RPC Theory

[CJ5Dave]

I have studied RPC’s for years. I can build good ones but still don’t understand them. I have decided they are magic. My faith in them is what makes them work.

 

[Eddyde]

I run both VFDs and an RPC in my shop. Both have their place, pros and cons. But if I had to choose only one, it would be the RPC. Cheap and easy to build, no need to rewire the machine's control circuits, no programming and virtually no maintenance.

 

[Chipper5783]

I run both VFDs and an RPC in my shop. Both have their place, pros and cons. But if I had to choose only one, it would be the RPC. Cheap and easy to build, no need to rewire the machine's control circuits, no programming and virtually no maintenance.

Yes, I have two machines on VFDs - they are fine (I got the VFDs for free when the company I worked for was upgrading equipment). I like it on the large drill press as I set it up with a reversing foot switch (though the VFD) - that is awesome. As pointed out, VFDs have their place.

I have 12 other machines (most of them have more than one motor - the surface grinder has 4 motors) - probably 25 - 3 phase motors in all. I’m running all that equipment on a phase converter (went to a Phase Perfect about 6 years ago, from the RPC I used for 35 years as it was too small) it is all working great. Switching to VFDs would be impractical.

 

[Larry$]

I built a RPC years ago. It fed two machines in my small shop for years. All parts except the caps were salvaged/used. Cost was nil. Sizing is based on the largest motor to start and the load it has to bring to full speed. Once the first motor is up to speed a 2nd one can be started. If I understand correctly the driven motor will not put out full power. How much of a hit does it take?? Does that mean things like pumps and fans will be overloading their motors??

VFD VS RPC: I'm moving my equipment to a place W/O 3 phase. I'm going to build a RPC again. I have four 3 phase machines. All will be connected to the one RPC. No changes need to be made to the machine controls. The coolant pumps will run just fine with no separate wiring needed. If I buy another 3 phase machine I won't need to add anything to my electrical system.

If I was using VFDs I'd have to have one for each 3 phase motor. I'd have to modify the controls. Expensive!
Down side to RPCs: The idler motor makes some noise & consumes some additional electricity. Normally they are located remotely and start stop controls are located near the machines. Just 24 Volt 3 light gage wires. Doesn't need to be in a conduit. Operates the pickup coil on a motor starter relay. When I built my last one, I cheeped way out and used a push button momentary contact switch to to connect the caps to the idler motor's 3rd leg. May not be "code!" I'll use a timer relay this time.

On Joe Pie's videos you can hear him start the RPC each time he is going to start a machine.

The advantage of VFDs is the ability to vary the speed of the one driven motor.

 

[strantor]

I think the classical way that these concepts are taught leads to confusion more often than not. And there seems not to be a consensus on the definition of "phase" either, which adds to the confusion.

First, what is "phase?"

If you're an Electronics Engineer, it's probably nothing more than a mathematical concept and you can use phasor diagrams and Spice simulations to demonstrate how North American residential mains are in fact, irrefutably, by the very mathematical definition of phase, two phases where one has opposite polarity or is shifted 180 degrees from the other (one is mathematically indistinguishable from the other, but mathematically one of them must be true).

If you're an electrician, when you say "phase" you're probably referring to an electromagnetic circuit. A single phase transformer has within it, a single EM field that is expanding and collapsing 60 times per second: IOW a single "phase." A single phase generator has a single coil inside being induced by a magnet spinning 3600 RPM to create that single 60Hz alternating EM field (technically it depends on # of generator poles but let's not make it complicated): IOW a single "phase." A 3 phase transformer has 3 EM fields expanding and collapsing and a 3 phase generator has 3 coils being induced.

I have to side with the Electricians on this one. We already know who won the debate, it's right there in the name, single phase power. So when we talk about phase, usually we are using the implied Electrician's conceptual definition but when we showing it plotted on an oscilloscope we are using an Electronics Engineer's tool so some of the Electronics Engineers' concepts get sneaked into the mix without announcing themselves.

What you're looking at on a scope screen is two halves of the same wave but one of them is being measured backwards. Nothing is out of phase. It's a single phase (a single EM circuit) and it can't be out of phase with itself. If you have a 240" long stick you can measure it from end to end and gets its length, simple enough. However the tool you've been given to measure it is a tape measure that has been nailed to the center of the stick, so you are forced to take two measurements and add them together. The Electronics Engineers' explanations, methods, and terminology will have you believing that there are actually two sticks end to end, with one having a positive length and the other having a negative length, and instead of adding the two measurements you subtract them because 120-(-120) = 240. And you will never win an argument with them in which you point out there's only one stick; they will math you to hell and back with Kirchoff's laws, phasor trig, and spite, repeatedly, until you give up. And they'll be right too, because they're using their definition of phase and you're using the other one. So make sure you know which "phase" you're talking about (and more importantly, which "phase" the other guy is talking about) before you start.



