3 Phase to Single Phase

[rhenders]

I have a trailer with a 30 kW, 3 Phase (wye), 208 V diesel generator. The generator's primary purpose is to supply power to a Variable Frequency Drive (VFD). The VFD is controlled by software to start/stop/speed control a 20 HP pump motor (also 208 V, 3 Phase). All works as designed with no problems. As you can see, the generator is considerably oversized for the pump application. That was done on purpose so that we could increase the pump size if needed (this is a prototype system).

However, as is often the case, requirements change. We now find ourselves needing to power some 120 V, single phase circuits in addition to the VFD/pump. The short and easy method is just to tap off of one of the generator phases and power the equipment since 208/(square root of 3) = 120V. The generator manufacturer concurs that this is acceptable, but mentions that we shouldn't "draw too much current" from just one leg/phase. He didn't define what "too much current" means in real numbers, but I understand that to the extent possible, we need to keep the 120V load balanced so as not to run the generator in a highly unbalanced load situation.

The 120V loads are a 2.5 HP compressor (usually needs its own 20A circuit), some interior lighting (8 x 100W incandesent lights), a computer rack (up to 700W), and a 15A convenience outlet circuit (who knows - coffee pot, lap top, ...). Back of the envelope estimating means I might put the compressor on one 20A circuit, the computer rack on a 15A, and the convenience outlet on a separate 15A circuit. Each circuit could be powered from one of the 3 phases to help balance the loading, but this is still pretty lopsided since the compressor dominates. Adding the max circuit currents (breaker sizes) for worst case design means that I would need 50 Amps (20A + 15A + 15A) of 120 V power which is 6 kW neglecting power factor and other losses. So, by my estimate, the 6KW 120V load is only about 20% of the total generator capacity, but with the compressor, it's pretty unbalanced.

I want to do this right and I'm concerned that the relatively large 20A compressor load is a problem because it will be cycling on/off throughout the day. It will be a major portion (40%) of the 120 V load. Is it acceptable to just run this equipment from the three phases of the generator as I've described (assuming appropriate individual circuit breakers/protection), or should I find some other means of converting the 208/3 phase to 120V/single phase? Am I making this too hard? Other ideas/thoughts?

Thanks.
rhenders

 

[augie47]

You should not need a transformer since you already have 208Y/120 output from your generator.
In a much simplified approach, think of your 30 kw generator having an out put of 10kw per phase.
Adding a true 2.5 hp 120v load to the 20 HP would, IMO, be pushing the limit too far, however, if your 2.5 hp compressor truly works off a 20 amp 120v circuit I suspect that it is one of those "special ratings" and not truly 2.5HP. Check the actual rated current and use that for your calculations.
If the compressor and drive on one phase does not cause a problem, the other smaller loads could be fed off the other phases.
(The actual load on your 2-HP motor will have an effect also)

 

[Besoeker]

To the OP.
Is the generator 30kW or 30kVA?
In my experience, ratings are usually given in kVA rather than kW.

 

[rhenders]

Generator Size

Generator Size

Thanks for the reply. The generator is a 30 KW (steady state) advertised as a 38 kVA. I got it from General Power (aka Diesel Generators Miami). They are a great company to do business with by the way. Very responsive support and very helpful in integrating into the entire system.

Generator Link: http://www.dieselgeneratorsmiami.co...e-bcjd30-60it4-open-skid-mounted-version.html

 

[Besoeker]

Thanks for the reply. The generator is a 30 KW (steady state) advertised as a 38 kVA. I got it from General Power (aka Diesel Generators Miami).

Thanks for the information. The 38kVA gives you the rated current, about 105A in this case. That's the maximum current you can draw steady state.
The 30kW is most probably a limitation of the prime mover.

You can't run at 38kW at 1.0 PF. The current would be within the generator ration but the power would exceed capacity of the prime mover.
And you can't run at 30kW at 0.5 PF - that would exceed the generator current rating.

 

[rhenders]

Phase Imbalance Problem

Phase Imbalance Problem

...If the compressor and drive on one phase does not cause a problem, the other smaller loads could be fed off the other phases...

I'd definitely like to go the simple route, but you hit right on my main concern. It's not clear to me what happens if, for example, the compressor and drive on one phase DOES cause a problem. What's the likely result if this happens? Phase imbalance alarm on the generator controller? Compressor won't start/run right? Generator explodes spraying shrapnel across the city?

If it's just an alarm or the compressor whining, I'll give it a try. But if it's that shrapnel spraying thing, I don't know...

 

[SG-1]

Is it possible to change the compressor motor to 208 ?
Or change the motor to 3 phase.

 

[augie47]

No shrapnel....
There are so many factors involved, specialty with motors involved.
PF, as Besoeker wisely points out, motor inrush, delay with motor getting up to full speed and, as mentioned, whether the air compressor is actually a full 2.5 HP or a "sales" 2.5HP.
If it's critical, gather ALL the nameplate data and talk with the generator folks or an engineer.
Not critical, give it a try My gut feel is you will be o,k, but Besoeker or others may know better.;

Report back, if you will

 

[jim dungar]

.... Adding the max circuit currents (breaker sizes) for worst case design ....

It is not a worst case design so much as it might be a misdirected one.

What are the odds the the circuits will be 100% loaded?
You may be trying to solve problems that do not exist.


You may be overlooking problems that cannot be fixed.
What if the reality is that the compressor circuit is 50% loaded, at maximum, but the receptacle is normally at 0%?


You need to know your actual loads. You need to know you maximum unbalance.

 

[rhenders]

Worst Case Design

Worst Case Design

Well, I'm not stuck on the 2.5 HP model I mentioned, it is just one I've been considering (right size, SCFM, price, etc.). Even if it is a "real" 2.5 HP, that's approximately 15.5 steady state Amps (2.5 HP x 745 Watts/HP = 1862.5 Watts / 120 Volts = 15.52 Amps). Allowing an additional 25% for soft start inrush current brings it to a little over 19 Amps. Compressor mfg. says it will run on a standard 20 Amp circuit so this makes some sense. Power factor claim by the generator mfg. is 0.97. Even if they are a little optimistic, I'm not convinced PF is really a big design driver. As for converting the compressor to 208V, 3 Phase, I suppose I could do that. It is a realistic option, though one that is likely to result in a much larger compressor than I need. Maybe a better consideration would be to try to go with a slightly smaller 120 volt compressor.

I am an engineer, but have mostly concentrated on small signal electronics, software etc., thus the request for advice from this forum (you're the folks who apply the technology on a daily basis!). I understand the principles involved (PF, inrush current, etc.) and agree these are the variables that make this more difficult. I can't characterize the loads any better than I have described. For example, the compressor will turn on automaticlly whenever the control system has called for enough air to trip the low pressure switch on the compressor - no way to tell when - this is a non-deterministic parameter. As for the convenience outlets, that's another non-deterministic situation. No telling what/when/how much the operators will plug in. Because of these unknowns, I'm forced to consider the worst case scenario which is with all three circuits fully loaded and with the compressor cycling on and off at some rate.

At this point the options that have been suggested are as follows:
1) Try the compressor test and hope for the best
2) Go with a smaller compressor or a larger 208V, 3 phase compressor
3) Go with a dedicated transformer or digital converter

I think the best course of action is to run the 120V, 2.5 HP compressor from a standard 20 Amp circuit in the building (not on the generator). I'll rig up an O'scope and current meter and capture what really happens when the compressor comes on. Hopefully, this will help me to decide if the actual current ramp characteristics are as bad as I fear. This will also give me some hard data to discuss with the generator mfg.

Thanks again for all the advice/discussion. Any other thoughts are welcome.