Phase help. Generated leg drops way low when starting compressor!

I'm using a Fitch Williams design. I can start a 3 hp buffer. But my 15 hp + 5 hp will not start my 7.5 hp compressor. Its a Ingersoll Rand 1 year old compressor I bought new. I had a store bought 3 phase convertor I bought back around 2000. It died but was struggling to run my compressor for the last year. That company was in Alabama and is no longer in business. I cant remember their name. The outside label of that unit said it was a 15hp convertor. But when i separated the panel from the motor, the motor was rated at 5hp? Thats the 5hp in my current configuration. By its self on the Fitch design , no way can it start the compressor. But the buffer it could. I then acquired a 15hp. I now start the 15hp then add in 5hp with its caps to balance.
Ive even tried running the buffer with both other motors and still no go.
I have tried a step up transformer to run my compressor on 480v to drop the amps and still no go.
I bought a scope with 4 channels to try to watch things but that has its own large learning curve.

At some point during all this i had a empty tank and one of these configurations sort of worked. When i stopped it around 100 lbs and tried restating it, no go. I believe mine has the dump valve on it.

I can get balance within 10% of L1 to do all these test Ive done but no matter where I try to watch voltage from, the panel, the idler or load my generated leg voltage will drop to around 115 when running 240v and 145v when running 440v.

With the money i spent on the scope i could have just bought a store bought unit but where is the fun in that.

So I don't understand how my last phase converter powered this compressor with only the 5hp? Everything is still located in there original spots with about 50 feet separating them using the same 6 awg wire.

Thanks John

Put a "fly wheel" on your idler, then try.

Your problem could be the result of a leaking check valve between the compressor pump and the tank. You may be trying to start the compressor against tank pressure. Check valves are cheap. Make sure yours is not leaking tank pressure back into the bleeder tube and pressurizing the manifold. Regards, Clark

Hi John, I would suggest the higher voltage transformer method and you will start and run your 7.5hp compressor without any troubles. You mention a 480v transformer, you should be able to reconfigure it to get 360v if your 7.5hp motor is 4 pole or 480v if your motor is 2 pole. Using thie method you will be able to run this compressor motor at it's full load name plate amps.
In building one of these type converters to run an air compressor, it can be very simple.

Jim

Hi and thanks for the quick responses!

First,
Calg. I have pulleys that would fit it or am I looking for more weight here? Does the spinning weight help?

Second,
Sealark37, I will pick one up and just replace it and let you know.

Third
Jim , I tried the higher transformer. stepping my voltage up from 240v to 440v. Which should of halved my amps. I also changed the motor over to the higher voltage. I didn't bother to recheck the amps because after trying this setup I still got the steep drop in voltage. But my logic like yours said this should have help.

I have had the best luck using a 6 pole motor for the phase converter motor.

Where are you measuring the generated leg to? Ground or one of the other 2 phases?

Another option to give a starting boost. Is to add a couple of 250 UF 250 Volt starting caps connected in parallel, to the generated phase and connected to the correct power line phase. A 1-2 Sec on delay timer with the NC contact in series with the caps to unhook them once the motor is started. Is all the control needed.

Messing with 480V only compounds the problem with transformer losses in the generated leg.

Bill

Increasing the inertia of the "pony motor" helps when starting other motors.

If it starts the unit empty, AND runs it up to full-pressure cut-off, BUT it won't re-start *with* tank pressure, then it must be starting *against* tank pressure. The RPC seems to be working.

Possibly in the mean time the unloader valve quit working right.... I didn't think a leak-back could close the unloader, I thought they needed a lot of air to shut them, which a leak doesn't give (you hope) but the compressor does.

Hi John,
I think you misunderstood my reply. This method still requires your motor to be connected for 240v operation. But you need an autotransformer, more than likely you would be able to reconnect the transformer you already tried as an autotransformer so you can get a boost voltage to the single run capacitor and then to the manufacturered phase terminal on the compressor motor. Yes you still need your start capacitors and a voltage sensing relay but with little experimentation you can achieve very reasonable phase balance with varying pressures in the receiver. Mainly you will consistantly start this compressor in any part of it's pressure cycle whether the unloader is working or not and if you require this motor to work at it's full load Amps it will.

Jim

I think the reason why one would jump to the conclusion that it's a converter issue, is because they're observing the generated leg sag on start.

Reality is, that it DOES. The generated leg of an RPC is not only limited in output, it's highly responsive to the load placed on the other two legs.

After the compressor reaches cutoff pressure, the m otor shuts down, then two things happen... first, is that the check valve between tank, and the outlet of the compressor pump closes, and the second, is that the unloader mechanism relieves all pressure between the pump and that check valve. They must BOTH work properly to relieve head pressure from the compressor.

As Bill says, take a wrench to the compressor-head fitting and back it off, see if it hisses. If so, stop messing with the electrical part and fix the unloader and check valve. Probably just dirty.

