What's new
What's new

15 hp RPC struggles to start 7.5 hp 3 phase air compressor - add cap?

metalmagpie

Titanium
Joined
May 22, 2006
Location
Seattle
I have had my RPC up and running for ten years or so without a hitch. Confident, I decided to rebuild a 3 phase air compressor despite persistent rumors they can be hard to start with an RPC. I spun it up today for the first time. It started fine with the tank empty. (There is no pressure gauge installed on the tank at present.) However, with the tank partly full the motor pulley spun quite slowly as if overloaded, and never came up to speed before the output breaker on my RPC blew. I tried again with 2 other 3 phase machines running and the motor started fine.

Let me digress into unloading systems for a moment, please bear with me. The air pump is from the Champion R series. Those have robust unloaders. The discharge tubes vent to outside air as long as the motor isn't running. As soon as it starts running, a centrifugal gizmo closes a valve and this allows air to begin building. It is difficult for me to imagine how if the centrifugal unloader seems to be working that it could somehow not really be working. It's pretty robust.

My current use model for this machine is for occasional use, so it's not enormously burdensome to have to turn on a couple of extra machines. But it's wasteful of power, and noisy. I'm wondering if I could simply add some run capacitors to boost the RPC's heavy starting capability. Does that seem possible to you?

metalmagpie
 
  • Like
Reactions: SAF
I have zero direct experience with your exact situation but, I do have a 30 HP RPC and this is what I think you're seeing:

If you think about single phase power, it turns off 120 times every second. That also means the mechanical power is stoppign that many times. It's like pedaling a bicycle up a hill in too high of a gear: you don't have enough speed and can't push hard enough for long enough to get to the top of the next crank.

Three-phase power always has another phase coming up (mechanical energy) when one is starting on its way down. It's like the same bicycle having three different people, each pedaling and at different points in the rotation. Someone is always pushing.

The rotary phase converter attempts to skew two of the phases to do the work of two cranks and capacitors to generate the missing third leg (pedal) of input force to spin your compressor. A heavy load makes this worse because the capacitor can only store so much energy to fill that giant electrical hole where the power is otherwise off 120 times a second.


It started fine with the tank empty. (There is no pressure gauge installed on the tank at present.) However, with the tank partly full the motor pulley spun quite slowly as if overloaded, and never came up to speed before the output breaker on my RPC blew.
Exactly. The capacitor just doesn't have enough ummph to keep the motor spinning.

But then...
I tried again with 2 other 3 phase machines running and the motor started fine.
This is because with all those other motors spinning, they are also generating electricity in the lines. Every time your compressor wants to stop, there is mechanical energy stored in all those other motors that is powering the compressor's missing leg. With all those extra motors, you made a much larger RPC: perhaps 25 or even 30 HP (combined of all the motors and RPC).

Not sure what the answer is here. If the pump is legit sized to a legit 7.5 HP of load, a single phase motor of the proper size may not be available.
 
I have zero direct experience with your exact situation but, I do have a 30 HP RPC and this is what I think you're seeing:

If you think about single phase power, it turns off 120 times every second. That also means the mechanical power is stoppign that many times. It's like pedaling a bicycle up a hill in too high of a gear: you don't have enough speed and can't push hard enough for long enough to get to the top of the next crank.

Three-phase power always has another phase coming up (mechanical energy) when one is starting on its way down. It's like the same bicycle having three different people, each pedaling and at different points in the rotation. Someone is always pushing.

The rotary phase converter attempts to skew two of the phases to do the work of two cranks and capacitors to generate the missing third leg (pedal) of input force to spin your compressor. A heavy load makes this worse because the capacitor can only store so much energy to fill that giant electrical hole where the power is otherwise off 120 times a second.



Exactly. The capacitor just doesn't have enough ummph to keep the motor spinning.

But then...

This is because with all those other motors spinning, they are also generating electricity in the lines. Every time your compressor wants to stop, there is mechanical energy stored in all those other motors that is powering the compressor's missing leg. With all those extra motors, you made a much larger RPC: perhaps 25 or even 30 HP (combined of all the motors and RPC).

