Transformer losses

[NicolasK]

If a transformer (three or single phase, pole or pad mount) is only 25, 50 or 75% fully loaded what amount of losses are there compared to a fully loaded transformer?? I'm trying to determine if it is cost effective to change out a transformer that is not fully loaded by comparing the cost of change out to the cost of losses over the lifetime of the transformer.

Thank you in advance.

 

[stevenal]

No-load losses remain the same. Load losses go down proportionately with the square of the load current.

 

[dpc]

Just to give you an order of magnitude feel for what stevenal is talking about I happen to be looking at some actual transformer test data:

Transformer size - 5000 kVA

No-Load losses = 4532.11 Watts

Load losses at 5000 kVA = 27,839 Watts

Total losses (at 5000 kVA)= 32,371 Watts

At half load, the No-load losses are the same, but load losses are only 25% of full-load value.

 

[NicolasK]

If the losses are half when a transformer is only 25 or 50% loaded, then why is the rule of thumb to install a transformer that is approximately equal to the load?

What am I missing?

 

[jghrist]

NicolasK,

Because losses are not the only cost. You have to consider the cost of the transformer.

 

[NicolasK]

True jghrist, but when the cost to change the transformer out is in the thousands of dollars (cost to my company and the customer due to down time) and the cost difference between the transformers is only a thousand......

I thought there would be more to consider.

 

[jghrist]

You also have to consider the load profile. If the load on the transformer is not constant, then the load losses will vary with time but the no-load losses will not.

 

[edison123]

Which trafo will have lower load loss ? 1000 KVA with 50% load or 500 KVA with 100% load ?

 

[jghrist]

edison123,

If both transformers had the same % load loss at rated load, then the 1000 kVA with 50% load will have half the loss of the 500 kVA with 100% load. Actually, % load loss is lower for larger transformers, so the 1000 kVA transformer will have less than half the loss.

 

[stevenal]

Nicolask,

You mis-stated dpc. He said a half loaded transformer had a quarter ((1/2)^2) the load loss of the same fully loaded transformer.

But it's the actual losses you pay for. The smaller transformer will usually have smaller full load losses, and no load losses. The no load losses can really add up. Unless switched off, no load losses are there 24/7, no matter what the loading (or revenue).

You are correct that installation labor and down time are valid costs as well. Isn't engineering fun? Your rule of thumb might have been valid when labor was cheaper. If load growth is in the picture, I'd throw it right out.

 

[tinfoil]

At our utility, when we are purchasing new transformers, we use a cost-of-loss equation allong the lines of

Cost = A * kW(no-load) + B * kw(full load) where A is about 6.5 times B


Logic:

Overall, our transformers wind up peak loaded somewhere in the ballpart of 50-60% of nameplate.

Our annual peak load factor is furthermore about 50%. That is, for a given part of the system (statistically each xmfr), the maximum load achieved at any time of the year is roughly double the average load for all 8760 hours of the year.

If the load factor is 0.5, then the Loss Factor is 0.5^2, or 0.25

Combine these together, and you get an overall factor of about 6.5

{Our actual calcs are a bit hairier, because we use real loading data, and those xmfrs loaded closer to 100% skew the weighted averages}.

In order to arrive at A and B, you also need to consider things like anticipated load growth rates, probability of premature failures, and (most importantly), the projected present-worth cost of each kWh of losses over the life of the transformer, given expected rate changes and the cost-of-capital. It is not a simple calculation.

 

[cuky2000]

If a transformer (three or single phase, pole or pad mount) is only 25, 50 or 75% fully loaded what amount of losses are there compared to a fully loaded transformer?

The example below calculates the efficiency of a typical pole mounted distribution transformer for 3 power factors. Considering the case of 100% PF, the reduction in efficiency when the transformer is loaded bellow 100% is around 8.2% @ 25% loading factor(LF), 5.5% @ 50% LF and 2.75% for 75% LF. For larger transformers this difference will be smaller. Keep in mind that the total efficiency is greater than 95%.

This made very hard to justify the investment to replace existing transformer since the payback may be too long.
For final conclusion it is recommended that you do the final detail calc.

I hope this could help