What's new
What's new

3-phase Xfrmr to condition output of RPC

specfab

Titanium
Joined
May 28, 2005
Location
AZ
I'm trying to determine what all my options might be to provide correct and balanced voltage to my Tree 325 CNC knee mill, from an RPC. I am hoping for some basic info about what is possible with 3-phase (or 3 single-phase) transformers.

1. Are 3-phase transformers typically built such that each phase coil could be tapped for a different voltage input?

2. In the case of supplying unbalanced voltage from a 3-phase source to a 3-phase transformer, does the transformer have any mediating effect on the unbalanced voltages between legs, or are the phases essentially independent within the transformer flux?

3. Would using a transformer on the output of an RPC create any oddball effects for the RPC, or other machines using the RPC output directly?

I have a commercially-built RPC that is feeding my shop, and no specific documentation for the unit (came with the CNC mill), and I can't afford a lot of "R&D" downtime sorting out the phase converter right now. I did get some good input previously about remedy for the specific balance issue, but it does involve dismantling the RPC and swapping caps in and out, so I haven't wanted to disable my shop power up to this point.

I have 240V residential supply coming in around 242V, and one leg of the RPC is working at about 265V. The Tree mill wants 230V +/- 5%; I would like to get pretty close to nominal, so I know that the power is not the problem while I'm trying to get the Tree servo/spindle drive back into operation.
 
1 They can be, but it may not be "typical",,, have to be made that way.

2) they are pretty independent, but not "entirely", because the typical type is made without the center leg that theoretically carries no flux if all are balanced. The better balanced they are, the more independent they are.......:D (in terms of input, which creates the flux via magnetizing current).

3) Not if the RPC starts with no load, not even the transformer on it.

Getting perfect balance on an RPC is a tough ask. They inherently start with an imbalance, because the generated leg is the back EMF, which is always less than the line input (has to be in order to draw current).

Possibly the best balancing method is an autotransformer on the generated leg. It is the most stable in voltage, with a known no-load voltage, but it does add series impedance.

The "balance capacitor" method makes for the least stable and most load-dependent generated leg voltage.
 
JST has it

IF measurements are taken under no load conditions, they are not meaningful.

Balance under load is the only real criteria.
 
Thanks for the info -- I hadn't really considered the fact that the transformer would actually be a load on the output of the RPC. The autotransformer idea is something to give serious thought to.

Any suggestions for good reference documents regarding RPC design and functionality? I think I need to get more details and background.
 
Thanks for the info -- I hadn't really considered the fact that the transformer would actually be a load on the output of the RPC. The autotransformer idea is something to give serious thought to.

Any suggestions for good reference documents regarding RPC design and functionality? I think I need to get more details and background.

Isn't there a subforum right here on PM?
 
JST has it

IF measurements are taken under no load conditions, they are not meaningful.

Balance under load is the only real criteria.

Sort-of.

The balance under load is actually what you want. You want it to BE balanced, of course.

The balance at no-load IS important, though, because you do not want a good loaded balance that ends up making the no-load voltage sky-high. That can happen if you ONLY look at the loaded balance, and try to get it to near perfection.

So you need to check both.
 








 
Back
Top