20 hp rotary phase converter conundrum

[NukeET]

Hello fellow home gamers,

I have found myself in need of wisdom regarding variable frequency drives, soft starters, and starting capacitors.

For context, I have built a 20 hp rotary phase converter and when running I am very pleased with its operation. The problem is starting the thing. Initially I tried using a pony motor but found that even when geared down, most 120v motors just stall. I could switch to DC pony motors but I would need to build an entire DC bus just to start the motor, not to mention mechanically coupling it being fairly difficult in the enclosure. I instead opted for a more traditional capacitor start system but had a very difficult time sourcing adequate capacitance. I eventually chained together 70 dollars worth of starting capacitors (about 1200 microfarad) which should have been adequate. They started the motor but vented within the 2 seconds they were energized. No good. I could just add more starting capacitors but at a cost of around 100 bucks I started wondering if there was a better solution.

 I can get 2-3hp VFDs for that amount of money and while grossly under rated for actual load, I’m wondering if that would work reliably for soft starting the motor. Obviously what I actually need is a soft starter, but finding one that converts single phase to 3 phase is no small order, and I wouldn’t even begin to know what rating it should be.

In short, can I use a significantly smaller rating VFD as a soft starter, and if so, what would be an appropriate rating?

[Kleinstein]

For starting one should first bring up the motor and only than electrically connect the load. I have a smaller rotary converter ( ~ 6 HP), that starts Ok this way, though running from 3 phase 400 V to start with.
The larger motors usually have a way to switching the windings in starr to triangle configuration. The usual way is to start up as star and than switch to triangle.

For current limiting in start a capacitor is unusual. The more practical way would be a series inductor.

For just startung a lower rating VFD should be OK. With a slow ramp up it only has to provide the power for the ramp up.
There is still the problem with connecting the secondary load.

[CaptDon]

You should never start the unit with a load on the secondary. Is this a single phase input to three phase output?
You 'built' the converter?? Tell us a bit about it. What is the input voltage and frequency? What is the output
voltage and frequency? What kind of output device is it? A big 3 phase alternator salvaged from an engine
generator set with a blown engine? I ask these questions mainly because my phase converter experience is
with self contained units with all of the windings on one stator. My 'shaft coupled experience' is entirely engine
alternator equipment including P&W and Solar (Caterpillar) turbine units and also three phase motor input
shaft coupled to a D.C. generator that in turn drives the D.C. field or 'excitation' of a multi-megawatt hydro
turbine. We can better offer solutions to your question if you tell us a bit more about your home built device.

[Stray Electron]

  A 20 HP rotary phase converter???  Is that a phase convertor capable of driving a 20 HP motor or a phase converter built out of a 20 HP motor?

  What kind of AC power do you have available to drive that thing?  20HP converts to about 15,000 Watts,, even at 100% efficiently and driven from a 220VAC source, that's roughly 70 amps. And that's not including the starting current which will be much higher!

[PaulAm]

well, you're not doing something right.

I have a capacitor start 7.5 hp rotary converter. The start switch connects the start capacitor into circuit until it gets up to running speed.  Generally, that's about a 1 second  contact.  That's manual, I didn't bother to use electronics for that.  There's also a contactor to disconnect the power and shut it down.

You get much bigger than that, the starting current starts getting ridiculous.  For a 20hp I'd use a pony motor.  If the bearings are decent you can let it stay mechanically coupled all of the time, however it should only be energized until the main motor comes up to speed.  Once at speed, disconnect the pony THEN apply power to the main motor.  No load should be connected when doing this.  You need some contactors and sequencing logic to do this automatically.

Once the main motor is at speed, then you can apply a load.

Use appropriate pulleys to get the motor to the correct speed which will depend on  the main motor and what you're using for a pony (this isn't too critical, but you should be within 10-20% of the main motor's rated speed). 

This apparently needs repeating: the main motor should not be energized while it is being brought up to speed by the pony.

