# Single phase current requirements and minimum transformer requirements

To start, a generator is not an option for my situation.

I am looking to produce 208 V 3 phase delta using a rotary phase converter to power two robots. This is my first attempt at generating 3 phase so naturally I have a few questions:

1. The system below is attempting to power two robots each with 2 kVA power requirement. I was quoted an ad(adx)10 with a 5 kVA rating. However, I was also quoted a 15 kVA rated transformer at ~\$2800. This seems oversized by quite a bit. Is there any reason I should not be able to go with a 9 kVA (or even 6 kVA), say from here:
2. What would be the current draw from the main circuit breakers 240vac 1ph.
• Assuming my calculations are correct: three phase current is 22 A but what does this mean on the single phase feeding the phase converter?
3. Please provide any additional information that you think I may be missing.
4. More precisely the robots have power option of 200-230 V 3 phase or 400-480 V 3 phase If there is any reason to select a 240 Vac delta to 4NNsVac delta transformer I would be curious to hear the reasons.

The easiest way to think about current, I find, is to first look at power, then convert that power to the voltage system that is supplying that power. The voltage system can have 1, 2, or 3 phases, and various voltage amplitudes (120, 200-230, 380-440, etc). Then look at steady-state current first, then power factor, then in-rush current (start-up), then harmonics.

You have two (2) robots, each rated 2kVA at 200VAC 3-phase. Let's assume you want to deliver full power to each of these, so that's a total of 4kVA. kVA is close to kW, but not quite, but for our purposes let's stick with kVA.

The output of the rotary phase converter must be capable of delivering 4kVA total to the 3-phase load, but it will incur some power losses doing this, lets assume the losses are 10%, so at its input it must receive 4.4kVA. So the single-phase supply must deliver 4.4kVA to the rotary phase converter. We can now calculate the current that the single-phase supply must deliver:
4.4kVA / 240V = 18.33A
And twice this (>36A) if the voltage was only 120VAC.

Interesting to compare that to the current flowing in each of the three phases to the 3-phase load: If the voltage at the load is 120V active-neutral (==208V phase-phase):
4kVA / 3 / 120 = 11.1A.
If the voltage at the load is 240V active-neutral (==415V phase-phase):
4kVA / 3 / 240 = 5.56A.

So you can see why three-phase is so much better than single-phase when the power levels get over a couple of kW.

That would be the whole story if the load was a pure resistor. But the load is not a resistor (heater), it is some "robot controller", and I don't know how that behaves - does it behave like a resistor, or does it look like an induction motor (with a power factor of say, 0.7)? Does it draw a large inrush current when it is first powered up? Transformers are notorious for inrush, so are old-school VSDs. Does it generate harmonic current?

If any of these "unknowns" exist, then the current that has to be supplied to the load needs to increase, but the power delivered to the load does not necessarily increase. So the current rating of the rotary converter has to increase - even though your load power rating is the same.

So the supplier is probably concerned about this and is offering a larger unit to ensure you don't have problems - but I would have expected the over-size factor to be, say, an extra 50% or so; not a doubling of the power of the load.

I suggest talking to your supplier about power rather than current ratings. Also, suggest asking the robotics supplier about what type of load the robot controller presents to the AC supply.

The first answer to this question provides excellent analysis of using power and PF calculations to work out likely current levels, so I won't repeat that element.

However, in terms of a cost effectively implementing the 240V 3ph -> 208-230V 3ph reduction, unless you absolutely need isolation, an auto-transformer setup would be much cheaper than your proposed delta-delta setup.

If you go for the upper end of the tolerable output voltage range (230V), then you only need to drop the 240V by ~5%. Given the total load of 4kVA, the auto-transformer element will only need to "source" around 200VA in total. You could either purchase a complete 3ph auto-transformer for this function (which will be much smaller than the isolating unit), or implement using three discrete ~75VA 1ph transformers.

In either case, the unit should be much smaller physically, and cheaper.

I have personally used a very similar setup to adapt a 3KVA 1ph 60Hz compressor to 50Hz usage, where maintaining the V/F rating required me to reduce the 240V local supply voltage to 200V. The auto-transformer setup with discrete toroidal transformers worked perfectly, and was much cheaper and lighter than a fully isolating alternative.