Rotary Phase Converter

To power my mill and lathe I bought a while back I’ve got to design and install a circuit in my shop that will provide 480V 3 phase power. I’ve completed the design portion with the help of some folks on Practical Machinist forum.  Part of this circuit will require a rotary phase converter (RPC) which will “convert” 240V single phase  into 240V three phase. Once I have three phase  I’ll use a transformer to step it up to 480V.  This is all AC voltage in case there is any confusion.

Rotary phase converters have been around for around 100 years and as such I’ve done nothing new.  Simply put, a three phase motor, called the idler motor, is run on single phase AC and back EMF is generating the third leg.  I looked at several designs and worked out what I wanted to do.  Eventually, I stumbled across a design on Practical Machinist by Peter Haas that was the same as what I’d come up with.

Here’s my diagram of the circuit as I intend to build it.  L1 and L2 are the 240V wires coming from the wall.  L1 is also called A, L2 is also called B, and the leg I’ll be generating is called C.  My design for the RPC will start and run when power is supplied.  The control portion is elsewhere and not shown in the diagram.  I’ll reference this diagram as I go.


Single phase power won’t start a three phase motor by itself.  Therefore, you have to find a way to get it started.  I chose to use some starting capacitors (Cstart in the diagram) to give it an initial kick which allows single phase AC to keep it going.  The starting capacitors cannot remain in the circuit and must be “turned off” after the motor has started.  To do this, I used a potential relay which is a switch that is normally closed until the voltage across a coil is brought to zero.  This will give me a rotary phase converter that will start at the push of a button.  In the picture below is an orange cord that supplies the 240VAC line.  It is fed into a contactor which is controlled by a normal light switch.  These are only used in setting up and testing the RPC.  From the contactor, two wires (red (A) and black (B)) lead into the small black distribution block.  Everything is connected to the distribution block for simplicity.  The black cylinders are my stating capacitors and the black box above them is the potential relay.  The gray motor is, of course, the idler motor.

As an aside: capacitors are an electrical device that stores and releases energy.  The value of how much energy they will store is measured in units called a Farad.  A Farad is pretty big and the capacitors I’m using are rated in micro-Farads which is 10^-6 times a Farad.  Micro-Farads are denoted by the Greek letter mu and F.  Typically, people just type uF because we don’t want to change fonts.  When people talk about the size of capacitors they’re referring to how much charge it can hold and not the physical volume of it.  Capacitors come in set sizes and to get a bigger one you can wire multiple ones in parallel.  This will add their capacitance.

To determine the size of the capacitor used to start the motor, you keep adding more until it starts reliably and gets up to speed quickly.  If the start capacitors are too small the motor will shudder or start slowly.  Too large and the motor will start very quickly causing shock to the motor.  This is based on what I’ve read.  While I encountered the too little capacitance, I didn’t have too much.


To make sure my potential relay was functioning properly, I placed my clamp ammeter around the wire coming from it.  When I started the RPC the ammeter showed a large current that quickly dropped down to zero.  Just what it is supposed to do.


At this point the RPC is said to be unbalanced because the voltages on all three legs are not the same.  To correct this, balance capacitors are added between the A-C and B-C legs.  These capacitors will change the voltage in the legs and improve the power factor of the RPC.  I don’t want to delve off into power factor but will say it is a value from 0 to 1.  It is a measure of how electrically efficient a circuit is.  Unbalanced converters run hotter and their unbalanced voltages can damage the motors they run.

Without any balance capacitors in the circuit I had voltages of VAB=245 VAC=227.6 and VBC=219.5.  Due to the nature of a RPC it’s impossible to get identical voltages in all three legs all the time.  As such I want to have all the legs within 5% of each other.  Preferably, I’d like to have them even closer but as the RPC is loaded the voltage on the generated leg will drop.  Since the load isn’t constant in my application, I can’t tune the RPC to perfection.

I used Fitch William’s instructions for balancing my RPC.  He recommends installing capacitors (CP in the diagram above) between the A and C legs to bring VBC up to VAC.  Next, he says to install capacitors (CS in the diagram ab0ve) between the B and C legs to bring VAC above VAB.  Finally, to balance the converter you change CP and CS to bring VAC=VBC which are both above VAB.

For my converter I incrementally added CP up to 145uF which brought VAB=245.5 VAC=257.6 VBC=244.4.  Next, I started adding CS up to 35uf which gave VAB=245.7 VAC=258.2 VBC=252.6.  This was bringing the voltages up a little higher than I wanted.  So, I reduced CP and increased CS to 125uF and 55uF respectively.  This gave VAB=246.4 VAC=254.6 VBC=253.0.  I stopped at this point and will have to see how the RPC does under load. If the voltages drop below what I’m happy with I will tweak the capacitors.

In the picture below, the capacitors encircled in green are CP.  The ones encircled in red are CS.


The last step recommended by Fitch requires, you guessed it, even more capacitors.   These capacitors (CPF above) are added between A and C to increase the power factor.  Increasing the power factor has the effect of decreasing the current in L1 and L2.  I measured a current of 4.1A in L1 before adding CPF.  As instructed, I increased the value of CPF until I saw a dip and then returned to the value which gave the lowest current value of 3.4A.  My final CPF was 15 uF.  These capacitors are shown in the picture below nearest the top in the center.RPC5

I shot a short video of the RPC in operation and of it powering a grinder.  They’re pretty boring videos but I consider that to be a good thing in this case.  The first video shows the RPC starting and then running for a few.


The next video shows my three phase grinder running off of the RPC.  It only has a 1/3hp motor so it doesn’t really load down the converter.  Still, you can hear the idler motor’s RPM drop a bit while providing the needed inrush current for the grinder.  With the grinder running I measured the voltages again and found VAB=245.2 VAC=251.2 VBC=252.1.  Given the drop, I may be adding some more capacitors to increase VAC and VAC but will have to wait and see.



This entry was posted in Uncategorized. Bookmark the permalink.

3 Responses to Rotary Phase Converter

  1. Such a nice post! Keep it up.

  2. Carl Garrett says:

    How big is your idler motor in hp and what hp load do you think your RPC is good for? Thanks!

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s