Transformer sizing (decent rule of thumb)

Status
Not open for further replies.
C

cubdh

Member
Hello everyone, I have a general question about sizing transformers. Currently we tell the technicians to figure out the max KVA load then load the transformer to 80% of the predicted max kVA then pick the next highest available transformer.
Now personally i feel we are grossly oversizing these transformers.

If you consider the ANSI Loading Guide (C57.91) it came to a sort of surprise to me that transformers can be loaded to 1.82 times its rated KVA for up to 1 hour. 1.57 x its rated KVA for 2 hours, 1.36 x its rated KVA for 4 hours 1.21 rated KVA for 8 hours. All this without compromising the Normal life span of the transformer.

These values are for a temp of 30 degrees celcius. You add 1.5% additional loading for every degree above 30. I will consider 37 degrees celcius so i will add 12 % additional loading to account for our peak hot climate.

Now this brings into the question our sizing practices. We are rating our transformes at 0.8 of expected peak KVA.

What would you consider a decent rule of thumb for us to rate our transformers. I am thinking that if the load is 110% to 120 % of the rated transformer size then it should work fine.

Example i am thinking if we expect a load of 110 KVA to 120 KVA peak load, then a 100 KVA transformer should be OK???

In your own personal experiences with transformers and how they handle loads what is the rule of thumb that we can go by to choose transformer capacity?????? I think our practice of rating transformers at 80% of peak load is not efficient. What do you use at your utility or what do you think is a good % tage?

Input would be greatly appreciated.
 
For peak loading losses may not be a concern, but it depends on how long your peak lasts (i.e cooling season vs.heating season is different than plant startup on Monday morning).

The winding (conductor) losses of the transformer vary with the square of the transformer loading. A transformer loaded to 120% will have 141% of the winding losses as one at 100%, while an 80% loading will only have 64% of the losses. Or course, you will have more core losses with a bigger transformer, so you need to take that into account.
 
Another pertinent issue would be future expansion..i have seen so many sites which upgrade their equipment ( pumps/HVAC..etc)...and if you don't have the cushion in transformer sizing you will end up buying a new transformer...( which takes at least 6 months for delivery nowadays...)

cheers.
 
Rule of thumb....

Rule of thumb....

I love rules of thumb! .. or rule of thumbs???

I spent two years with an independent Utility where they ran everything, and there was a history of old experts who did things their way.

The Line Crew foreman would get me to size a service transformer, and I would apply a similar method of calculation as the OP.

Unbeknownst to me until 6 months into the job, he would then halve the size I gave him to pick the transformer! And truth-be-told, I never heard of one being undersized!

But what he had going for him was that if a service did overload at some point in time, he had an inventory of tranformers to pick from and he'd swap it out in a couple hours.


John M


A
 
You will notice that I stayed out of this conversation? We have lots of rules of thumb for sizing our transformers but none of them fit the NEC or this conversation. :smile:
 
Thank you all for your input.
I will look into winding losses vs core losses and see where i can find a compromise. Yes we do allow for future load maturity. I think what i can do is take sample of commercial loads and also residential loads look at their 5 to 10 years history and figure out at what rate the load matures.

Mayanees brings up a good point. The few transformers that are overloaded can be swapped out with a larger one, now the question would be how do we know they are being over loaded. We do not have demand meters out there, only in special cases.
I need to figure out what the core losses are. Not to mention we pay in most cases about 90% of the transformer costs. And the bigger the more expensive.
 
Loading/sizing of transformers is going to be different depending on whose perspective. Utility has sets of guides they use, and each utility may use a different philosophy. If your designing it, then you need to consider your loading per NEC, and decide if you need spare (future) capacity. The size you give to the utility will invariably be different than what they install.
 
Well in regards to the NEC. Just like you say not all utitlies operate under the same guidelines. Even more so when considering different countries and customer demands. You will notice a drastic difference in the type and amount of usage from country to country. A 30,000 sq foot office building located in Austin Texas and a 30,000 sq foot office building located in my country will not at all have similar electricity usage.

In Austin residentials pay 5.8 cents US per kwh. Our utility charges about 25 cents US per khw. Almost 4 times as much. Then take into consideration that the average income in Austin Texas is $42,689 US per year and the avg income in my country is $6,000 US per year. 7 times less.

Our office buildings are not as well air conditioned and lighted as one in the Texas, (6 VA per sq ft? in our dreams) and i use Texas as an example because of similar weather.

I do use NEC guidlines when estimating load but I have noticed that the actual usage is slightly over 40% less than estimated by NEC. In this regard, NEC is definatly overkill for me. And i know that even in America NEC is used as a worst case scenario for as safe as possible operation.
 
Having been burned more than once with transformer and power supply sizing, and considering the cost of a larger unit is usually not a deal breaker, I usually am pretty generous in sizing them.
 
cubdh said:
I do use NEC guidlines when estimating load but I have noticed that the actual usage is slightly over 40% less than estimated by NEC. In this regard, NEC is definatly overkill for me. And i know that even in America NEC is used as a worst case scenario for as safe as possible operation.
Click to expand...

I bet that in general the same thing applies here as well. The NEC load calculations are designed to allow for the likely worst case of what might all be on at the same time. It is very rare that ever actually happens, but it is still safe if it does happen.
 
cubdh said:
Hello everyone, I have a general question about sizing transformers. Currently we tell the technicians to figure out the max KVA load then load the transformer to 80% of the predicted max kVA then pick the next highest available transformer.
Now personally i feel we are grossly oversizing these transformers.

If you consider the ANSI Loading Guide (C57.91) it came to a sort of surprise to me that transformers can be loaded to 1.82 times its rated KVA for up to 1 hour. 1.57 x its rated KVA for 2 hours, 1.36 x its rated KVA for 4 hours 1.21 rated KVA for 8 hours. All this without compromising the Normal life span of the transformer.

These values are for a temp of 30 degrees celcius. You add 1.5% additional loading for every degree above 30. I will consider 37 degrees celcius so i will add 12 % additional loading to account for our peak hot climate.
Why would you lay off sizing to a "techician" rather than an EE?
Now this brings into the question our sizing practices. We are rating our transformes at 0.8 of expected peak KVA.

What would you consider a decent rule of thumb for us to rate our transformers. I am thinking that if the load is 110% to 120 % of the rated transformer size then it should work fine.

Example i am thinking if we expect a load of 110 KVA to 120 KVA peak load, then a 100 KVA transformer should be OK???

In your own personal experiences with transformers and how they handle loads what is the rule of thumb that we can go by to choose transformer capacity?????? I think our practice of rating transformers at 80% of peak load is not efficient. What do you use at your utility or what do you think is a good % tage?

Input would be greatly appreciated.
Click to expand...
why would you have a technician make the call rather than a EE or yourself, the bidder?
 
Well our technicians are the ones that travel the areas and decide what type of infrastructure would be needed to connect new customers. They have been trained by management on transformer sizing practices. I am an EE with computer engineering background. I am very new to the Power Industry < 1 year. We have a Load management program in another department to reduce oversized transformers to smaller sizes, to reduce losses. I actually came across these things while doing a needs assesment for a GIS. For this reason i decided to take a closer look at our sizing practices. I would like to be directed and have good evidence to present to management about what i believe is over sizing in the hopes that we change our sizing methodology. If anyone has any literature on this topic, or can direct me to the right place, i would appreciate it.
I am also thinking of picking a sample of services, use customer billing (CIS) and do my own little research on load growth and how our transformers are sized. Maybe at 15 - 20 commercial services. That might be proof enough.
 
Status
Not open for further replies.
Top