleed resistor burned out to open circuit, the other shorted down to 10 ohms - in the run capacitors. Maybe I should have used 2-watt ones instead of half-watt resistors...

However, my query is that when I put the multimeter across the 3 phases while running the idler, I got L1-2 = 245Vac(mains), L2-3 = 123Vac (half mains) , L1-3 = 365Vac (very high) - at idle

When I switched in another 3hp motor to help balance the phases better, then switched in the lathe, and then turned the 3hp helper off, the line voltages are now 246,(mains line in) and L2-3 which had been 123 volts was now 376 Vac when the extra motor was running, and was 407 Vac when I turned the 3hp extra balancer motor off.
(I didnt measure the other phase L1-3)

I realise the varying loads will alter the induced voltages, but I'd been reading where others had tweaked the capacitor sizes to get up to 5% variation between phases.
I have variations of up to 100% between phases at idle, and some darned high figures under load...

Wondering if maybe I should have set the idler motor on delta configuration instead of star, for our Aussie 240Vac mains line in L1-2 ?
and why the momentary starter capacitor works best if I do not use it at all? The idler gets up to speed in 2 seconds or less without it being momentarily pressed.
and why the bleed resistors burned out?
and If I should have a smaller capacitor across the other induced phase line

I'm no electrical expert, (or I would not be asking here), nor a total fool either...(I don't think!)
Thanks!

Any advice or suggestions appreciated!

Similar Threads:

• Robodrill 7.5hp spindle. Single Phase Power input No Rotary Phase convetor doable?
• driving a single phase capacitor start/run motor from three phase vfd or power grid.
• 3-Phase Quality From Rotary Phase Converter
• Pros and cons of rotary table sizes: 6" vs 8"
• Need Help!- what size 3 phase coverter haas vf-2

[Stewey]

No replies yet... oh well.
I swapped the terminals in the idler motor, from star to delta connection, and retested it.

It booted up to speed in a second - again no need for the start capacitor.
The idler ran a bit more quietly.

Line voltages with only the idler running were 455, 236 385. 236 is the line in - our mains AC power, nominal 240Vac
When I turned on the 3hp mortising machine, it brought them closer together: 390, 243, 350.
When I then switched on the metal lathe as well. they got a bit better: 386, 244, 333
When I then turned the other motors off and left the lathe running alone (plus the idler), 350, 244, 362
and it now starts and runs the chain mortiser as well - while before, with the idler in star or Y, it would not.

Of course all safety procedures are in place, and we have earth-leakage circuit breakers on the mains coming in, as well as a separate one on the line to this machine.
It is unplugged in between changing things etc.
Disclaimer - these notes are for academic study and analytical purposes only - not instruction.

[handlewanker]

Hi....I went down that path too some years ago with a Bridgeport 1.5HP motor......built a rotary as well but experienced the voltage unbalance when off/on load.......eventually bought a professionally built one that adjusts the capacitor bank electronically according to the loading, and it's as silent as a tomb.
Ian.

[Stewey]

Thanks, Ian, I'm an Ian also...!
I bought a rotary converter in a box - the size of an esky, about 8 years ago $2300 from a mob in Bendigo... 3hp capacity. (2.2kW)
It is dead quiet, but it never performed too brilliantly - and If I wanted to start the 3hp drill press, it would have hissy fits with the magnetic contacts in the starter button, and the 3hp wood lathe was a struggle too, even under no load and belts off. The machine would trip out under 'overload' to get started, unless I manually gave it as spin then ran back and pressed the on button a few times, to get it up to speed.
I bought that RPC to run a 3hp cnc spindle, which it does, but you have to take that gently - so I leave it for that task alone, and don't try and use if for other 'cruder' or less sensitive tasks.
We have a 20kva genset, and needed to run that for any other 3 phase needs - but running out of diesel in the middle of a job is not fun, hence the idea of making a rotary converter...

Thanks for the reply!

[handlewanker]

Thanks, Ian, I'm an Ian also...!
I bought a rotary converter in a box - the size of an esky, about 8 years ago $2300 from a mob in Bendigo... 3hp capacity. (2.2kW)
It is dead quiet, but it never performed too brilliantly - and If I wanted to start the 3hp drill press, it would have hissy fits with the magnetic contacts in the starter button, and the 3hp wood lathe was a struggle too, even under no load and belts off. The machine would trip out under 'overload' to get started, unless I manually gave it as spin then ran back and pressed the on button a few times, to get it up to speed.
I bought that RPC to run a 3hp cnc spindle, which it does, but you have to take that gently - so I leave it for that task alone, and don't try and use if for other 'cruder' or less sensitive tasks.
We have a 20kva genset, and needed to run that for any other 3 phase needs - but running out of diesel in the middle of a job is not fun, hence the idea of making a rotary converter...

Thanks for the reply!

Hmmmmm..........back in the late 90-'s I knew an Ian Stewart who was the secretary of the Melbourne Society Of Model & Experimental Engineers club when I attended the club nights at the Glen Iris AMRA club site......coincidence?
Ian.

[sajurcaju]

This is the best reference I know of:
Fitch Williams Converter Design & Balancing Instructions
Fitch knows what he is about.
Note that the cap isn't all across one phase.