[Ground_Loop]

Hello fellow home gamers,

I have found myself in need of wisdom regarding variable frequency drives, soft starters, and starting capacitors.

For context, I have built a 20 hp rotary phase converter and when running I am very pleased with its operation. The problem is starting the thing. Initially I tried using a pony motor but found that even when geared down, most 120v motors just stall. I could switch to DC pony motors but I would need to build an entire DC bus just to start the motor, not to mention mechanically coupling it being fairly difficult in the enclosure. I instead opted for a more traditional capacitor start system but had a very difficult time sourcing adequate capacitance. I eventually chained together 70 dollars worth of starting capacitors (about 1200 microfarad) which should have been adequate. They started the motor but vented within the 2 seconds they were energized. No good. I could just add more starting capacitors but at a cost of around 100 bucks I started wondering if there was a better solution.

 I can get 2-3hp VFDs for that amount of money and while grossly under rated for actual load, I’m wondering if that would work reliably for soft starting the motor. Obviously what I actually need is a soft starter, but finding one that converts single phase to 3 phase is no small order, and I wouldn’t even begin to know what rating it should be.

In short, can I use a significantly smaller rating VFD as a soft starter, and if so, what would be an appropriate rating?

Great user name.  I was a Nuke ET on Enterprise.

[james_s]

  A 20 HP rotary phase converter???  Is that a phase convertor capable of driving a 20 HP motor or a phase converter built out of a 20 HP motor?

  What kind of AC power do you have available to drive that thing?  20HP converts to about 15,000 Watts,, even at 100% efficiently and driven from a 220VAC source, that's roughly 70 amps. And that's not including the starting current which will be much higher!

That's not THAT unreasonable. 70A circuit breakers are widely available, and the typical American house will have a 200A 240V service, so running a 15kW load is totally feasible. A lot of houses built in the 60s-80s had electric forced air central heat which would typically be 15-25kW so it's not uncommon to have that extra capacity once the home has been upgraded to a gas or heat pump system. Also a rotary converter made from a motor needs a motor rated for significantly more power than you intend to draw from it.

That said, I can't think of many reasons for messing with a rotary converter anymore. They were once common for home machine shops but these days VFDs are cheap and give you additional features. It usually makes more sense to just use a VFD on each individual machine.

[NukeET]

Thanks for the feedback!

Additional system information:

The idler motor (that’s the motor/generator) is a 20hp 3 phase induction motor. It is a variable voltage motor and is currently wired into 230v delta according to the wiring diagram. T1 and T2 of the motor are wired to 230v L1 and L2 from a 200amp service. T3 of the motor is the generated leg. The phase converter control cabinet has a start, run, and output contact that are all controlled by an arduino and relay board to control timing and some safety features. The output contact remains open until the idler motor is stable to ensure it does not start loaded nor expose equipment to startup transients.

The phase converter was rated at over double the load it would be supplying as that seemed to be a good rule of thumb from what I saw. It drives a large 3 phase transformer welder that at most would be a 10hp equivalent. I also have a large assortment of other 1-2 hp 3 phase motors it will end up powering as well. The reason I built the phase converter over just using a VFD was suitability of application. A VFD is great at starting and running motors, not so great at handling instantaneous full power spikes like would be seen with a welder. The phase converter is great for this because the rotational mass of the rotor acts as a flywheel during surges. The entire build also costs a fraction of what a single 10hp VFD would cost. In addition, I can power an entire shop with a 3 phase bus instead of dedicating VFDs to each motor.

Hope that helps.

[james_s]

Yeah I don't think a VFD would be ideal for running a welder, it might work but it certainly would not be optimal.

[NukeET]

So you’re starting the phase converter with 3 phase power? Is that three phase from a VFD or do you have dedicated lines?

[NukeET]

It’s made out of a 20 hp motor. It could in theory output 20 hp, but I have rated for continuous duty at 10hp for the exact reasons you mention.