Now that's out of the way, from now on when I say "phase" I mean an EM circuit.

How does a RPC turn single phase into 3 phase? let's say you have a stool with two legs. It can pivot/rock in one axis only, left or right. Now add a leg. Now it can pivot/rock in 3 axes. By adding a single leg we added two "degrees of freedom" (or "constrains" depending on how you look at it). In the same way adding a single leg to single phase turns it into 3 phase.

Right, but how, physically, does it do it? Some people find it helpful to think of a RPC like a rotating transformer. But that doesn't explain where the 2nd and 3rd phases come from. I find it more helpful to think of it as a 3 phase generator, because a 3 phase generator has 3 separate coils arranged mechanically 120 degrees apart and that makes it intuitive to understand where its 3 electrical phases come from and why they are 120 electrical degrees apart. A 3 phase permanent magnet motor, like a BLDC or an AC servo motor, is also generator; 3 coils 120 degrees apart, with a magnet spinning in the middle. If you have one laying around, chuck it up in the lathe, spin it up, and measure the input (output). Nice, clean 3 phase power. A 3 phase induction motor also has 3 coils spaced 120 degrees apart and also can be used as a generator in the very same way, if you have a way to establish a magnetic field in the rotor. And it just so happens that we do have a way to do that: by sending single phase power into it!

A 3 phase motor will typically run on single phase as long as it is already spinning when you apply the single phase. It has trouble starting though. It's like riding a bicycle that only has one pedal. You try to take off on that bike, you're probably screwed unless the pedal is in just the right spot. 9 times out of 10 you'll just buzz, humm, overheat, and let the smoke out, just like a 3 phase motor try to start on single phase. But if someone gives you an extra pedal to get started, you can maintain cruising speed once you reach it and then remove the pedal. Those capacitors in a RPC are the temporary starting pedal. They aren't "push from behind," they aren't adding any power, just allowing the mains another temporary angle to push on the rotor to get it up to speed.

Some RPCs leave the capacitors engaged and some disengage them, some do even more. My RPC (American Rotary ADX30) has a whole bank of separately connected caps and an electronic monitoring circuit that switches different caps into and out of the circuit as needed, depending on load, to keep the phase voltages balanced. Why it needs to do that and why it works, is a little more technical than I am prepared to go right now, but it does work and is marginally better than not having the feature.

So, you have a 3 phase induction generator that functions as it own prime mover. It uses the single phase input both to spin itself and also to excite the rotor so that it can generate the 3rd leg. Instead of single phase power -> motor -> pulley -> belt -> pulley -> generator -> 3 phase power, it's just single phase power -> 3 phase motor acting both as a single phase motor and a 3 phase generator -> 3 phase power.

 

[rabler]

VFD vs RPC approaches a religious debate. I use both.

 

[kokoroko]

I have been reading about RPCs and trying to understand how they work. Most explanations have mis-statements or gloss over how they make true 3 phase power. For example, several sources say 'they take single phase power and add a generated phase to get 3 phase power.' So basically 1+1= 3. Since when? I realize the standard 240V power is "split phase" but it is not 2 phase.
I guess my fundamental misunderstanding is how does the RPC take the single phase L1 and L2 lines and offset these by 120 deg? Especially since these appear to be wired directly to two legs of the 3 phase in most diagrams?
I know we have a lot of people here who are very knowledgeable on this. Please help!

View attachment 498772
You can see in this diagram that L1 is connected to T1 directly and L2 is connected to T2. By definition these phases are 180 deg apart?

Rotary Phase Converters (RPCs) work by using a motor (idler) to generate a third phase from single-phase power. Here's a simplified explanation:

1. Single-phase input: L1 and L2 provide single-phase power.
2. Idler motor: The idler motor generates a third phase (L3) by rotating and creating a phase shift.
3. Balanced output: The RPC combines the original two phases with the generated third phase, creating a balanced three-phase power.

The idler motor's rotation effectively creates the necessary 120-degree phase shift, turning the single-phase input into three-phase output.

 

[rwm]

The clouds are starting to part for me. The biggest concept shift is to get away from measuring the potential from ground. I seems that the neutral point for 3 phase is above ground potential. Everyone who has contributed has helped in some way.

 

[strantor]

The clouds are starting to part for me. The biggest concept shift is to get away from measuring the potential from ground.

Agreed, this is where a lot of the confusion comes from. But know that if you start expressing voltage measurements referenced to anything other than ground, you're going to upset a lot of people whose entire grasp of electricity is built on a foundation of the sacred reference point.