Best of luck,
Steve

[handlewanker]

This is the best reference I know of:
Fitch Williams Converter Design & Balancing Instructions
Fitch knows what he is about.
Note that the cap isn't all across one phase.

Best of luck,
Steve

Hi, it isn't just balancing the voltages between the generated phases that is hard, it's the variable factor when a load fluctuates........the circuitry that does this is no DIY capable build.

Anyone with a bit of electrical knowledge and a multimeter can adjust the capacitors to get phase voltage balance........for one occasion only....... as when the converter is idling or on load, but you can't have it both ways with a simple DIY set up.
Ian.

[sajurcaju]

I should have remembered this problem; for some reason I am unable to post links. Anyone know why?

Attached is the pdf itself on the Fitch approach. Hopefully this will go through.

Handlewanker, if the converter starts out of balance it will get a lot worse when the load changes. If it is in balance it will only change a little with load.
Stewey--prove it!

Steve

[handlewanker]

Hi....DIY rotary converters have always had this problem of the inability to adjust the capacitors and hence the voltages generated for load variance......for a constant loading they work very well.......a mill or lathe drive is constantly varying and you cannot adjust the capacitor values manually while they are in circuit, one reason why they cost around $3K for a 2 HP commercially made model.

If you have the ability and knowledge to design a load sensing 3 phase circuit, then a DIY rotary is feasible.
Ian.

[sajurcaju]

Ian, you are ignoring all the people who have their mills and lathes running fine on fix-tuned rotary converters.

Steve

[handlewanker]

Whatever you say.....I have learned otherwise and chose not to go down that path.
Ian.

[Stewey]

Thanks, Ian, I'll just stick with the unit the way it is for now... it beats starting up the genset & burning diesel every time I need the lathe.
The expensive computer-controlled rotary converter I bought 8 years ago will remain just for use with the cnc spindle.

Sajurcaju, thanks for the PDF, I'll enjoy reading it. But the voltages will of course fluctuate depending on the load, which by nature will vary.
The more motors I switch on at the same time, the better the phases balance and the more quietly and cooler everything runs, due to induction...
I'll add another slave motor to the line...

Over and out!

[Stewey]

Here's an update...
It worked tolerably, the idler being rather noisy till other motors to idle and balance things were switched in - that was the key...
I decided to gut it all, and go for Mark2. I bought some line current and voltage gauges from Evilbay, and an E-stop switch, and some neutral links and set it up more logically.
I put a meter on the incoming power line. Then a switch.
That puts power to the idler motor (5hp)
I had a 300uF momentary on switch and capacitor (450v rated) but never needed it.
I had 2 of 70 uF run capacitors in L1-L3. It got up to speed within 2 seconds.

I put line meters on the output lines to the 3 phase socket.
Reading the above theory on the 10kp Fitch system, and others, I decided on another capacitor of 40% the run cap, across L2-L3, and 10% across L1-L2 as power factor assistance.

At any rate the voltages at startup were pretty horrible. incoming was 240volts, and current 12 amps as the idler groaned away. L2-3 was around 420v, and L3-1 was 380v. (L1-2 was same as supply)
I put a magnetic starter and another 3 phase idler in the system.
Starting that helped tremendously, and brought the two high line voltages down to the mid to upper 300v range.

I then started to try and tweak the capacitors more.
Strangely, when I took the L1-2 capacitor (first try 12uF, 2nd try 8uF) out, the voltages were better with no power factor cap in L1-2.
I then pulled off the L2-3 capacitor which was about 80uF, and used a 40.. Better. Tried 25. better again. Tried none - best of all.
I then went back to the run caps - 140uF - (2 of 70 uF.) and took one off. It would not start, with or without the starter cap.
So I added in assorted others in parallel, dropping the total from 140, to 120, to 107, to 95, to 12, and it would not start at all.
Each time I lowered the total capacitance in the run cap, it improved the voltage differences..
I ended up settling on a 70 + 25 uF = 95uF for the run cap. And no other capacitors at all, but two idler motors.

That gave me, when I started and ran the lathe + frill press, voltages of 240 incoming, and current 4.5amps, incoming,
and line voltages of 240/285/282 and line currents of 1.7, 2.1, 3.5.

With the loads turned off and just the one idlers running, the incoming current was up over 10 amps.
Running the 2nd idler brought that down to 8 amps draw.
______________________
No matter how I tweaked the capacitors, I couldn't get the L2-3 or L3-1 voltages down closer to 240. Altering the capacitance, and putting caps in other lines only left the voltages higher, and the mains current drain also higher.
I'm content with it as is, but at some stage I might add another idler.
It would not start the magnetic starters of the chain mortising machine, but I'm not overly concerned. Oddly, it did start that when I had the other capacitors which I removed - connected in.

This update is purely for academic purposes... and the incoming line is protected by earth leakage circuit breakers.
It was an interesting exercise.

No matter what I did, the run capacitor (300uF) was never of any use- never needed.

[Stewey]

Further update - I added another idler motor. And the inline voltage & current meters for each line. And better switch gear.
It runs beautifully, quietly, cool, and starts most things. I have an old wood lathe that the magnetic starter for won't energise, and a 3hp chain mortiser - with the same issues, that's all.
But the rest of the equipment goes very well.

[hemipower]

Good point on the workings of a converter at idle or under load been; down that road. This might be a good view of the load sensing circuit. =