RPC snub regen current possible?

[Garwood]

I use a 60HP RPC fed from a 45KVA single phase transformer. It does well supplying power for all my CNC's and manual machines, however, it does a bad job of handling regen current.

I have a Kitamura 5X that is a 10K RPM Cat50 machine with 35HP and I've got the Fanuc spindle drive setup for global current limiting at 100 amps. If I run it on 240 volts straight from the RPC it performs great except when I stop the spindle from higher RPM's it occasionally alarms out for overvoltage. To get around that, I have it on a transformer dropping the voltage at the machine down to 225. I notice at the lower voltage the performance isn't great with 100 amps current limiting. And it still puts the voltage spike on the line if I have other equipment running.

Would it be possible to use a regen resistor bank as a load to pull the wild leg voltage back down with an IGBT maybe? Anyone ever done anything like this? I thought I heard of someone doing this awhile back, but can't find it in the search results.

Thanks!

 

[rons]

A quality VFD has an IGBT that is used to pull down DC bus voltage. It's not turned on all at once, The IGBT will be pulsed. I think.

 

[Garwood]

I figured the voltage wasn't increasing much on the legs tied to the grid.

Kitamura recommends real phase I'm sure. I have 5 CNC's currently. The two smaller ones (10HP spindles) have braking resistors. The 3 big ones use dynamic braking.

I was figuring if I could turn that energy into heat it would help heat the part of the shop where the RPC lives in the winter and I could put the resistors outside in the summer.

 

[JST]

Was not able to find info on that model.

Are you sure it does regen to the power line?

If it does, it should be able to get rid of a lot of power to the regular two lines. The generated line will not accept as much or maybe any significant regen power, due to how it works. A lot depends on whether it monitors the bus, or each power line.

 

[johansen]

Friend of mine has a brother speedio and when stopping the spindle, it sends 70 amps rms back up the grid for about 120milliseconds. This is enough to push the 208v at the 20hp rpc up to 250vac (measuring from generated leg to neutral) (he has buck boost transformers to reduce the voltage at the machine to 200 volts.)

So anyhow I built a circuit to turn a solid state relay on, energizing a 5600 watt (240v) furnace heating element. I verified it does turn the heating element on within 1millisecond, it doesn't wait for the next line cycle. Anyhow for his system the 20 amps of diverted current is enough to bring the voltage down.

I have an unfinished unit that has 2, 5kw resistors, to be connected line to line rather than line to neutral. This would be 41 amps on the generated leg instead of less than 20.


Anyhow your spindle drive might have an option for a dump load resistor.

 

[Garwood]

Friend of mine has a brother speedio and when stopping the spindle, it sends 70 amps rms back up the grid for about 120milliseconds. This is enough to push the 208v at the 20hp rpc up to 250vac (measuring from generated leg to neutral) (he has buck boost transformers to reduce the voltage at the machine to 200 volts.)

So anyhow I built a circuit to turn a solid state relay on, energizing a 5600 watt (240v) furnace heating element. I verified it does turn the heating element on within 1millisecond, it doesn't wait for the next line cycle. Anyhow for his system the 20 amps of diverted current is enough to bring the voltage down.

I have an unfinished unit that has 2, 5kw resistors, to be connected line to line rather than line to neutral. This would be 41 amps on the generated leg instead of less than 20.


Anyhow your spindle drive might have an option for a dump load resistor.

This sounds extremely applicable to my situation.

Would you be interested in selling another one or sharing your circuit?

Thanks!

 

[johansen]

Yes can finish it and ship it.