Rotary Phase Converter Voltage Outputs and Step Down Transformers

Hello all! I started this post to discuss the specific experience "By users only please" That are useing Rotary Phase Converters in their shops to run Fadal (Newer Taiwan Built Machines)or any other manufacturer as well. I just installed a Northern Power PL30_30HP RPC in my shop to power a 2015 Fadal (Taiwan Built Series 1) I am running the RPC off od 240 volt single pahase residential power (200 amp service) On the output of the RPC I am getting between 240 to 255 volt 3 phase power. Machine requirement is between 208-235 Volt max. Because I was told by many CNC techs I could eventually burn out my Syntec spindle drive @ $5k (or even my Yaskawa axis drives)I decided to play it safe and now have set up a 240 to 208 step down 30KVa isolation transformer with "Taps" to adjust the range of power out out for my needs. Dadal 4020B VMC has a 23 KVa requirement as I have the high Torgue spinslr motor option.
It seems no matter what I tried there is approximately a 15 volt variance from the 3 legs coming out of the RPC. My understanding is the at the "Manufactured leg" by the RPC is the higher voltage leg and the RPC engineer from Northern Phase Converters is telling me that when the converter is under load from the machining that higher voltage will drop down closer to the others. My ? to all of you with a RPC experience what have you seen with your output voltages measuring "Leg to Leg" on your Three Phases? Such as T1-T2= T1-T3= T2-T3= Again I get around a 15 volt variance (unloaded). American Rotary claims to be closer to all 3 phases with their digital control. It would be nice to hear back from all you RPC users out there. It would be great to compare the numbers unloaded and then voltage measured under machine load. Thanks for all your input thus far!

Gman4405 said:

........
It seems no matter what I tried there is approximately a 15 volt variance from the 3 legs coming out of the RPC. My understanding is the at the "Manufactured leg" by the RPC is the higher voltage leg and the RPC engineer from Northern Phase Converters is telling me that when the converter is under load from the machining that higher voltage will drop down closer to the others. My ? to all of you with a RPC experience what have you seen with your output voltages measuring "Leg to Leg" on your Three Phases? Such as T1-T2= T1-T3= T2-T3= Again I get around a 15 volt variance (unloaded). American Rotary claims to be closer to all 3 phases with their digital control. It would be nice to hear back from all you RPC users out there. It would be great to compare the numbers unloaded and then voltage measured under machine load. Thanks for all your input thus far!

Click to expand...

It is really about percentages, not so much volts. At 208, 15V variance is about 7%, which is a bit higher than ideal.

However, that is not unknown in RPCs. The generated leg is just that, "generated" by the motor, while the other two are just passed through from the power company. So the larger the RPC, the less drop at a given load.

Your options are to adjust the no-load/low-load voltage higher with capacitors, boost it with an autotransformer, typically using a step-down to boost by the paralleled secondary volts (for low boosts), or just accept it.

The transformer method is the most controlled means. I do not know your CNC, nor how sensitive it is. If you want the 15 volts back, a 12V/24V type transformer set with secondaries in parallel for 12V boost is probably the best.

Even though the actual boost will likely be less, the 24V connection is probably way too much.

Because I have only regular motors to run, I just tolerate the imbalance here.

I have an American Rotary RPC and the phase-phase voltages are very close on all 3. I think maybe a few volts. I needed to step it down for a machine so I recently bought and am using an American Rotary 15kva, 3ph auto transformer for this purpose. I have the same situation. Seems to work pretty nice. It's the size of a fat lunchbox.

Three Phase Transformer Archives - American Rotary

JST said:

It is really about percentages, not so much volts. At 208, 15V variance is about 7%, which is a bit higher than ideal.

However, that is not unknown in RPCs. The generated leg is just that, "generated" by the motor, while the other two are just passed through from the power company. So the larger the RPC, the less drop at a given load.

Your options are to adjust the no-load/low-load voltage higher with capacitors, boost it with an autotransformer, typically using a step-down to boost by the paralleled secondary volts (for low boosts), or just accept it.

The transformer method is the most controlled means. I do not know your CNC, nor how sensitive it is. If you want the 15 volts back, a 12V/24V type transformer set with secondaries in parallel for 12V boost is probably the best.

