Actual Power vs. Installed Capability

[zdas04]

I'm looking at acquiring a plant that includes a big air compressor. The motor on the compressor is rated at 9,000 kW, but the compression requirement is closer to 6,000 kW.

When I'm estimating steady-state electrical cost will I be seeing the 9,000 kW motor-rating or will I be seeing the 6,000 kW compression load.

David Simpson, PE
MuleShoe Engineering

http://www.muleshoe-eng.com

Please see FAQ731-376 for tips on how to make the best use of Eng-Tips Fora.

The harder I work, the luckier I seem

 

[dpc]

You'll see roughly the mechanical load plus the motor losses.
This will be much less than the 9000 kW motor rating.

 

[zdas04]

Thanks, dpc. That is the way I was starting, but I got nervous that my tiny knowledge of electricity would cause me to understate op costs by 1/3 (when the expected electric bill for the whole plant is over $30,000/day that mistake would be major).

David

 

[rbulsara]

Is this an existing facility? If so isn't there a way to get old bills?

Also I would think you have accounted for other loads in the plant. Also the electrical demand charges makes significnant part of industrial billing. Also demand charges are racheted, meaning once you reach ceritain maximum, you pay for it for next 3 or 6 months. The starting current of this motor will most likely determine the peak demand.

 

[zdas04]

rbulsara,
We're buying an existing plant that was in a facility with a huge co-gen (which we're not getting) so the old records are missing or bizare.

We're looking at several peak-shaving possibilities (we're leaning towards standby generators). On the utility we're going to be on, the "demand charge" is set for the highest 15 minutes in a calendar year and it is set for the following calendar year.

Luckily the motor has a pretty effective soft start so it shouldn't be too hard to "peak shave" while starting it and several other very large loads.

Right now we're at the "feasibility" stage (mostly doing economics), when we get to the design stage there will be competent electrical types involved. Right now it is just a flange-head out of his depth.

David

David

 

[jghrist]

I would expect starting current, being of short duration and low power factor, to be of little significance in the 15 minute integrated kW demand.

 

[itsmoked]

Would it be worth it to actually interlock the big loads so that no two start at the same time? Or no two start in the same 15 minutes?

 

[dpc]

I agree with jghrist that motor starting is rarely an issue for reducing peak demand.

Just a word of caution regarding "peak shaving" using generators - I would not base any business plan or cost-benefit analysis on money saved by peak shaving through on-site generators, especially when you have a 12 month ratchet and a 15 min demand window.

If your equipment has one bad day,or even one bad hour, you pay the price for the next twelve months.

I have seen a lot of generators installed with an eye toward peak shaving, and I've never seen a single installation that lived up to expectations with regards to demand reduction. If the generator doesn't make sense from an energy-production standpoint, I would be very skeptical about its merits.

On the other hand, if you can reduce your peak by shutting down non-essential loads, that is a strategy that is much more reliable and can provide some payback. But even that is tough when you have that 12 month ratchet.

Just some observations based on scar tissue I have picked up over the years and, of course, YMMV.

 

[zdas04]

Thanks dpc, we'll be looking at options with a careful eye.

David

 

[jghrist]

An exception to dpc's scepticism about using generation for peak shaving is when the demand charge is based on a coincident peak with the supplier's system demand. This would be unusual for an industry, but not so unusual for municipal utilities.

We have some municipal utility clients who have made very profitable use of peak shaving generation with a coincident peak rate structure. They have real time access to their suppliers system load data and can cut a lot off of their demand charge with as little as 3-4 hours operation per month. In one case, the utility has a ratched coincident peak rate structure and saved millions with a single summer hour of operation.

With a non-coincident peak rate structure, I agree with dpc. You can eat up a lot of fuel trying to reduce your peak because you have to run the generators every time you approach your peak use. Unless you have a well-defined, sharp load peak, you have to run a lot of hours.