Electrician Talk banner

Rotary Phase Converter - Small Motor

9K views 51 replies 21 participants last post by  acro 
#1 ·
I have a customer with a small welding shop. No access to 3 Phase power, so he has a rotary phase converter which he uses to power a few pieces of equipment. The phase converter has a 20HP idler and multiple run caps.



One of these is a small twisting machine he purchased a couple years ago to twist square bar for ornate gates/railings etc. The twister utilizes a small 3P motor with a FLA of 3.7A @ 220VAC.


Apparently a couple of weeks ago he attempted to use it, and one or more of the windings shorted out.


The company he purchased the unit from came and pulled the motor, rewound it (or had it rewound), and reinstalled. Their tech said it failed due to a power issue at the shop, causing the motor to draw too much current.


I was forwarded an email from the tech which stated that when the motor was bench tested (unloaded) at their shop it drew ~2.5A @ 212V, yet when connected on location it drew 4A @ 244V.


I stopped by on Friday on my way to another site, expecting to find a failure with the phase converter that was causing the motor to single phase. However, that doesn't seem to be the case.


I did some quick tests as I was short on time, and found some strange variations with the voltage and current between phases. I expected the higher voltage reading between A and C, due to the phase converter being so lightly loaded. I didn't expect the values between B and C, nor the current being so high and varied.


No load was the motor pulled from the transmission. Base load is the motor connected to the transmission. Light load is twisting some 1/2" square bar. Light load 2 is twisting the same size bar, but this time with an additional 9A load on the phase converter due to another piece of equipment running at the same time.
 

Attachments

See less See more
1
#3 ·
Voltages look normal. Not any kind of expert either but I would start by measuring the caps to see if any are going out of spec.

Wealth of knowledge available here:

Electram Rotary Equipment Ltd.
18112 - 107 Ave, Edmonton, AB, Canada

Phone: 780-451-0418
 
#4 ·
Voltages look normal. Not any kind of expert either but I would start by measuring the caps to see if any are going out of spec.

Wealth of knowledge available here:

Electram Rotary Equipment Ltd.
18112 - 107 Ave, Edmonton, AB, Canada

Phone: 780-451-0418

I was going to call Complete Electric Motor Repair in Calgary on Monday, they've helped me out before when I had motor questions. Thanks for the info on Electram, always good to have.


All the caps have been replaced recently. The run caps are sized for the other (much larger) equipment, so I'm not expecting them to be applicable for this little motor.


The only voltage reading that seems curious to me is between B (L2) and C. Should it also not read high like A to C does? Or is that because the caps are over compensating the phase angle under such light load?


I find it strange that the current above FLA, even with the motor just sitting on the bench.
 
#8 ·
Yes, the new caps are identical to the old ones. I metered a couple of the other pieces of equipment with the phase converter running, and they test fine. I don't remember off hand what the brand is for the phase converter.


i just looked at some notes I took of a similar setup only the converter is feeding a group of motors. Funny thing: AC voltage on my setup was lower than AB/BC loaded or unloaded.

Which I guess is similar to what I'm seeing, with the difference being which line we're calling A and B.


Why not use a 350 dollar vfd. Single in 3 phase out.

Sure, but I'm trying to understand what is happening here. I'm not one to say "I know you just spend $600 having the motor rewound, but lets throw another grand at it and see what happens".



Is it a compatibility issue due to the size of the idler and run caps that is causing the motor to draw up to 120% FLA with no load? With so much current variation between phases?



Did they screw up the rewinding?
 
#32 ·
To follow up a bit on the RPC output and theory, I point to a drawing from Peter Hass, that is a vector drawing that better illustrates the concept.


When you look at it, keep in mind that most US rotary phase converters are built with 230/460 Wye connected motors. Hi voltage is connected in a single Y. Low voltage is connected in a YY. And normally there is no connection available to the common of the Y connection.






It is here, that you can visualize the difference between the 1Φ grounded center tap, and the center of the Y connected stator windings.

And that they are 69V apart.


For a more through explanation, than I have given, take a look at Peters explanation in these two threads.

Nobody does it better, that I have seen.

https://www.practicalmachinist.com/...ree-phase-vs-single-phase-illustrated-102416/


https://www.practicalmachinist.com/...n-rpcs-their-capabilities-195352/#post1265580
That diagram makes it so much easier to understand :vs_laugh:
 
#7 ·
Electram specializes in phase converters. Ask for Ron Abma.

i just looked at some notes I took of a similar setup only the converter is feeding a group of motors. Funny thing: AC voltage on my setup was lower than AB/BC loaded or unloaded.

Capacitors were replace like for like?
 
