I have a customer installing a new computer / server room.
They will have (280) pieces of equipment @ 1.5 Amps /120 volt and (10) pieces 2A / 208 volt - All single phase loads....
Plus they would like to have a 60A /3 Phase Rooftop unit to run of the generator also.

280 x 1.5A x 120V = 50,400 Watts
10 x 2A x 208V = 4,160 Watts
1 x 60A x 208V x 80PF x 1.732 / 1000 = 17,300 Watts


Calculated load = 71,860 Watts.

Is this correct?

Not exactly, you are assumming a PF 0f 1 on your single phase loads. Which is an error on the good side. You used VA on your single-phase loads. If you assume a .9 or .8 PF you will get something less. Long story short, you need a 100-KW generator based on the numbers you gave with not much room for growth.

Is there a UPS between the generator and data room. If not your client is really waisting a lot of money for nothing gained. If there is a UPS you size the generator at 150% of the UPS input power requirement with walk in capability, plus 125% of the HVAC load.

Edited to ask UPS?



[This message has been edited by dereckbc (edited 07-01-2004).]

So you would figure a PF of .9 for electronic equipment? Thanks for the response.

Go back and read my edited statement first. I would not go lower than .9, but I would not even talk to the client unless they are willing to put in dual mode UPS.

YES! A UPS will also be part of this installation. I figured I would start with the generator first to keep the question as simple as possible. UPS has not been sized yet.

Any suggestions are greatly appreciated.

Please discribe "Walk-in capability"
Thanks.

Higher end UPS designed to work with generators like Powerware and MGE will have control circuitry to "Walk In" the rectifiers, rather than all at once to prevent the generator from tripping. Also look for UPS systems with power factor correcting circuitry. Get with one of the UPS manufacture reps and let them get invloved in equipment specification.

Consider getting quotes from at least a couple of genset vendors based on your list of loads. Motor starting, switchmode power supplies, [UPSs] and generator automatic voltage regulators can be a problematic combination.

Producers like Kohler and Cat have sizing software with a ‘built-in’ checklist on stuff to consider for genset loads—often free for asking.

Otherwise, it can be a no-win dogfight—first between connected equipment, and soon to follow—between you and your client.




[This message has been edited by Bjarney (edited 07-01-2004).]

All comments above are good. But experience has tought me to make sure the UPS controls and the generator's regulator are compatable. Make sure you get this in writing from both the generator and UPS suppliers. If they are not compatable in the load enviroment you will have the dog fightmentioned above.
Load sizing alone is not going far enough.

Boy Howdy, NE. Excellent advice. Sounds like you may have learned this 'the hard way.'

That is part of the "Walk In" capability.

Just equipment? No other building systems, or lighting? The building is useless without them. You did mentioned... "Plus they would like to have a 60A /3 Phase Rooftop unit to run of the generator also" I would be wary of how that load interacts with the UPS. Maybe 2 gennies?

Then there's the Service Contract.... Maintenance on the Batteries, Generator, Testing, etc. If you're not willing to take that on, you might want to have a "quiet conversation" with a Company that would that specialize just in maintenance and system installs. Sub them the install with a little mark up for your trouble, bring them the inside circuits, and hand them a lifetime of maintenance..... One hand could wash the other there. Just an idea... You come out ahead, they take care of the problems.



[This message has been edited by e57 (edited 07-02-2004).]

I ran into this problem during a remodel of an existing building. Part of the remodel was to install an owner furnished UPS. The existing generator had been on site for over 20 years. Just getting tje generator up to code and operational was an PITA. The previous owner had not maintained it very well as he went broke.
The new owner already had the UPS from another site. This was a major headache we could not get the two to provide stable power at any time. Once the generator took over the load voltage started fluxuating wildly. The ups or gen seemed to compete to send voltageany were but at the correct level (480 3phase 150KVA). The manufactuers reps just pointed at each other and both told us to scrap the other item.
We finally convinced the owner to scrap them both. The cost over run was over 35K and the owner took a long time to pay. The owner had to get a new bank loan and the bank wanted to know why and argue about it. Owner knew he would have to pay since we did not supply any of it and it was hooked up exactly as specified.
I never want to hear that kind of machine screaming and see a gen shaking and pulling out tie down bolts again Also the UPS started stinking , thought it was going to burn.
The longest time they ran together was about 45 sec. the first time. Did not allow then to run that long ever again.
By the way both units ran very well on thier own if the other was not in the circuit.

We all have different levels of experience with equipment operation, but not all at the same time, so often hard-learned advice like nesparky wrote should be considered. Thin specs and low bids can precipitate this kind of insanity. Went through these kind of problems about 10-15 years ago at several facilities.

With an accumulation of switchmode supplies and UPS support, this site may be a “data center” with an inexperienced electrical designer writing bad specs and nearly useless drawings. Sometimes his client has high expectations that he has not detailed to the design/engineering folks figuring he’ll get a low-bid system that will somehow do the job to his unstated expectations. Some “fast-track”/”just-in time” commercial property management outfits are notorious for attempting to try to pull this off, and often it will involve gotta-have-it-last-week rehabilitation or tenant-improvement contracts, trying to snow another contractor with their obnoxious, urgent-but-still-tight-budget railroading. Some never learn, or they learn very well to feign the poor innocent dummy over and over.

Although to some degree everyone is on their own learning curve with standby systems and how they interact with loads, there are some basic ways that all but very precisely specified combinations of equipment operate. On the load end, motors and switchmode supplies being commodity items work best with a “stiff” electrical source that is a pure sinusoidal voltage, but their load on a given AC-power system can cause short {sub-cycle/harmonic} or longer-term [sluggish/dimming] currents that affect voltage at their AC-power terminals.

Generally on the source side, utility transfomers like to see ‘linear’ and fixed or slowly varying loads—ideally with a sinusoidal current that is best in phase wit the voltage waveform. Gensets are not as “stiff,” being pickier about load character, and UPSs are even less tolerant of changing load character; being even a less-stiff source. Gensets and UPSs can fight like the Hatfields and the McCoys if there haven’t been careful negotiations prior to the “meeting.”




[This message has been edited by Bjarney (edited 07-03-2004).]

My job involves working with experimental electric motors and variable speed drives. We regularly push systems into strange states to see what happens. If you are ever in the greater Boston area and want to see the simplest situation of _two_ devices duking it out, drop me a line.

A VSD has the same sort of output as a UPS inverter, taking an internal DC source and running it through transistors to produce AC. In the case of a UPS, you are trying to synthesize fixed voltage, fixed frequency AC, and in the case of a VSD you are synthesizing variable voltage, variable frequency AC to change the speed of the motor.

One operating state that we regularly see has the motor with only an inertial load (very low drag, but high mass) spinning at very near to synchronous speed. We can see tremendous sub-harmonic pulsing as the shaft speed oscillates up and down. We are seeing an oscillation where energy is going back and forth between the DC rail capacitors and the rotating mass on the motor. Current to the motor fluctuates between near zero and very high. The noise from the motor is something else, as we hear the VSD switching frequency, overlaid 60 Hz, and the sub-harmonic variation.

-Jon

Thanks for all the replys. Tho we have installed several engineered UPS / Generator systems for our clients in the past I needed to make sure I had my concerns covered in this design/build venture. I have had meetings with our generator & UPS distributers and everything is on track to working in harmony to get the design rolling. I was really looking for a quick reference to sizing for budgetary reasons and you all have provided plenty of info. I'll keep in touch!