Pony Motor Question - 25 HP RPC

[Strostkovy]

Thanks. Been a while since I was in an EE class.

400uF.

According to Capacitive reactance calculator reactance capacitor frequence - sengpielaudio Sengpiel Berlin that's 6.6 ohms of reactance. At 240V that is 36 Amps. Since RPCs with capacitors tend to have some very wild generated legs it's possible to see a fair amount more than that, especially when figuring out balancing. The actual current you see through the capacitors will be high at low load, low at moderate load, then high at high load again, typically.

In any case that is far too much current for 14 gauge. 6 gauge is nearly fool proof. 8 gauge won't cause any problems unless something is already seriously wrong. 10 gauge is probably adequate unless a very specific set of entirely possible scenarios of load and balancing occurs. Anything less will get too hot.

If it is unwieldy to make connections to capacitors with that thick of wire, use thick wire to a distribution block, then use thinner wires to individual capacitors or banks of capacitors. Remember, every time you double the amount of conductors the effective wire gauge size drops by 3. So two 9 gauge wires are equivalent to one 6 gauge wire, as are four 12 gauge wires.

 

[Balvar24]

Thanks. That's a big help.

SO.. I'm currently running from a distribution block to my Idler, Capacitor bank, and out to my load. But, I'm running out of taps.
I can hook two 100uF capacitors (13.3 ohms) in parallel on a 14 Ga THHN (rated for 25 amps) back to the distribution block.
This x 2 will get me back to my 400 uF. AND, for a little safety, I could wire it in at 12 or 10 awg.

Sound about right? Trying to make due with stuff on hand, as long as I don't build a grizzly bear.

This was probably one (there were a few others) of the downfalls of this set up it was with it's previous owner. He had 500 uF wired up in 12awg. When I removed it all, I could tell that it had been hot a few times.

 

[Strostkovy]

Thanks. That's a big help.

SO.. I'm currently running from a distribution block to my Idler, Capacitor bank, and out to my load. But, I'm running out of taps.
I can hook two 100uF capacitors (13.3 ohms) in parallel on a 14 Ga THHN (rated for 25 amps) back to the distribution block.
This x 2 will get me back to my 400 uF. AND, for a little safety, I could wire it in at 12 or 10 awg.

Sound about right? Trying to make due with stuff on hand, as long as I don't build a grizzly bear.

This was probably one (there were a few others) of the downfalls of this set up it was with it's previous owner. He had 500 uF wired up in 12awg. When I removed it all, I could tell that it had been hot a few times.

Two 12 gauge wires is as small as I would go.

 

[jim rozen]

Two 12 gauge wires is as small as I would go.

+1

Two issues - if a wire comes adrift from the capacitor bank and gets to ground, if it's a small enough wire it may blow up before the breaker feeding the setup trips. This is a good reason to always have that stuff in a steel enclosure per code. The other issue is you can have large circulating currents in the run capacitors.

1) see if the wires get hot. Don't electrocute yourself.
2) sneak an amp-clamp in there. It does a great job of reading the reactive currents.
3) upsize the wires to near the size the main breaker is rated for.

They don't typically teach this stuff in EE school.

 

[JST]

+1

........ The other issue is you can have large circulating currents in the run capacitors.

........

Yes. Typically the currents are larger with a larger boost in voltage from the caps, so if the voltage is in a reasonable range it won't be an issue, but......

That's an issue with the cap boost/balance system. The current subtracts from available current for the load, and it tends to load down the idler as well, raising the input current, but the good news is that the heavier the real load the less the boost and the less the current, so it self-compensates.

But if you are setting up a pony start, and have too much capacitor on the system, it can bog the pony.

 

[Balvar24]

I'm going to run (attempt) a 10HP compressor on it. I know that may come with it's own issues.

I've currently got everything in a steel enclosure. My additional capacitors came yesterday. I'll string it up with #12 and report back after I get it balanced and load test it.

Thanks again.

 

[Strostkovy]

I'm going to run (attempt) a 10HP compressor on it. I know that may come with it's own issues.

I've currently got everything in a steel enclosure. My additional capacitors came yesterday. I'll string it up with #12 and report back after I get it balanced and load test it.

Thanks again.

Honestly compressors aren't as bad as people seem to think. If you do have issues, a timed unloading valve is an easy solution.

