What's new
What's new

Big RPC stalling out on pony start

Strostkovy

Titanium
Joined
Oct 29, 2017
I have an RPC I built that has a 10hp motor and a 40hp motor started by a 5hp pony. The 10hp idler starts up without issue and I use it to power a 6.5HP brake just fine.

I only now got around to wiring the 40hp motor in and even though the pony brings it to very near synchronous speed it suddenly begins to stall out. When I first setup the smaller idler I wired it in a way the reconnected the start capacitors when the motor was shut off, which resulted in DC injection braking, and the big motor sounds exactly like that by louder.

I can't take a picture of the nameplate because it is obstructed but the important bits are as follows:
230/460 V
102-92.2/46.1 FLA
SF 1.15
PF 0.89
NEMA EFF 91.7%

The pony is a 5hp 230V 3450RPM compressor motor, connected by a 1:1 pulley arrangement.

The line impedence at the panel is 0.23 ohms, so my electrical service has a hard time delivering large surge currents.

My first attempt to start it was by disconnected the start capacitors but leaving the run capacitors connected. I started the pony, then the big idler, and it immediately slowed down. Power draw was around 230 Amps but it's hard to actually tell because the meters update slowly. The goal was to then bring on the small idler with no start capacitors and check rotation before reconnecting the capacitors, but that didn't happen.

Second attempt was like attempt one but with no capacitors.

Attempt three was to start the 10hp idler first, then blip on the big idler from standstill to find the direction of rotation. I swapped polarity so that the direction matched, then started the small idler, the pony, and then the large idler after it was up to speed. The voltage dropped low enough that some contactors dropped out and it stopped before I did.

It seems like the motor would start from the 10hp idler alone, with no pony, but that would take maybe 20 seconds at 250A which obviously isn't feasible.


My best guess is that this motor has very low resistance, especially relative to my high line impedance, which doesn't allow me to give it the inrush current it requires to establish the magnetic field in the rotor properly, and is likely just inducing an AC current into the rotor that often opposes the field windings.

If that is the case then starting the small idler first, then starting the pony and large idler at the same time should allow it to start, but I just don't have the power for that.

The thing I would try next would maybe be to put some beefy resistors from the pony motor contactor to feed a limited amount of power into the windings of the big idler to allow the magnetic field to form in the rotor as it was brought up to speed.

I checked and rechecked and then rechecked the wiring to the big idler and I'm positive I wired it to the correct voltage and in the correct way. It is delta only, no wye start.

The motor is too heavy to move somewhere with real three phase. It does turn over with the smaller RPC power alone, but the generated leg is too weak to bring it up to speed fast enough before burning up everything.

jogbw6r.jpg

Muqvhmf.jpg

Hs02kIo.jpg
 
Last spring I was setting up a 10 HP idler to start with a pony motor. I had a spare VFD available, and so used it for the pony motor. I anticipated having a lower inrush current due to using the pony motor.

What I found was that as I adjusted the VFD to bring the idler motor to synchronous speed, the inrush current had increasing variability- both high and low. The low was 11.4 amps, and the high was 173.5, and the average was 93.4.
I had meters set up to look at the voltage between the phases- and eventually noticed that I was generating a significant voltage between all the legs BEFORE I connected the single phase line. My idler was acting as a generator, and when the single phase current connected the phases sometimes matched, but often did not.

The initial set up was with a belt drive, with mismatched pullys. To hit 1800 rpm on the idler, I needed to set the VFD hz to 65. I later set it up with a chain drive with a 1:1 ratio thinking I would use a single phase pony motor, and could "index" the idler to the single phase waveform powering the pony motor. This did not work, as they ran at a lower rpm. I next tried going back to a 3 phase pony motor, and running it off of a 3 HP RPC I already had. This did not work either. I found that the pony motor would bring the rpm up- but when the idler started acing as a generator, the rpm dropped...

I tried to find a way to use the generator effect to consistently reduce the inrush current, but do not know enough to devise a way to make the connection at the right time.

I would suggest you try measuring the idlers voltage when spun by the pony motor and see what you find.

From what you have written, you have more electrical smarts than I do. If you can come up with a way to consistently reduce the inrush current it would be a great help for those of you who need the large idlers on a limited electrical service.
 
Do you have a full, not wished-for, 200A service available to it, and if so over what conductors?

BTW what is the 40 HP idler's RPM?

Neat guarding idea on the tower.