Not sure what the answer is here. If the pump is legit sized to a legit 7.5 HP of load, a single phase motor of the proper size may not be available.
This isn't really very accurate. Capacitors provide voltage support & reactive current but little more in an AC system. The energy storage to provide the third phase is coming from the rotation of the phase converter. It is generating the additional phase(s), not the capacitors.

In addition, the motor is certainly not stopping 120 times per second. The *torque applied is*, but the acceleration/speed is a matter of the average torque.

The electrical issue here is that the starting power draw (both real and reactive) is causing the voltage on the generated phase to sag under load, reducing starting torque, and decreasing acceleration. More capacitors can fix this under locked-rotor conditions, but as the reactive current draw reduces and the motor gets under speed, the RPC will now have far too much capacitance and suffer over-voltage.

The mechanical issue is that the unloader closes quickly after the compressor begins to spin. Any revolutions done after the unloader closes are then filling up the discharge line to the non-return valve, with each piston stroke adding more pressure to that line.

With a rapidly accelerating motor, this isn't an issue. It only takes a couple of revolutions for the motor to reach full speed, and it is pumping at near full speed into that line *before* the pressure in the discharge line reaches tank pressure and the valve opens.

With a slower acceleration, this is a problem. It's putting tens or hundreds of revolutions worth of air into that discharge line, quickly reaching full output pressure, so the machine has to accelerate against full head pressure.

Options to fix it could include:
  • Delay the existing mechanical unloader somehow.
  • Replace the unloader with an electrical or otherwise timer-based one.
  • Add more air storage to the discharge line (e.g. bigger diameter, longer pipe, or a small reservoir) so that it can pump more air before reaching full pressure. Downside is you'll lose more air at end-of-cycle when the unloader operates.
 
This isn't really very accurate. Capacitors provide voltage support & reactive current but little more in an AC system. The energy storage to provide the third phase is coming from the rotation of the phase converter. It is generating the additional phase(s), not the capacitors.
I was trying to keep it basic and bite sized. You're not wrong but sheesh.
 
If you want, i can wind you a 1 to 4 turn, 10 awg modified 40 amp contactor.

Put the 2 turn coil in series with one of the pass through phases on the compressor. When it starts it will pull in, when the line amps decrease it will release.

Connect additional run capacitors connected from generated leg to whichever phase your rpc start caps are on, through the contactor.
 
This is really not an issue for more capacitors. The capacitors can only do so much in terms of voltage boosting, and then adding more becomes counterproductive, actually adding a current draw to the generated leg.

The centrifugal unloader should be good, it ought to delay closing until a certain pump RPM. But if the unloader is closing at a lower speed (weak spring, etc depending on construction), it may not let the motor get up to a speed where torque is sufficient to keep running.

If the unloader closes early, then the motor may still be in an RPM range where the slowing when the load is applied reduces torque instead of increasing it.

There is another form of unloader, which is "flow based". Mcmaster has had them, although I don't know if they have these in a useful size for you, and I could not find them just now when I looked.

These let air out of the input pipe through a "leak orifice", delaying the buildup of pressure. When the flow through the "leak" gets near the compressor CFM rating (meaning the compressor is nearly up to speed), the pressure on a piston is high enough to move it. The piston closes the "leak" orifice and opens a passage to the tank.

That may work for you, if the rating of a McMaster unloader is suitable for your pump.
 
I had a similar issue trying to use a VFD to start a 10hp quincy. farted around with the unloader valve forever, no joy.

Installed an electric air solenoid valve on the unloader and delay timer set long enough to get motor up to speed plus a few seconds. works a charm now for many years.

PS. Its been too long. But I seem to remember the unloader comes in when it sees air pressure. releasing the air lets compressor run. Anyway get the air solenoid type that allows this. I had a pile of these parts, used on the shelf, and the first one I tried did not work correctly.
 
Last edited:
OK so I'm envisioning a solenoid valve connected in parallel with the mechanical unloader valve from the discharge tube to the atmosphere. The idea would be when the pressure switch toggles calling for the motor to start, that the solenoid valve would then open for a few seconds and then close until the next time the pressure switch turns on. This is simple mechanically - plumb a tee into a 3/8" copper line, run 3/8" copper tube to solenoid valve input, put muffler on solenoid valve output. The complication is in controlling the valve. I need a circuit which senses AC power turning on and which passes AC power for a few (adjustable?) seconds and then shuts off said AC power until the next time it senses AC power going from off to on. Does that make sense?