I seems that the neutral point for 3 phase is above ground potential.

Eh, not really, or at least it shouldn't be. I think you are still trapped in the ground-referenced mindset even though you've recognized the trap and are trying to break free. Maybe you have defaulted to a neutral-based mindset which is functionally the same thing. You need to bust completely out of the box for the epiphany to fully materialize.

Think about it this way: Let's measure the output of a 240V 3 phase Delta isolation transformer.
You measure from L1 to ground, you get nothing. It is ungrounded. There is no ground.
You measure from L1 to neutr... wait, there's no neutral either.
So, does this transformer then have no voltage?
Yes, it does, measure from (L1 to L2) and (L1 to L3) and (L2 to L3) and you will see 240V each time.
If you have an oscilloscope with differential isolated probes, you can even look at the 3 waveforms (L1 to L2) and (L1 to L3) and (L2 to L3) simultaneously and see that they are 120 degrees out of phase with each other.
If you have an oscilloscope without differential isolated probes, you can still look at the 3 waveforms (L1 to L2) and (L1 to L3) and (L2 to L3), but not simultaneously. You'll only be able to look at two phases at a time.

Once you've got your head wrapped around that 3 phase delta Isolation transformer with no neutral and no ground, let's perform some unsanctioned modifications to that mental model of a transformer. Open the cover, choose one of the phase windings, and right in the middle of it, scratch off the enamel and solder a wire onto it. You can name the wire. You can call it Bob, Winston, neutral, whatever you like. Most people would name it neutral. I'm going to choose the phase between L1 and L2, install my wire, and name the wire Ronald. Now I can measure from Ron to L1, and see 120V. Ron to L2, also 120V. Ron to L3, 208V? What the heck? Where did that oddball come from? Here's a hint: this is just geometry. You don't need to know the electrical formulas for polyphase power, you can get these numbers by sketching triangles in CAD.



The black lines are the phase windings. The blue measurement line is the 208V measurement.

The effect of our modification is that now we can use that center tap (Ron) to connect 120V loads.

There is no voltage between any of these points and ground, because this is still an isolation transformer. L1 to ground: nothing. Ron to ground: zilch. L3 to ground: nada. L2 to ground, nope.

Now let's connect Ron to ground. We have just made the center grounded delta distribution transformer (AKA high leg delta AKA wild leg delta) that many are familiar with . And since we are playing on the public court now, let's use house rules and rename Ron to Neutral. If you measure between neutral and ground, you should not measure any voltage because they are connected together and should be at the same potential.

That high-leg delta is what is being generated by a RPC. We feed it this:


which is a single phase, already center-grounded, and it adds the third leg (notice I didn't say "it adds the third phase" - because it actually add two phases) to form the triangle above. That's why the graph on the first page of this thread shows the generated leg being so much bigger than the other two. It really isn't; the voltage output from a RPC on (L1 to L2) and (L1 to L3) and (L2 to L3) should each be 240V as expected, but it just looks like a "wild leg" when you use neutral or ground as a reference point instead of (L1 to L2) and (L1 to L3) and (L2 to L3), which makes more sense but most people aren't in the habit of doing it.

P.s. Instead of connecting Ron to ground, we could have connected L1 or L2 or L3 to ground, and then we would have had a corner grounded Delta distribution transformer (with a highly irregular center-tapped winding).




Ever measured a 480V circuit to ground? did you get around 277V? If so, it's because that's a WYE circuit and the geometry works for those too:


Again the black lines are the phase windings. The blue lines forming the triangle are the the 480V measurements.

While we're doing thought experiments, let's modify that 480V transformer above. Let's delete that center neutral. We can't actually delete the node where the phase windings are connected together, but we can cut the neutral wire and ground wire off, turning it into an isolation transformer so that no ground-referenced measurements are possible, and no neutral-referenced measurements are possible.
Can we still measure the output? Yes! From (L1 to L2) and (L1 to L3) and (L2 to L3).
We still have 480V 3 phase power available even in the absence of the holy reference point!

 

[Larry$]

Thank you for a god explanation. Can I ask for more information?
I had a shop in the old part of town where the electrician said to be careful when adding a circuit and not get the "wild leg." So it must have been a Delta transformer supply. The center tap would have been neutral and "bonded" (attached to earth ground.) True?? 240V 3 phase. When I moved to a newer location I got 208 Volt 3 phase and had to install transformers to get higher voltages (460+-.) needed by other machines.

I had machines that used a shift between wye and Delta motor windings to improve acceleration (?) and reduce inrush current (?) 6 wire motors.
How does that work? How does it reduce in rush current? There was a timer that was set to make the switch after a few seconds.