As to "too heavy to move" Belt it to a hand truck or barrel dolly. Should then be able to tilt it, slide it up a ramp into pickup or van, take it to where you can "borrow" utility-mains 3-Phase and vet THAT much of it, yah?
 
Last spring I was setting up a 10 HP idler to start with a pony motor. I had a spare VFD available, and so used it for the pony motor. I anticipated having a lower inrush current due to using the pony motor.

What I found was that as I adjusted the VFD to bring the idler motor to synchronous speed, the inrush current had increasing variability- both high and low. The low was 11.4 amps, and the high was 173.5, and the average was 93.4.
I had meters set up to look at the voltage between the phases- and eventually noticed that I was generating a significant voltage between all the legs BEFORE I connected the single phase line. My idler was acting as a generator, and when the single phase current connected the phases sometimes matched, but often did not.

The initial set up was with a belt drive, with mismatched pullys. To hit 1800 rpm on the idler, I needed to set the VFD hz to 65. I later set it up with a chain drive with a 1:1 ratio thinking I would use a single phase pony motor, and could "index" the idler to the single phase waveform powering the pony motor. This did not work, as they ran at a lower rpm. I next tried going back to a 3 phase pony motor, and running it off of a 3 HP RPC I already had. This did not work either. I found that the pony motor would bring the rpm up- but when the idler started acing as a generator, the rpm dropped...

I tried to find a way to use the generator effect to consistently reduce the inrush current, but do not know enough to devise a way to make the connection at the right time.

I would suggest you try measuring the idlers voltage when spun by the pony motor and see what you find.

From what you have written, you have more electrical smarts than I do. If you can come up with a way to consistently reduce the inrush current it would be a great help for those of you who need the large idlers on a limited electrical service.

That's interesting. Perhaps the generated phases are due to some amount of magnetization of the rotor? I didn't do many tests, but judging by the lights the inrush current varied greatly when I kicked power to the idler.

I actually already have meters between the generated leg and each mains connection, and between the generated leg and neutral, but they can only have power with at least one idler running. I'll just use a multimeter I suppose.

If this is the case though then my two options are to attempt to degauss the motor prior to startup, feed in a "primer" signal (just a little voltage to align the field) or attempt to synchronize the two waveforms. Phase velocity may be too high to do that in any meaningful way.
 
Do you have a full, not wished-for, 200A service available to it, and if so over what conductors?

BTW what is the 40 HP idler's RPM?

Neat guarding idea on the tower.

As to "too heavy to move" Belt it to a hand truck or barrel dolly. Should then be able to tilt it, slide it up a ramp into pickup or van, take it to where you can "borrow" utility-mains 3-Phase and vet THAT much of it, yah?

I have a wished for 200A service (won't blow the transformer, at least) fed by 2 gauge copper THW from the pole to the service entry panel (150' or so), and then 4/0 Aluminum another 250 feet to the shop.

I measured the voltage drop with 50 amps of load at the point just before the converter and calculated that a dead short would only provide 1000 Amps. (0.23 ohms)

Idler RPM is 3530. That's the highest full load speed of any two pole induction motor I've ever seen.

I do have a scope and I may use that, but attempting to align the field in the rotor through electrical means is going to be my first attempt. (Beefy resistors feeding the motor a wee bit of power from the idler circuit should *hopefully* be enough to get the eddy currents flowing)

The tower is kind of weird. Half of it is guarding due to kids wandering around in the shop often, half of it is afterthought structure, and the whole shape is weird but it takes up very little floor space. It weighs around 550 pounds and while it won't tip over you definitely can't get it upright with only two people. Also my license is on medical suspension, so there's that.


It's worth noting that this has some serious braking torque. Unpowered it takes about 15 seconds to stop. Powered it stops in about 2 seconds.


The purpose of this is to power a 30hp timesaver. Machine load will be much lower than that though once running.
 
Now another question: If I were switching power on completely 180 degrees out of phase with the generated voltage from the windings, would it cause it to stall?

It just seems to me like there would be massive inrush current and a shudder but I wouldn't expect it to do the DC-injection-braking-like thing it's doing. Unless of course my pathetic line service can't even fight out the permanent magnet effect in the rotor before it has moved away.


A thought: I know they partially magnetize rotors on stepper motors to improve performance, even though fundamentally it's not required. (That's why they are often called hybrid steppers). Do they do the same thing with induction motors? I would expect that would give higher starting torque with less rotor losses, and this motor is designed for high inertial load (pump) and abnormally high (in my limited opinion) efficiency.
 