[NukeET]

I have built all the contacts and sequencing logic myself. It worked fine with a pony motor, but was not very reliable. A 3/4 hp 120v ac motor would stall even when at a mechanical advantage of 3 to 1. Starting torque on single phase induction motors is just not suitable. I didn’t have any universal motors handy either. It was also bulky and loud and I didn’t love the idea of having to replace a belt that runs constantly for no good reason. I even looked into electric clutches but at that point there are better options for cost. It did start with the capacitor bank, but it killed at least 2 of the caps. Seems they don’t make start capacitors for niche applications like this. If I continued to add capacitors it would probably work, but for the cost I’m wondering if a VFD start is a better way to go.

The system itself has push button start and stop, low voltage protection, and failed start lockouts I designed and tested myself. I foresaw this being used with machining equipment and decided to invest the time in adding the most important safety features.

[NukeET]

Ha! That ship was something else. I was on the Reagan.

[NukeET]

I watched a video of someone doing it. It didn’t work well and then it didn’t work at all.

[CaptDon]

Is this a true rotary phase converter like a phasemaster? From your description it sounds
as if you are feeding a standard three phase motor with single phase and expecting a
phantom 3rd phase to be generated by the motor. That doesn't work well since L1-L2 are
180 degrees out of phase not 120 and the phantom phase will be way out of proper
sequence. In some ways it also sounded like you had a big single phase motor shaft
coupled to a 3 phase alternator. B.T.W., secure that thing to the floor! We had a phasemaster
rip itself off the mounts and hurl across the floor when its 480vac output arc flashed
ahead of the protection circuit during a lightning strike at a transmitter site on a mountain
top. The rotating mass was probably around 100 pounds and it didn't like coming to a stop
from 1800 syncronis rpm's to stop in about 1 revolution!!

[PaulAm]

VFDs work well for single machines but not well for things like CNC that may contain their own  VFDs on one or more axes, or things like welders that may have significant instantaneous loads.  VFDs also do not work well if you are interested in "plugging", ie throwing the motor into reverse to rapidly stop it or back out.

That doesn't work well since L1-L2 are 180 degrees out of phase not 120 and the phantom phase will be way out of proper sequence.

Never ran one of these, have you?  You put in single phase on L1-L2 and you get the other two phases generated on L1-L3 and L2-L3.  They are not perfect, but you can tune them a bit with run caps.  It works well enough to be able to run CNC equipment and anything that uses 3ph motors.   There's enough energy stored in the rotating mass to be able to handle some peak starting currents, but it's important that the idler not be too small for the load.

If I had unlimited money, I'd get a Phase Perfect which essentially synthesizes 3ph power and is built to handle machine shop loads.  Rotaries are cheap and work well enough for most cases though.

Back to the original problem: gear your pony down so it can get the idler moving.  If it's moving fast enough it will pull into synchronous speed when power is applied.   You might have to decouple the pony then.  One thing to try is to have a loose belt between the pony and idler and use a lever to pull the pony tighter to the belt.  Start the pony, apply tension and let the belt slip until the idler is up to speed then apply power.  Not elegant, but will work.  Just don't get caught in the belt.  You need to get the pony to speed and then have some kind of slipping clutch to get the big motor up to speed.

You could wind a rope around the motor shaft and give it a good pull which would probably work.  Be sure the rope is clear before applying power

Edit: Here's a thought.  Get a 1hp 3ph motor and a small VFD.  Use the VFD to drive the small motor.  Couple the motors then you can ramp up the VFD until the idler is at synchronous speed.  Depending on the mass of your idler,  you may still need to gear things down a bit, but you can program the ramp time on the VFD and it might just work fine.

[NukeET]

All great ideas. It seems a 3hp VFD is going to be a good way to go. Worst case scenario I have a 3hp VFD I can use for other motors. Thanks guys.