#9 ·
I don't know how it works. I've never really thought about it. But I know an electrician that had a problem with small three phase motors running weird on a phase converter. It was discovered that things ran better with a minimum load greater than the small motor. In other words, if they first loaded the circuit with a larger grinder motor, the small motor ran fine.


Could it be this converter is behaving badly with such a small load?
 
#10 ·
As seen in my included measurement spreadsheet on my first post, it definitely evens out the current a bit with additional load on the phase converter, which definitely makes sense to me.


However, the motor current is still way over the spec plate FLA, even at a higher than rated voltage (244 vs 220).
 
#11 ·
Many RPC’s use caps on one side only, as in from one single phase leg to the manufactured leg. This can introduce a bit of imbalance in voltage and phase shift. If the RPC consistently runs lightly loaded, take out a couple caps and lower the voltage on the manufactured leg. I have a plant that runs entirely on a 60 HP unit and it has a pump load of 10 HP constantly and it had a way high voltage on the manufactured leg of 275 and 280 volts. I took out a couple caps and got the voltage down, then put some back in on the conveyor and air compressor that switch in with the motor starter. It makes for a much quieter and cooler running system that runs with a balanced voltage.
Lastly, do they have 208 motors by chance? They will run hot on 240 either way.
 
#12 ·
Many RPC’s use caps on one side only, as in from one single phase leg to the manufactured leg. This can introduce a bit of imbalance in voltage and phase shift. If the RPC consistently runs lightly loaded, take out a couple caps and lower the voltage on the manufactured leg. I have a plant that runs entirely on a 60 HP unit and it has a pump load of 10 HP constantly and it had a way high voltage on the manufactured leg of 275 and 280 volts. I took out a couple caps and got the voltage down, then put some back in on the conveyor and air compressor that switch in with the motor starter. It makes for a much quieter and cooler running system that runs with a balanced voltage.
Lastly, do they have 208 motors by chance? They will run hot on 240 either way.

I always had in my head that running 220/240V motors on 208 would cause them to run higher current and therefore hot, not the other way around? In this case it is a 220V motor running on 244V, would that not result in less motor current?



I can see that running the motor on a phase converter requires it to be de-rated due to the less than ideal phase angles, but this one won't even freewheel without being over the FLA on two of the phases.
 
#14 ·
Ya, this one works great for the loads it was originally installed for. It's this little motor that is piggybacking off of the system that is having trouble.

I'm unsure if the fact that the motor is running over FLA with no load is due to the phase converter, or if the problem lies elsewhere.

Unfortunately I have no point of reference, as this piece of equipment was purchased and installed a couple years ago, and for all I know has been running over its rated specs the whole time. And of course, no overloads installed.

Sent from my SM-G960W using Tapatalk
 
#20 ·
At undervoltage the motor generates a lot less torque at a given current so current goes way up because it is overloaded. At minor overvoltage just the opposite happens and current drops. But eventually the excessive flux increases the current even though torque current decreases.

But in this case we have a much bigger problem. Every 1% imbalance in voltage causes a 6-8% current imbalance. The current though is out of balance so it just contributes to overheating the motor. At more than 1% voltage imbalance we derate. This curve shows how much to derate.


https://duckduckgo.com/?q=nema+volt...ndustrial-electronics.com/images/epe_10-14.jp

So I’m not surprised at all.
 
#23 ·
I can't take time to read all the other reply's, but I did read the op . When you try to read between phases on the load side of a rotary phase converter, you gonna get garbage readings on the derived phase to the other two phases. Normally those converters use C for the derived phase. If you are good between A and B don't be worried if either AC or BC looks off, it's normal for phase converters to do that chit. As long as you have 120 to ground on C you should be ok.
 
  • Like
Reactions: Mobius87
#27 ·
I can't take time to read all the other reply's, but I did read the op . When you try to read between phases on the load side of a rotary phase converter, you gonna get garbage readings on the derived phase to the other two phases. Normally those converters use C for the derived phase. If you are good between A and B don't be worried if either AC or BC looks off, it's normal for phase converters to do that chit. As long as you have 120 to ground on C you should be ok.

This is not correct, a rotary phase converter has an output that is similar to a center tapped delta. There is no neutral, but when measured to ground, the generated leg should measure at least 208V to ground. Typically, if measured when unloaded the reading will be higher. Should head lower once load is applied to the output.
 
#26 ·
Rotary Phase Converter Loads And Balance

A lot of the good points have been mentioned already, but I'll add a few more.

Rotary phase converters can only be well balanced at one point of load. The balance of voltage and current is load dependent. Being you have a 20HP idler, it's likely balanced for some load about 10HP or higher, so that it can support the load capacity it was built for.