I have a wished for 200A service (won't blow the transformer, at least) fed by 2 gauge copper THW from the pole to the service entry panel (150' or so), and then 4/0 Aluminum another 250 feet to the shop.
I take it you meant 2-ought as I use #2 for my 10-15 KVA gen set at about 50-60 feet,
..

Resistors that are affordable and not fussy about heat sinks are to be found as electric range replacement elements. Paralleled in your case, probably. My largest is a 9 KW "canning" element.

My SE .. 200A residential, about 60 - 90 feet of buried 4-ought Al to the powerco transformer pad is midway garage/shop wall, so 15 feet to the RPC or P-P. Right decent, actually.

Even so, I decided to not even try a 20 HP start off that. Shop isn't my only load in an "all electric" home!

How hard would it be to find the 40 HP a new home?

I'd want to do this with staggered start of a 20 HP primary, then another 20 HP, or two tens or a 5 and 15 or some such combination as supplementary idlers instead of that BFBI hammer of a 40 HP starting load on the service and long drops you are stuck with, pony, VFD, or any other start.

Timesaver starting demands are going to be ....what?
 
Using your numbers, that 230A draw was dropping your "220V" by a bit over 50V.

The motor if started across the line might pull 3x that, maybe more, with attendant voltage drop.

Most soft start systems only cut the current to 30%, so you would be back to the 50V drop.

If you DO get it started, it will pull about 40A on idle, which sets you at about 10V drop just from that.

It sounds to me as if you just have too much motor there. The description is of exactly what yoke u get when a motor being started pulls the line down to a point that prevents it from starting. And most "high efficiency" motors should be worse on starting than standard ones (but may make good idlers).

Maybe if you can change the pulleys to get up to full speed, or above, the motor may "catch" without a huge surge, but I am doubtful that you will find it useful.

Do you NEED a 40HP idler?.
 
SO you guys are way ahead of me, and I have never done one as fancy or large as this, but the ones I have done, the pony motor is powered until the second the idler is fired, because single phase motors run slower than 3 phase. The high current start up draw is from getting the thing up to speed, you should not, I would think, see high draw from a motor spinning at 3400 rpm to get it to 3600 rpm.

My first[pre internet] try at building a converter did much as you describe until I figured out you needed to shut off the pony at the moment you put power to the idler

If it is wired this way and cannot get itself from 3450 to 3600 on its own and the wiring is sound I would think there must be something wrong with the idler
I'm most likely wrong but thought I would throw it out there in case someone was missing the obvious
 
In my case the run capacitors contributed to its "generating". It also took time for the voltage to climb. I was using a timer to kick in the mains power to the idler after starting the pony. The pony only needed to be on for about 2 seconds- but I had the timer at 10 seconds, which gave plenty of time for the voltage to rise.


With the initial run capacitors I had connected I was seeing 240 to 270 volts. At one point I hooked it up to a small coolant pump- and it ran, so it could provide current as well as volts.

Last spring I started to write up a post about what I found, but never finished it. I will review it and post it in a different thread as it may be helpful to someone.

Everything I read indicated a pony motor decreased inrush current. It may not do that on a consistent basis.

The data I got was that it could, but the rpm that produced the lowest inrush current was below synchronous RPM. This may be due to my particular setup, but there was noting out of the ordinary with regards to having run capacitors connected when starting the RPC.

I do not claim to have an understanding of what I observed. Hopefully it will make sense to others
 
Are all the motors rated at 3450 RPM or thereabouts? A 1:1 pulley ratio implies that the larger motors are also running at that. 3450 motors that large are made but uncommon.

Re residual magnetism, it isn't going to last long when power is connected.

Bill
 
Thermite- Nope, 2 gauge. It's in open air which is the only reason I'd even consider this. I really like the stove elements idea, especially since if voltage comes up as expected then the voltage across the elements begins to drop, lowering their temperature and lowering the resistance for a "stronger" link.

I do have authority to shut off every other load on the service prior to starting anything.

The 10hp idler will probably start the timesaver exactly once.


From my best estimates I expect the motor would want to pull 400 Amps to start with a single 600-700 amp half cycle magnetization surge. I expected the 400 amps to go away but the very high inrush current would stay there, but not matter since it would just be a cycle or two.

I have not considered turning off the pony just prior to starting the motor since I have a lockout wired in where you can't start the large motor without the pony on. I don't think it would matter though, because the pony motor draws less than one amp more than it does at idle once spun up, which tells me I have very little slip and am almost certainly above the 3530 rpm of the large motor.