I would prefer to avoid complicated and expensive time delay relays in favor of something simple ideally with an external resistor and capacitor for timing, so I could change them. If this control circuit were to be on all the time, it would be preferable for it to draw very low power, because this is going on my own personal power bill. :-)
 
I had a similar issue trying to use a VFD to start a 10hp quincy. farted around with the unloader valve forever, no joy.

Installed an electric air solenoid valve on the unloader and delay timer set long enough to get motor up to speed plus a few seconds. works a charm now for many years.

PS. Its been too long. But I seem to remember the unloader comes in when it sees air pressure. releasing the air lets compressor run. Anyway get the air solenoid type that allows this. I had a pile of these parts, used on the shelf, and the first one I tried did not work correctly.
Yeah that is straightforward, and as you suggest, less screwing around. I have a small lab type pump that I plan to use for general 50 psi or so air around the "small work" part of the shop, and it won't start against any load. The unloader I have is not quite sized correctly, I may go with a solenoid.
 
Side note: if you go with an electrical unloader, you can consider doing what a lot of smarter large compressors do: at part load, run the compressor continuously and maintain pressure control using the unloader, significantly reducing start cycles and (potentially) improving pressure regulation.
 
Say, NOW there's an idea I can use. I run a sandblaster on occasion and the unit kicks in/out constantly. I am going to have to think this over for the electric logic, but the idea is have a "MODE" switch. standard, and always run. In always run bring in unloader instead of drop motor contactor. That's how gasoline contractor's compressors work.

<EDIT> simple: use a second pressure switch set 5 PSI lower. wire this one to unloader. Bypass it for standard operation.
 
I have had my RPC up and running for ten years or so without a hitch. Confident, I decided to rebuild a 3 phase air compressor despite persistent rumors they can be hard to start with an RPC. I spun it up today for the first time. It started fine with the tank empty. (There is no pressure gauge installed on the tank at present.) However, with the tank partly full the motor pulley spun quite slowly as if overloaded, and never came up to speed before the output breaker on my RPC blew. I tried again with 2 other 3 phase machines running and the motor started fine.

Let me digress into unloading systems for a moment, please bear with me. The air pump is from the Champion R series. Those have robust unloaders. The discharge tubes vent to outside air as long as the motor isn't running. As soon as it starts running, a centrifugal gizmo closes a valve and this allows air to begin building. It is difficult for me to imagine how if the centrifugal unloader seems to be working that it could somehow not really be working. It's pretty robust.

My current use model for this machine is for occasional use, so it's not enormously burdensome to have to turn on a couple of extra machines. But it's wasteful of power, and noisy. I'm wondering if I could simply add some run capacitors to boost the RPC's heavy starting capability. Does that seem possible to you?

metalmagpie
I'm afraid you are missing some of the electrical concept of an RPC. As someone that has built many, as well as dealt with compressor issues, I might try to help.

Before you do anything, I recommend you find a place with real 3P power to test that compressor. I have dealt with several issues that were related to unloaders or leaking checking valves to tank. You should see only a small increase im amperage over the compression range as PF goes up, which is why amps don't change too much, but power does.

You need too know what amps are doing as well as voltage during load conditions. An RPC is NOT capable of maintaining phase balance. You need to check all L-L voltages for unloaded and loaded up against that compressor. You will find it to move more than you think.

You cannot add caps to 'boost' power. Doesn't work like that. Current through a cap is fixed by frequency but if you go adding, you will disrupt the 3P voltage balance. You first need to see where the issue is. It is true that you are current limited on an RPC on that 3rd leg, which is why they must be oversized for the load.

Is your RPC commercial or home built? People here always feel insulted when I tell them their RPC is F'd, even if they spent big $ on it. You need to learn how to tune one.
 
.............................. I tried again with 2 other 3 phase machines running and the motor started fine.
.................................

metalmagpie
Missed this the first time through.....