When it runs loads outside of its, tuned point, things become more and more unbalanced. You didn't mention what size his bigger loads are, but this is likely where his RPC is sized and balanced for, and it works fine tuned for that load. Running a very small load as compared to the idler motor size, leaves things very unbalanced.


It was mentioned to move some of the caps to the larger motor loads. This can help, but it takes a fair bit of trial and testing to figure out what to move where, depending on the load sizes, and what runs when, or are the loads run one at a time or have a duty cycle. The more random the load, the harder it is to hit that moving balance point.


Your converter is 20X the size of your small motor, so you have a severe mismatch in size for any reasonable balance voltage and current. Like stated earlier, a small voltage imbalance produces a much greater current imbalance. The current is what heats and cooks the winding's.


On a RPC, the load motors should never be expected to produce the maximum rated HP. It might be likely, that the bar twister was designed that way to keep costs as low as possible. Then your feeding it voltage that is higher than your typical +/- 10% for 1.15 service factor motors. Then you have the import factor, they are known for not being the most robust.


The other things not mentioned is the phase angles shifting under different loads, this creates more imbalances. Then you have the PF (power factor), it is normally very low when the RPC is unloaded, and gets better when the load increases. With a small motor as a load, the PF stays very low.


Much of the current you will read with an amp clamp meter is reactive. It flows back and forth between the capacitor bank and the inductive motor windings, it don't come from the meter, but the capacitor bank. Hence moving some of the bank to the largest motor can help, remove some of the reactive current from the lines.


Mitigation strategies for the small motor load to prevent overheating.


Run other large 3Φ motors concurrently, to keep the system better balanced, where it's designed for.

Get a bigger and better motor so that it's not running at more than 5% of its voltage rating, and sized to not exceed 60-70% of it's HP rating, to supply the given mechanical load.

Use a 1Φ motor and skip the RPC supply, provide some overload protection, in either case to prevent failure.

Lastly a VFD 1Φ to 3Φ could be used.


For 1HP bar twister load, I would be thinking a 1Φ motor, with built in overload protector, and not mess with the RPC balance that has been working fine for the customers other larger loads.
 
#30 ·
The 3Φ output, at the idler motor connection box, would be an ungrounded delta.
The idler motor stator windings are insulated from the frame, and have no bonded connection to earth or neutral.

The reference to ground voltage, you do read on the output of the phase converter, is due to the single phase supply service, being center tapped grounded.
The neutral to ground bond at the service is what provides the reference to ground voltage, hence this is why it gives voltage readings similar to a center tapped delta.
 
#31 · (Edited)
Rotary Phase Converter Theory of Output

To follow up a bit on the RPC output and theory, I point to a drawing from Peter Hass, that is a vector drawing that better illustrates the concept.


When you look at it, keep in mind that most US rotary phase converters are built with 230/460 Wye connected motors. Hi voltage is connected in a single Y. Low voltage is connected in a YY. And normally there is no connection available to the common of the Y connection.






It is here, that you can visualize the difference between the 1Φ grounded center tap, and the center of the Y connected stator windings.

And that they are 69V apart.


For a more through explanation, than I have given, take a look at Peters explanation in these two threads.

Nobody does it better, that I have seen.

Three-Phase vs. Single-Phase, Illustrated

Another Explanation of RPCs, and Their Capabilities
 
#33 ·
Update on this:


Pulled the motor and sent it away to be checked and rewound to 250V, with a new FLA of 3.1A.


I installed the rewound motor today, and before I connected it to the gearbox I checked the current and voltage. 3A/2.5A/2.5A at 244V/244V/253V. Still way too high for a freewheeling motor, but better than before.


I swapped the phases around to see if that imbalance would follow the phase, and it does.



Now for kicks I bolted it back to the gearbox and tried it again (no load). It now won't even get up to speed. 25A+ on all phases, with the voltage dropping to 235V/233V/222V.

I called the motor shop back and was told that it's part of their policy to check for a cracked rotor, and that it would have been bench tested at 230V (they don't have 250V available). He didn't have the results of the bench test handy.


Not sure where to go from here with this thing?
 
#34 ·
It sounds like the reducer is misaligned with the motor causing the rotor to rub the stator. The voltage drop sounds a little high but with 25+ amp draw on a likely #14 wire who knows how long fed by a DP contactor well, it probably is on target. See if running the motor with the bolts loosened on the reducer helps in any way.
 
#35 ·
Rotary phase converters are yesterday's solution.

As this thread shows, they're not worth the hassle.

If anything goes sideways -- caps// windings// load balance -- the EC ends up chasing his tail -- without pay, I should add.