I definitely have a lot of things working against me on this one. I do know that high efficiency motors require higher starting currents, and that the two pole motor is going to be harder to start. Also given that this is a high rpm pump motor, it may be designed for much higher starting torque than normal.

I anticipated idle draw to be 20 amps for the 40hp and 20 amps for the timesaver. That puts me at 60 amps idle with everything allowing 40 amps to the actual load. Which is terrible but not impossible.

To my knowledge the timesaver came from a place running 208, and the idler seems to be 208 rated as well (given the range of low voltage FLA). With a typical utilization voltage of 200 volts, and a no load line voltage of 234, I should be able to draw 147 Amps before line voltage drops too low. This does assume derating for both motors, and no winding above FLA.

If I can't get the 40hp idler to work I would just throw a 10hp single phase motor on the timesaver. But I'm $1800 into this converter and would like to be not much more into it than that. A single phase motor would be around $1200 shipped.

On this converter the run capacitors are on the output side, not the motor side, so it does not spin up with any capacitors connected.


But given the information I now have, let's analyze this:

The motor wants to draw 600 Amps for a single cycle on 230 V, given an instantaneous load for that cycle of 0.38 ohms.

Given line impedance of 0.23 ohms and base voltage of 234 V, that only gives 145 volts to start. This seems right given a contactor did dropout on one attempt.


I was assuming that the line would dropout for that cycle or two and then recover and allow the motor to start, but that voltage will never get any field going in any meaningful way.


So, next attempt will be as follows: When the pony motor is started, the large idler contactor will be bypassed with electric range heating elements. This will allow about 20 amps to establish rotor current from a standstill instead of at full speed and hopefully negate the need for inrush. If the elements stop glowing, then I know the voltage across them has reduced enough that direct engagement should be possible.

I could probably bring the idler online at the same time as the pony motor, but that would make for very hard starting for the pony, though it would probably work. I'm not going to try it though.


The rotating mass of the idlers should be enough to supply the magnetizing inrush needed for the timesaver.
 
Thermite- Nope, 2 gauge. It's in open air which is the only reason I'd even consider this.
I am not sure even this will work, but it is cheap to try.

I'd big-wheel hand-truck that RPC up to the head-end of that #2 drop so it had direct-access to the service-entrance of the heavier powerco feed. Run one more conductor to the shop, since it is not buried. Presuming 230-245 VAC off the RPC, head-end, leave the Timesaver at 208, and it might be OK with the drop.

The next "accommodation" would not be to drop the Timesaver to 10 HP single-phase, but rather to 15 HP or even 20 HP 3-Phase.

In EITHER case, I would change the ratio of that 5 HP pony so that it spun-up the idler to around 4,000 RPM, got TF OFF the line itself, and a "smart" sensor smacked the idler's contactor across the line as it was coasting down THROUGH the phase & RPM sync point. Sort of an "inertial starter with eyes" as it were. Sensors are cheap. Bit of perf board. Take the contactor lag as reasonably repeatable?

One spool of #2 wire and some sweat to give that the first try with the Timesaver as load. Hardly ANY out-of-pocket cost to at least see if the RPC starts better.

The pony ratio change should not cost much out of pocket, either.

Won't be rocket science to weather-proof snake and rodent proof, (don't ask.."warm place to shit" thing, not loose shoes, nor tight snatch, I'd guess) that RPC rig if there is no space indoors.

Another option puts a transformer into the mix to run that #2 at 480 V or so, Timesaver probably dual-voltage and re-strappable?

But that's another "I wouldn't". I don't even tolerate above-ground at all, and am not a fan of home shops having to get into 600 Volt Class switchgear and all the rest, with shock-hazards bad enough, already.
 
I am not sure even this will work, but it is cheap to try.

I'd big-wheel hand-truck that RPC up to the head-end of that #2 drop so it had direct-access to the service-entrance of the heavier powerco feed. Run one more conductor to the shop, since it is not buried. Presuming 230-245 VAC off the RPC, head-end, leave the Timesaver at 208, and it might be OK with the drop.

The next "accommodation" would not be to drop the Timesaver to 10 HP single-phase, but rather to 15 HP or even 20 HP 3-Phase.

In EITHER case, I would change the ratio of that 5 HP pony so that it spun-up the idler to around 4,000 RPM, got TF OFF the line itself, and a "smart" sensor smacked the idler's contactor across the line as it was coasting down THROUGH the phase & RPM sync point. Sort of an "inertial starter with eyes" as it were. Sensors are cheap. Bit of perf board. Take the contactor lag as reasonably repeatable?