Even though the actual boost will likely be less, the 24V connection is probably way too much.

Because I have only regular motors to run, I just tolerate the imbalance here.

Click to expand...

So have you actually checked the Leg To Leg volts as I have described above? Can you please take a current test and provide leg to Leg voltage at the machine with the RPC running. Then do it again while under load such as taking a rough cut? This will truly help my self and others running a RPC and seeing the real life data . Thank you so much for your reply..........G

Gman4405 said:

"By users only please" ...!

Click to expand...

My converter is not the same as yours, nor is it running the same type of machine - so its data would probably not help.

jim rozen said:

My converter is not the same as yours, nor is it running the same type of machine - so its data would probably not help.

Click to expand...

My goal here was to compare other users setups so it doesn't matter if you have different equipment. Please share what you have and it will benefit all of us. If you could also tell us if everything is running great wityh the setup you have and /or if you had to make any adjustments along the way. Any information would be a positive contribution. Thanks for posting.

Pete Deal said:

I have an American Rotary RPC and the phase-phase voltages are very close on all 3. I think maybe a few volts. I needed to step it down for a machine so I recently bought and am using an American Rotary 15kva, 3ph auto transformer for this purpose. I have the same situation. Seems to work pretty nice. It's the size of a fat lunchbox.

Three Phase Transformer Archives - American Rotary

Click to expand...

Thanks Pete for your reply. Can you please take some reading with your voltmeter such as measureing from leg to leg of the 3 phase terminals @ the machine input such as T1-T2= ? volts T1-T3= ?volts T2-T3= ?volts
It would be great to see your real world experience and actual data.

L1, L2 = utility power.

For unloaded converter:

L1, L2 = 240.5 volts
L2, L3 = 216.3 (worst case 10 % inbalance)
L3, L1 = 223.8

For converter running one hp motor (lathe, spindle not loaded)

L1, L2 = 239.2 volts
L2, L3 = 213.5 (worst case 10 % inbalance)
L3, L1 = 220.4

For converter running one hp lathe motor (1/4 inch DOC in aluminum, agressive feed)

L1, L2 = 237.5 volts
L2, L3 = 212.2
L3, L1 = 210.2 (worst case 11.5 percent inbalance)

jim rozen said:

L1, L2 = utility power.

For unloaded converter:

L1, L2 = 240.5 volts
L2, L3 = 216.3 (worst case 10 % inbalance)
L3, L1 = 223.8

For converter running one hp motor (lathe, spindle not loaded)

L1, L2 = 239.2 volts
L2, L3 = 213.5 (worst case 10 % inbalance)
L3, L1 = 220.4

For converter running one hp lathe motor (1/4 inch DOC in aluminum, agressive feed)

L1, L2 = 237.5 volts
L2, L3 = 212.2
L3, L1 = 210.2 (worst case 11.5 percent inbalance)

Click to expand...

Thanks so much for takeing the time to check your voltages. Gives all ud RPC guys some feedback in real life scituations.Is your primary input to the RPC 240 volt single phase? Are you useing any type of stepdown transformer before or after the RPC? If you are what type is it? Thank you much for posting!

Input yes, single phase 240, no transformers, no capacitors, 5 hp idler started with a 1/8 hp pony motor:



What the ten percent looks like:

RPC vs VFD

American is what I run and very little variation in legs. Set up was done 7 years ago; if I was going to do it again I would look very seriously consider a VFD. Did think the VFD aspect?
Just for what it is worth.

I just measured mine. Machine running and cutting.

This is for an American Rotary 30hp RPC. No input step down to the RPC. Single phase input voltage is 250vac.

At transformer input (out of RPC)
A-B 248v, B-C 248v, C-A 244v

After transformer-
A-B 215v, B-C 215v, C-A 212v

Load was about 3-5 amps (Brother Speedio).

jim rozen said:

Input yes, single phase 240, no transformers, no capacitors, 5 hp idler started with a 1/8 hp pony motor:

What the ten percent looks like:

Click to expand...

Thank you much for posting specs!

doug8cat said:

American is what I run and very little variation in legs. Set up was done 7 years ago; if I was going to do it again I would look very seriously consider a VFD. Did think the VFD aspect?
Just for what it is worth.