 

[Balvar24]

I've got a small tank to install on the pressure switch line to let me run unloaded for a bit.

 

[Balvar24]

Did a preliminary balance on the RPC. Unloaded amps came in at 8.5. I fired the compressor (unloaded) and it looks like I'll need to do a bit more balancing. The coil for my motor starter is bad (used). I've got a new one on the way. I had 200 uF on one leg and 400 uF on the other.

 

[JST]

Did a preliminary balance on the RPC. Unloaded amps came in at 8.5. I fired the compressor (unloaded) and it looks like I'll need to do a bit more balancing.........

How do you know you need more balancing?

 

[Balvar24]

245 volts on the non-generated leg. 238 volts on th e generated legs. All with the compressor unloaded. Prior to running true load motor, the converter was running 262 volts on the generated legs.

 

[JST]

245 volts on the non-generated leg. 238 volts on th e generated legs. All with the compressor unloaded. Prior to running true load motor, the converter was running 262 volts on the generated legs.

Sounds to me as if you have it just about right. You really do not want it any higher with no load or light load, you are at the UL max for 240V right now.

 

[HappyWyo]

I know this is kind of late in the conversation, but we made a slip apart joint like on an old crank car or crank tractor. It was 2 piece, the first piece was attached to the pony shaft with a set screw. The second piece engaged the first piece in a "slide in " manner. The second piece was matched but mounted on bearings and was an integral part of the pulley. When the main motor surpassed the pony motor's speed it would "kick" free and the pulley would continue rotating on bearings on the pony motor's shaft. This allowed the pony to stop rotating eliminating wasted energy, and possible eddy currents. I am sure there are better ways but that is what we dreamed up. We used it that way for about 15 years. It may give someone some ideas.

 

[Strostkovy]

Capacitors also change the phase angle so a balanced voltage may not be a balanced current, and therefore may be an unbalanced winding load.

I suggest using your amp clamp on each motor lead under typical load conditions and basing your adjustments on that.

You can also balance with inductors (and reverse capacitor run motors by replacing them with inductors) but that gets weird.

 

[Balvar24]

I hated phase angles in school. Good point. When I ran it, I had it wired up temporary for a function test on the compressor/motor. I wanted to make sure it wasn't DOA before I put any more work into it. I'll get it set into place and wired in. Then I'll re-check voltage and check amperage on each leg. Thanks.

 

[JST]

Capacitors also change the phase angle so a balanced voltage may not be a balanced current, and therefore may be an unbalanced winding load.

I suggest using your amp clamp on each motor lead under typical load conditions and basing your adjustments on that.

You can also balance with inductors (and reverse capacitor run motors by replacing them with inductors) but that gets weird.

The entire point of "balancing" is to have the output of the RPC behave as much like powerco 3 phase as possible.

There is an inherent imbalance in ANY RPC. The lines which come straight through have no added impedance. The voltage/current from the generated leg comes from the motor, locally generated.

The generated leg voltage is lower than the line voltage under any sort of a load, because the motor has both resistance and inductance, and both cause a voltage drop. The voltage is going to be lower also because it is "back EMF" and inherently is lower than the incoming voltage, or the motor cannot run.

Then also, as current is drawn, there is a phase lag in the current, due to the RPC idler inductance.

When you use a large idler motor relative to the load, the lag is not a big deal, because it is small in relation to the load motor inductance.

With smaller RPC idlers, the inductance of the RPC is more comparable to that of the load motor. The balance capacitor corrects that back toward the other two legs by drawing "leading current" and counteracting the inductance (much the same as power factor correction capacitors). You cannot make a perfect correction because your no-load voltage will get out of range.

One could also insert an inductance in the other two legs (the straight through wires). That would increase voltage drop under load, but also give all three legs a similar phase shift, ultimately making the output more similar to what the powerco supplies.

 

[Balvar24]

I'll report back with what the amp clamp says. Hopefully I'll have time to do some set up this weekend if my parts all come in.

 

[jim rozen]

I'll report back with what the amp clamp says. Hopefully I'll have time to do some set up this weekend if my parts all come in.

I'm gonna go out on a limb here and, with your load motor connected but not heavily mechanically loaded, the two utility phases will be much larger than the manufactured leg. But the voltage balance will be about right and the load motor runs fine.