One spool of #2 wire and some sweat to give that the first try with the Timesaver as load. Hardly ANY out-of-pocket cost to at least see if the RPC starts better.

The pony ratio change should not cost much out of pocket, either.

Won't be rocket science to weather-proof snake and rodent proof, (don't ask.."warm place to shit" thing, not loose shoes, nor tight snatch, I'd guess) that RPC rig if there is no space indoors.

Another option puts a transformer into the mix to run that #2 at 480 V or so, Timesaver probably dual-voltage and re-strappable?

But that's another "I wouldn't". I don't even tolerate above-ground at all, and am not a fan of home shops having to get into 600 Volt Class switchgear and all the rest, with shock-hazards bad enough, already.

You misunderstand. The service drop is 2 gauge copper overhead. The shop is all 4/0 buried Aluminum after the meter. I have dealt away with as much voltage loss as possible. And if that much loss was from a loose connection it would have made itself known.

The 2 gauge should only be 0.03 ohms though. I'm guessing the transformer and line supplying it are the rest. I am at the end of the line, but I confirmed with the neighbor who shares a pole that they can't tell when we pull 100 amps. It does go through 6 breakers in a odd branch-to-panel arrangement, but there is no way in hell each breaker is dissipating over 500 watts as they would have to if that is where the loss is.

A ratio change would be easy and I already have sensors that can do that on hand. But given how suddenly the motor brakes when power is applied, I don't think more RPM will help. I've made the mistake before of brute forcing things that should have been calculated and resolved more carefully.

I quite like not dealing with 480, and that probably still wouldn't help alleviate the problem. But yes, everything on the entire system can be configured to 480 if needed. I'm already not too happy playing it so fast and loose with 240 (which is really only 120 on two of the legs) and I don't want to do that with even more voltage.



I still think my best way forward is to try and slowly form a rotor field while the pony is kicking up, and use the pony power to overcome the braking force that would occur during that time. It would be very clear by heater coil glow and pony current whether it was working or not. The oversized pony lends itself to this quite well.



This was originally intended to stand next to the air compressor which was to be placed in a little shed thing on the other side of the wall, but that didn't happen and probably never will. The tower is okay where it is though.
 
You misunderstand. The service drop is 2 gauge copper overhead.

One supposes that helps shorten the local winters and prevent birdie feet from getting frostbite, but as far as running a heavy RPC or motor?

You'd be much better-off it was the reverse - Powerco-installed and professionally terminated 4-Ought Al the first leg, #2 Cu the last leg.

Not a lot of help having lower Ohmic losses after they have already taken their toll. Akin to a Weir in a water line.

15 HP is going to be about the prudent limit.

For an RPC that can START a 30 HP Timesaver? Nominal 45 HP idler, yah?

Lybarger's Corollary applies (All else being equal, YOU LOSE!)
 
Strostkovy;3268523 I still think my best way forward is to try and slowly form a rotor field while the pony is kicking up said:
How about doing a soft start on the 40 HP idler?

You need two motor contactors for this. Start the idler motor with the coil windings set for 480 volts. Bring it up to speed with the pony motor. Turn off the pony motor off and then apply power to the idler. Once the idler is running switch the coil windings to the 240 volt connection.

I use the soft start (1/2 voltage start) on my lathe to avoid the current rush which stalls the shop motor generator set. The starting inrush current is much greater than the motor full load current.

There are paired contactors made specifically for this purpose that use a mechanical interlock to avoid shorting out the motor windings. The interlock can also be established by routing the contactor coil power through a double pole relay. This insures that only one contactor can be engaged at a time.
 
Reading that thread I like my idea because it operates to synchronize and low voltage start.

I could implement a low voltage start with contactors but it would be a very significant change in wiring and increase in complexity. With stove coils, (only up to 20A or so though) I don't need to change any contactors.

Power company says $8000 for 200A, won't quote on 400A and laughs at the mention of three phase.


If this fails, I will put some neons across the contactor and see if that tells me anything.

Otherwise next best option is just a single phase motor on the timesaver and leave the RPC with just the small idler functional until I get a load that requires maybe an additional 15-20hp of idlers, and deal with it then.

Diesel is out because I don't want to maintain it and I have neighbors. Also, I don't pay the power bill but I would pay the diesel bill.
 








 
Back
Top