If you don't find a failed (leaking) compressor check valve, I'd look closely at the mains supply TO the RPC system, for a bad connection somewhere... anything that would limit input current to the RPC. If input current is slightly limited, the generated leg will get REALLY flaccid.

God how I hate a flaccid generated leg. I have seen situations where a compressor would not start on a RPC, but would on a static converter. I don't understand it either. Bob

Hi Again thank you, to all you for your help.

Ok, as Dave summed up in his post to back off the compressor head fitting. I did. ALL the way way off. no leak. BUT i don't have a dump system on this compressor. I still have it all connected in the 460v configuration, so both motors and then after the rpc i bump it up to 460v. With the head nut totally off it fires up beautifully, sounded strong! I did this at least 15 times to take the below readings with no hesitation. As soon as i connect the head nut, no go. To much load.

Here are the readings right at the contactor of the compressor,
These numbers are to one leg and ground.

L1 L2 L3
no load 276 204 266
drop to 223 63 105
w / load 256 169 224

amps (the motor rated at 9, i believe. Its hard to read)

L1 L2 L3
inrush 44 30 23
w load 10 9.2 4.6


The RPC ( before transformer and across two legs and no load)

L1 = 244 L2 = 280 L3 = 269

Yes, I admit these numbers could be a lot better with some more balancing, which I can do. But I guess I haven't rebalanced for all these different configurations I could use. Just trying to find the configuration that looks like it may work then I will tighten it up.

So is there something I could use or fix on the compressor to make this work? Such as dump valve?

One more thing.

When im trying to add the tank of the compressor back in, it has about 90 psi in it. It is not empty. Thats what it is pushing against.

I agree with Bill on this one- a 7.5hp compressor WILL have an unloader... if not, it won't start. Typically, it'll be a small diameter metal line going from the check-valve or pump-to-tank line, up to the control switch or contactor. A tab on the armature of the contactor will also actuate a valve that relieves pressure 'tween motor and cylinder to allow load-less startup.

Now another thing that comes to mind... the check valve- these things get bombarded with crud, and will sometimes get very sticky... not only not close, it can also stick shut and not OPEN until it gets a whole lot of pressure... and many of them double as the reaction discharge for the unloading mechanism. If yours is stuck, it would certainly cause this problem.

Last... on an RPC system, do NOT measure anything from one of the three phases, to ground or neutral. The measurement there, is totally irrelevant. Measure from phase to phase only.

This is clearly a compressor issue and not a converter issue. Stop monkeying with your phase converter. Take good pictures of your compressor and post them. Then we can help.

metalmagpie

You need a pressure switch that is equipped with an unloader fitting, a short tube, and a check valve with a similar fitting. When the compressor reaches cut-out pressure, it cuts power to the motor, as well opening the unloader valve, which bleeds off the pressure in the line between the pump and the check valve. This allows the compressor pump to start when needed against little or no line pressure. Regards, Clark

vintagetools said:

Ok, as Dave summed up in his post to back off the compressor head fitting. I did. ALL the way way off. no leak. BUT i don't have a dump system on this compressor. With the head nut totally off it fires up beautifully, sounded strong! I did this at least 15 times to take the below readings with no hesitation. As soon as i connect the head nut, no go. To much load.

Click to expand...

So when you removed the "head nut" did any air psst out? Is any air escaping from the line to the tank when the head nut is off, you may not hear a small leak?

The " As soon as i connect the head nut, no go. To much load." sounds a little fishey here. The motor should have no load for the first few turns so it can get up to speed. And any leak in the check valve can keep pressure in the line to the compressor.

The cheapie unloader valve doubles as the check valve, and looks like a check valve. And may be what was installed when the compressor was made.

The air flow stops when the compressor stops, then the check valve shuts, a small vent hole is exposed to vent the pressure off on the line to the tank. The vent hole is small and can get clogged with crud, mud dobber gunk or other stuff.

Any parts that have been changed or removed in this area since the compressor was new are suspect for not being the proper part.

Good troubleshooting is verifying assumptions. There are some basic assumptions that are tripping you up here.

So the question remains is: "When im trying to add the tank of the compressor back in, it has about 90 psi in it. It is not empty. Thats what it is pushing against."
How does the tank pressure now get to the compressor before it starts???? Leaking check valve? Check valve replaced with a ball valve? Something else???

Bill

Once you have determined there is a functioning unloader, if it still won't start, try starting by pushing and holding the contactor in with a wooden dowel or similar insulated pusher. In some cases momentary voltage sag will trigger the phase loss protection in the overload block for the motor starter and pulsate the contactor so it won't start the motor. If you hold in the contactor (as a test) it may start fine after the few seconds the voltage takes to return to normal. You may be able to rig something to do this automatically as a long term solution. I had this problem testing a 7.5 on my 15hp rpc recently....
tom