Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations SSS148 on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Efficiency standard for Distribution Transformer 1

Status
Not open for further replies.

a10jp

Electrical
May 18, 2005
150
I have a general question on efficiency on distribution transformers. In our specification, we specifiy 20000/25000kVA transformers to be factory tested to the standard of NEMA TP 1. In this case, according to the standard, which specifies the minimum efficiency levels for NEMA Class I liquid filled distribution transformers, it says for for transformers 2500kVA and beyond, the minimum efficiency shall be 99.4%. This is from Table 4-1 in Section 4, which listed reference condition for Load Loss at 55% and at 50% nameplate loading.

The manufacturer factory test reports came out to, as an example, 99.37%/99.36%/99.35% forone transformer and 99.35%/99.34%/99.32% for another, etc. The results are close to 99.4%, but not quite.

When I asked the manufacturer to justify these test results, and to clarify what standard was used in conducting these tests, the reply was they categorize this as a power transformer vs a distribution transformers, the standard does not apply. The efficiency tested just happens to come out close to NEMA TP-1, even though the standard does not apply.

FUrther, the manufacture says that ANSI, NEMA nor IEEE do not specify efficiency for power transformers above 25000kVA for the reason that hihger efficiencyies require superior core steel and that may not be readily available. Current standards on power transformers govern only safety and construction but not on efficiency. The Mfr tried to optimize the design of power trans taking into account material cost that result in the least costs of ownership to the end-user under normal conditions....

This sounds like a standard industry response. My questions are:

1. Is it true, that efficiency does not exist for large transformers similiar to what we used?

2. Is 2500kVA the maximum limit on transformer capacity where one can truly comment on efficiency, and anything else larger than that is merely luck?

3. Even if it is true no standards exist for such large transformers, there must be a minimum efficiency you can use to determine the level of construction we are getting from the mfr?

4. One last thing, am I being too picky? Is perhaps 99.32% considered acceptable? Afterall, the standard only goes up to 2500kVA. It must be for a reason.

I like to understand more of the economic as well as technical justifications based on what was discussed above. If anyone has any comments, I will be glad to hear them. Thanks.
 
Replies continue below

Recommended for you

1. Efficiency exists, but minimums are not required by standards.

2 through 4: How much are you willing to pay for a high efficiency transformer? At some point, the economics don't work when the extra first cost exceeds the value of the losses saved over the transformer lifetime. You need to evaluate losses with a dollar figure that will allow a manufacturer to design a unit that will suit.
 
You can get very high efficiency, but you have to pay for it and that is done by specifying an efficiency or evaluating the present value of losses as stevenal references. This is commonly done for units of this size.

Chalk it up to experience.
 
When buying a transformer, efficiency has to be agreed upon.Because transformer cost/price depends on the efficiency.For a rating,within the efficiency band normally achievable,the price band is 1:2.ie a very high efficient unit can cost you double than a low efficiency unit.Some times efficiency is suggested by industry standards,othertimes it will be specified by buyer or qouted by vendor based one loss capitalisation specified by buyer.
 
You usually can get the efficiency that you want. What I would do in this situation is to compare the TOC of the transformers at 99.4% efficiency with the ones coming in at 99.35% etc and make a decision based on the ecconomics. You may very well find that the TOC of the transformers are very close to or a bit lower than the one at your bench mark efficiency, as high efficiency transformers as explained by prc can be pretty expensive and thus the initial capital outlay may cause the TOC of both transformers to be close or the ones at the little lower efficiency to be more ecconomic to own.
 
Hi everyone,

I was thinking about what OPMGR1 has said earlier, too. I really like to know can you determine exactly what the upfront cost is vs exactly what efficienty you are going to get. That is, can you design your transfomrer construction so well, that the manufacturing can give you exactly what efficieny you asked for? Or is some allowance/tolerance permitted at the end?

Earlier in this discussion by stevenal, he said that minimum is not required by standards. Is this statement true? I thought what NEMA TP 1 is specifying is the minimumm efficiency for the specific class of transforms?

Just like to hear what everyone's opinion is?

Andrew
 
I wasn't aware of any efficiency standard, but I don't have access to the referenced NEMA standard. Is this a new standard? The DOE has issued a ruling on distribution transformer energy conservation but this wouldn't apply to 20/25 MVA power transformers. We normally recommend evaluating losses instead of specifying them - this allows each manufacturer to optimally design the tranformer for their particular cost structure. The evaluation factors can be determined from the user's energy costs and the expected load profile of the transformer.