Click to expand...

Im running a 15HP High Torgue motor on a North America 30HP RPC and my understanding is a VFD is not a good choice for my application.

The problem, G... is that no matter WHO you ask, or WHAT they've got, or WHAT they're powering, the answer is basically irrelevant, simply because the rotary converter is not an 'exact' device... it is effectively a type of 'spinning transformer', and the generated leg's output is dynamic... meaning... it changes with load.

As current load changes, the generated leg voltage changes... AND... PHASE ANGLE changes. In systems without capacitors (like Jim's plywood-archetecture), the amount of change that occurs as a result of capacitive reactance is minimal... because there's very little capacitance in his circuit. IF you apply balancing capacitors, then a change in load results in a change in current, which changes phase angle and voltage.

SO... start your machine, take all your measurements, then put your machine under a load, take measurements again... you'll find that your numbers are NOT the same. The generated leg is a moving target. Just like my single-phase welder conversion, Fitch didn't say it was 'perfect', it was just a slick and easy way to make 3-phase shop tools work in a single-phase environment.

Most guys who have CNC machines facing this issue will study the wiring, and figure out how to move all the voltage-picky circuits to the main leads, and save the generated leg for non-sensitive (like motor) loads. This not only moves the controls to the most stable power, it relieves the generated leg of phase-unnecessary loading. Still others will move ALL the controls to the two hots, and then slap a VFD on the spindle motor(s), and feed the VFD with single-phase, and that gives the CNC controller yet-another-aspect of machine tool operation that can be set up with a G-code...

On the subject of 240-255 on a machine set up for 208-235, yes, there's some machines out there (particularly imported) that won't like being up at 250ish, but in many cases, it isn't a huge deal. Running a drive motor on INSUFFICIENT voltage is oftentimes more a problem than having an EXCESS... basically, if the motor's insulation isn't weak, it should be fine at that point, and higher... and as the motor's applied voltage goes up, the current demand goes down.

That instability is exactly why the best plan is to set up any compensation ("balancing") for the loaded condition, and then accept the phase changes and voltage changes for the more lightly loaded condition where the maximum power is not an issue.

The problem comes where you have large loads and considerably smaller loads which may need full power. Then the optimum condition for the largest load may have an imbalance that potentially might affect the light load, when the heavy load is not applied.

The possible solution is to associate the "balance " capacitors with the load itself, so they are only applied when the load is in use. That can be a bit tricky to do, but will improve those cases where the load conditions just do not allow a "one size fits all" set of compensation capacitors.

Using a transformer to do the voltage compensation has fewer problems of that sort, but may have a higher overall voltage drop.

jim rozen said:

L1, L2 = utility power.

For unloaded converter:

L1, L2 = 240.5 volts
L2, L3 = 216.3 (worst case 10 % inbalance)
L3, L1 = 223.8

For converter running one hp motor (lathe, spindle not loaded)

L1, L2 = 239.2 volts
L2, L3 = 213.5 (worst case 10 % inbalance)
L3, L1 = 220.4

For converter running one hp lathe motor (1/4 inch DOC in aluminum, agressive feed)

L1, L2 = 237.5 volts
L2, L3 = 212.2
L3, L1 = 210.2 (worst case 11.5 percent inbalance)

Click to expand...

This single post has just saved me from weeks of stress.
Thanks.

be sure to check your phase converter caps monthly apparently. just had one go dead short almost on me and decided to mess up the spindle drive slightly.

Jim's numbers are a perfect example of the fact that the generated leg voltage is always lower than the input voltage, unless corrected with transformer or capacitors.

The RPC generated leg is often higher in voltage, because it usually has "balance capacitors". Jim's RPC does not, so he gets the raw motor output voltage, unmodified in voltage or phase. And, as you see, it is a bit lower than the input.

I do not really like the common statement that an RPC is a form of transformer, or "like a transformer", etc.... It is really a form of generator.

The input voltage/current induces a "field current" in the rotor. The resulting magnetic field in the rotor then induces a voltage in each of the three windings, just as a generator does, and in the correct phase for each. The generated leg has no externally applied voltage, so the only voltage from it is what results from the generator action.

I got the control to come on.

Will find out this weekend what my limits are.