All of these dinky motor loads should have their own VFD -- with an isolation transformer if necessary because the load is strangely sensitive to harmonics -- a very rare event, BTW. Motors will never need such a filter... just their controls... if even.
 
#36 ·
For a dirty fab/welding/machine shop, especially one with numerous motor loads, a RPC makes plenty of sense. All those VFD’s, maybe a dozen or so in such a shop doesn’t make good economic sense. The issue with RPC’s that I have stated numerous times is that the larger sizes are not plug and play right out of the box. They need a little tuning to get them right, then they are generally good for say 10-15 years or so, just have to listen to them hum away.
 
#38 ·
While I don't have complete understanding of motors, Stallcup's motor book gives extreme amperage imbalance when three phase voltage is out of balance.

Maybe this is unrelated. An industrial facility has a capacitor system quite primative. A big capacitor system is on line all the time. A secondary group of capacitors are switched by power factor need automatically.

On Sundays when nothing is on except lighting, voltage spikes way higher than I want it to. T5 fluorescent lighting fails far too often.

Rotary phase converters are full of capacitors sized for a consistent load. Could the motor shop resize capacitor values to correct voltage imbalance?
 
#39 ·
While I don't have complete understanding of motors, Stallcup's motor book gives extreme amperage imbalance when three phase voltage is out of balance.

Maybe this is unrelated. An industrial facility has a capacitor system quite primative. A big capacitor system is on line all the time. A secondary group of capacitors are switched by power factor need automatically.

On Sundays when nothing is on except lighting, voltage spikes way higher than I want it to. T5 fluorescent lighting fails far too often.

Rotary phase converters are full of capacitors sized for a consistent load. Could the motor shop resize capacitor values to correct voltage imbalance?

The phase converter capacitors are sized for the other, larger 3 phase equipment. When they bought this equipment it was assumed they could just tap into the existing power and run it when they had the occasional need. Which they did, until it fried.
 
#41 ·
As was mentioned (by CMP and others), going to have to switch gears on this one. Likely with a VFD, easier to set up for reversing than fitting and rewiring for a single phase motor.


Although now that it has been rewound to be closer to the shop voltage, I don't understand why this motor nearly pulls FLA still when it is freewheeling unless there is something wrong with it that the motor shop didn't catch.
 
#42 · (Edited)
A few things I see
220v motor @ 60hz. seems odd even if it is from China. I have seen this with European machines before that they change the motor to 60 hz, BUT did not resize the loads for faster speed. example: Blower is designed for 2800 RPM, Install a 60 Hz motor on same blower it is running at 3400 RPM. Curve shows not enough HP at that speed.

At the cost of a Small VFD I would not waste any more time, it will not depend on other stuff running to add load. You can also adjust the v/Hz ratio if needed if they did use a 50 hz motor and just remarked it. I know you said it was rewound but they may of used data from the original motor that may of not been correct.

Have your customer stay away from junk machines, we bought straw machines that were a Chinese knock off of a Chinese knock of of a Chinese machine.

Cowboy
 
#43 ·
Update:


Still fighting with this machine. As mentioned before the motor ended up being rewound twice. I took it back to the motor shop a third time and they went over it in detail and found no issue.


Wired it up to a VFD and it ran without issue on the bench. Connected to the gearbox it won't get up to speed and pulls 25A+, regardless of being connected to the RPC or the VFD.


While testing it using the VFD, suddenly I hear the speed start to increase but it trips out on overload before it reaches full speed. (not the VFD settings, ramp was set to 0.1S).


I reset and try it again. This time it does make it up to speed almost immediately, but the motor is still pulling 3x the rated FLA and trips out on OL.


At this point I'm convinced it is a gearbox problem, not an electrical/motor problem. This would explain why it worked fine before it fried the motor the first time (which started these shenanigans, no OL present for the motor from the factory either. Was advertised to my client as plug and play.)


Anyway, we pull the motor again. The gearbox still turns easily by hand. We decide to pull the gearbox as well (that was a heavy/awkward bitch). It has a sight glass on the back (not visible when installed) with no oil visible and the data plate tells us it is a cycloidial reducer.


We crack it open, and it appears to be grease lubricated rather than oil. Either that or the oil has turned to a grease like consistency. It appears to be otherwise pristine inside.


Anyone here have experience with cycloidial reducers, and could shed some light on whether this is the culprit or not?
 
#45 ·
It's been a while since I saw this thread and wondered if you found a solution? We have a worm gear reducer on a radial stacker that gave the guys fits in the winter months tripping the overloads. I saw how thick the oil/grease was, and got out all I could then filled it back up with ATF, no more nuisance trips after that. There's a good chance the churning friction of too thick a viscosity oil is the culprit.
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top