I currently have a project underway involving 17 geo package units serving 2 warehouse like structures. My question is, can anyone here point me to where I can learn about the efficiencies of step-down transformers that are used to supply the 208Y 3-phase power to the geo units?
 
These 2 warehouse structures house an indoor sports business. Plenty of soccer, football and baseball fields, all with AstroTurf laid down.
 
The 2 structures are served by a closed water loop, of unknown pipe diameter, using a borehole field of unknown quantity of boreholes. Geo (package) units are evenly spaced around the fields, hung from the ceiling, each connected to the water loop. They are 208 VAC 3 phase, along with 2 water pumps configured in a push-pull fashion.

Electric service to the 2 structures is 480 VAC 3 phase.  The geo package units were supposed to be 480 VAC 3 phase, but were incorrectly ordered and received as 208 VAC 3 phase, and, installed.

The building owner chose a solution of leaving the geo package units as is and instead changing the supply voltage at the circuit breaker boxes to 208 VAC 3 phase.

The building owner had on hand many small 30 KVA 480 Delta to 208 Wye transformers, and installed these, versus installing one larger, appropriately sized transformer.

My question is, where could I go to research if there's efficiency gains to be had by removing the 6 small 30 KVA transformers and installing instead one much larger transformer.

The question being put to me is, how much am I (building owner) paying in utility cost for the inefficiencies of 6 small 30 KVA transformers, and, is there an economic case to be made for changing to one large transformer?

I know this is a different subject than usual here.  But there are a lot of really smart people here, and, this still is a geo related subject.  Any guidance or direction anyone can provide would be highly appreciated!

Many thanks!

Best regards,

Bill

Maybe the electrical engineers that work for the power company could answer your question.  In my area, with my subdivision, I have found the power company to ve very responsive to similar questions.

Bill,
One devils-advocate question that occurred to me is,
Do the six small transformers offer any redundancy at all?
Or are they just divided among the 17 loads?

I don't know anything about the failure rate of transformers :-)

Should be around 97% efficient, 98% if combined. Compare that to perhaps a 10% jump if you could combine all the electric motors into one big one.

The fact that these were spare transformers may indicate they are older. You should be able to find a spec online for the model number that indicates their efficiency. If I look at a generic NEMA premium efficiency transformer today, I see 98.25% for a 30 KVA vs. 98.95% for a 200 KVA, so on like equipment it is not a show stopper. Also need to consider if you centralize to one transformer and start pushing more amperage at 208 further distances, you need to look at increased line losses. Probably no net savings and if there were any, it probably couldn't justify additional distribution run costs.

It might be worth looking at replacing 5 of the heat pumps with the latest high efficiency ones and then stage them so they run a lot more than the older ones. A 20% improvement in something that runs nearly all the time may be a pretty good ROI.

Much appreciate the comments.


My initial conclusion:

If six 30 KVA 480D-208Y/120 VAC transformers were replaced with one 180 KVA transformer of the same pri/sec, the energy reduction benefit would be approximately $42/month, assuming 13 cents/kWh.


Calculations:

One 30 KVA transformer specs at 186 watts loss at 35% loading.  (Square D Premium 30 product family)
One 150 KVA transformer specs at 576 W.
One 225 KVA transformer specs at 764 W.

Six 30 KVA transformers = 1113 W loss

The equivalent of six 30 KVA transformers would be theoretically a 180 KVA transformer. 180 KVA is not a standard size, so, interpolating between a 150 and a 225 yields 670 Watts.

1113 W - 670 = 443 W additional loss due to using six 30 KVA transformers versus one 180 KVA transformer.

Assuming constant 7x24x365 loading at 35%:

443 W * 24 * 365 / 12 / 1000 = 323 kWh loss / month

At 13 cents:

323 kWh/mo * $0.13/kWh = $42/month


Comments:

What I don't know is the economics of changing out six 30 KVA transformers for one 225 KVA transformer (first standard size above 6*30=180 KVA).

What I do know is that these kinds of projects need 3 year or less break even periods to get the attention of the kind of customers I have.  Thus:

36 months * $42/month savings = $1512 savings in 3 years.

While I don't know the cost of a 225 KVA transformer, nor the associated electrical engineering and installation costs, for a '6 for 1 swapout,' I think it's a fair statement, with just this 'back of the envelope' calculation, that it can't be done $1512 or less, and thus would not be a cost effective change.

In summary, I think the building is stuck for a long time with the six 30 KVA transformers, and, that this was probably a reasonable action in response to the mis-ordered geo units, assuming the 30 KVA transformers were available to the building owner at no or little cost (versus putting in a 225 KVA transformer).


Many thanks for the comments.

Best regards,

Bill

P.S.: After typing the above, and after further reviewing the Square D Catalog, I see that I incorrectly assumed full load losses on a 7x24x365 basis.  In fact, the business is only in operation between 5 PM and Midnight.  That's 7 hours per day, not 24.  During the remaining 13 hours of the day, there's no load on the transformers.  They do in fact have some 'no load' losses, but, it's about half.  And therefore, $42/month savings is probably much too high of an estimated savings.  Still, there's no point in redoing the above calculations because the potential project's economics were already too woeful for serious consideration.  Bill

Posted By pachai on 06 Nov 2013 12:33 PM
Bill,
One devils-advocate question that occurred to me is,
Do the six small transformers offer any redundancy at all?
Or are they just divided among the 17 loads?

I don't know anything about the failure rate of transformers :-)

This is a good question, as the type of business in the structure would not be able to proceed in the summer time if the geo units were lost.

Turns out, the geo units are evenly distributed across the six transformers, and that there isn't a '5 + 1' redundancy implementation.

Best regards,

Bill