The RPC is "undersized" for the demand. You knew that, but that sentence proves it. You provbably know the rest of this also, but it may as well be stated.

OK... "undersized" has details. The RPC can run that motor fine.... and it is significantly oversized for the "perceived load"... you'd think a motor 2/3 the power of the idler would be OK, and you have double.

What oversizing the idler gets you is less voltage drop under load on the generated leg. You are then using a motor with a higher current rating, having lower resistance windings, lower inductance windings, really a lower impedance on the generated leg, so the capability to generate more output amps on the generated leg.

As you know, the other machines, if just idling, will provide extra current on the generated leg. You can assume that they add approximately the same amount more current, proportionately as their extra power suggests. So a 15 HP idler, with two 2 HP machines idling, will in total, act like an RPC of about 17.5 HP or so (I figure them to be about 85% or so effective).

Every induction motor has a torque curve. Torque increases as speed increases, until a maximum occurs. Above that, torque decreases, since current is less. Below that, torque decreases, due to effects like saturation of the iron by the extra current, etc.

You need enough current to get the load motor past that peak of torque. If you get there, the motor will speed up and take the load. If you miss, the motor will bog and current will increase.

It may not take much more current to do it. Just the added current from a couple machines did it for you.

However, lowering the load (using an unloader, for instance) will do as good a job as a bigger idler, and costs less, especially when you have the RPC already.
 
I'm afraid you are missing some of the electrical concept of an RPC. As someone that has built many, as well as dealt with compressor issues, I might try to help.

Before you do anything, I recommend you find a place with real 3P power to test that compressor. I have dealt with several issues that were related to unloaders or leaking checking valves to tank. You should see only a small increase im amperage over the compression range as PF goes up, which is why amps don't change too much, but power does.

You need too know what amps are doing as well as voltage during load conditions. An RPC is NOT capable of maintaining phase balance. You need to check all L-L voltages for unloaded and loaded up against that compressor. You will find it to move more than you think.

You cannot add caps to 'boost' power. Doesn't work like that. Current through a cap is fixed by frequency but if you go adding, you will disrupt the 3P voltage balance. You first need to see where the issue is. It is true that you are current limited on an RPC on that 3rd leg, which is why they must be oversized for the load.

Is your RPC commercial or home built? People here always feel insulted when I tell them their RPC is F'd, even if they spent big $ on it. You need to learn how to tune one.
My phase converter is home-built. I stand behind the design. Your tone leaves something to be desired.
 
My phase converter is home-built. I stand behind the design. Your tone leaves something to be desired.
Ok, I see I have hurt your sensitive, feminine feelings. I'm sure you will figure it out. Just add some caps.
 
In the original description the op said the pump initially started unloaded (0 pressure in the tank) then as it pumped up running full speed it started to bog down. If I’m understanding that correctly this also points to the rpc being over loaded and not an unloader issue.

If I was in this situation I’d buy a single phase motor (not cheap I know) and put that on there. This is the way mine is.
 
In the original description the op said the pump initially started unloaded (0 pressure in the tank) then as it pumped up running full speed it started to bog down. If I’m understanding that correctly this also points to the rpc being over loaded and not an unloader issue.

If I was in this situation I’d buy a single phase motor (not cheap I know) and put that on there. This is the way mine is.
It depends how you read it. I focussed on this:
and never came up to speed before the output breaker on my RPC blew
I take it to mean that it's subsequent starts that never reach full speed - not that the motor reaches full speed initially, then starts to slow down again as it approaches full pressure .

It's potentially both an RPC and unloader issue. You can either strengthen the RPC so it can slam the motor up to full speed before it starts to build pressure, like a proper stiff 3~ grid supply would. Or you can delay the unloader closing a bit so that it doesn't matter if the motor takes another second or two to reach full speed. Or both.

The second option is likely to be cheaper and easier.

I still stand behind the option of just putting a bit of a reservoir, say, a capped pipe of about half a liter/a pint, between the unloader valve and the non-return on the tank. Now the compressor can fill this while it's coming up to speed. This is all but free.
 








 
Back
Top