Knowing the efficiency at a single load point is not as useful as knowing the no-load and load losses. With the no-load and load losses, you can determine the efficiency and any load and the loss costs for any load profile.
 
There are NEMA standards for smaller transformers, but not this size.

Evaluation of losses is the best approach. This requires you to estimate what the energy loss costs actually are to you.
 
Hi jghrist, I have uploaded the NEMA table showing the efficiency for size up to 2500KVA.

Because losses and efficiency are related, one calcuates the other, I assumed they mean the same thing.

Our spec did say that if the efficiency is lower than the NEMA Class 1 Eff, the product is unacceptable. Based on what we discussed above, guess we have a defective spec.

Switching the topic, how do you determine what is the optimal losses you should have from a manufacturer viewpoint? What is the design criteria to optimize ownership cost vs upfront cost of construction?




 
What dpc and jghrist indicated is the best way to qualify a transformer. Looking at the NEMA extract, the std. mention other ways of evaluating transformer and even though I dont have that standard I would assume that the Total Owning Cost (TOC) method is described. In addition to this you should build tolerances within your specification using these other stds IEEE, NEMA etc as a guide. The book "Energy Efficient Transformers" by Kennedy gives a good account of how to determine TOC. If I am in your shoes I would work with these "out of standard efficiencies" and apply a penalty to the initial cost of these transformers based on the new losses figure over the life of the transformer.
 
Our spec did say that if the efficiency is lower than the NEMA Class 1 Eff, the product is unacceptable. Based on what we discussed above, guess we have a defective spec.
The problem is that NEMA Table 4-1 does not define a Class 1 efficiency for transformers above 2500 kVA.
Because losses and efficiency are related, one calcuates the other, I assumed they mean the same thing.
You can calculate efficiency at 50% load from the losses, but you cannot calculate the no-load and load losses separately from the efficiency at 50% load.

See stevenal's 12/16 link or opmgr1's reference for the method to optimize ownership costs. You need to ask a manufacturer how to optimizing losses from their standpoint; I just know it can be done.
 
Just got me thinking about some of our larger power transformers. Similarly to the rest of the posts, our standard on minimum efficiency performmance for distribution transformers (Australian Standard - AS 2374.1.2) goes up to 2500kVA.

Using the same formula as for the standard;

eff = 0.5*KVA/(0.5*KVA+0.25*LOADLOSS+NLL)*100

We get for the following transformer sizes, the following efficiencies. A number of each MVA transformer were calculated. They are all 3 phase core type.

120MVA - 99.68%
50MVA - 99.60%, 99.58%
33MVA - 99.54%
19MVA - 99.72%, 99.71%, 99.7%

Hope they are of interest to the forum.

ausphil
 
It should be kept in mind that these efficiencies only apply at full load, or at whatever load they are based on.
 
jghrist,thank you for putting in that link.I have been looking for these Rules for sometime.
 
Hi ausphil, may I ask at what % loadloss were these values computed, just like what dpc has stated? The NEMA TP-1 uses 50% nameplate load for upto 2500KVA in their standard.
 
One more thing to consider when calculating power transformer efficiencies and economics:

I'm guessing that a10jp's 20000/25000kVA unit is an OA/FA transformer. The overall efficiency will change for the load regions where the cooling fans are or are not operating. The same holds true for an FOA transformer. The efficiencies in each region should be calculated and compared to the expected load profile to determine the economic costs of losses.
 
sorry, I should have put that in, efficiencies are at 50% NLL.

ausphil
 
Hi PHovnanian, what you said is true. Under normal working situation, the transformer is expecetd to operate in 20MVA capacity under natural cooling mode, KNAN, and 25MVA for KNAF. The efficiency calcuation done by the mfr was conducted at 20MVA and at 100% NLL. WHen we did the calcuation using 50%NLL the result came out closer to the figures shown in ausphil's tabulation. In any case, the evaluated efficiency data all came out ot less than the expected "loss" profile supplied by the mfr, (For the lack of typical loss data, I used the mfr loss profile instead.) The discussion presented above by everyone are all correct, and I am grateful for such a great discussion on this subject.
 
Also be aware that the ANSI and IEC standards differ slightly on the definition of transformer efficiency, IIRC. I believe for fan-cooled ratings, the ANSI standards require the cooling fan power requirements to be included in the transformer losses, while IEC standards do not require this. But I'm going from memory, so don't quote me.
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor