transformer overload

[drdanshore]

A third world trasformer re manufacturer has built me an 25kw/28kva 3ph oil filled, convection air cooled, distribution transformer. I would like to put 38kva through it occasionally for 50% 10 min. cycles lasting one hour twice a day but I an planning to manage the load to the limit of the transformer.

If the transformer lives for only one year then I would be happy but I am not in a position to monitor the oil with an instrument to evaluate it's health.

the ambient is 20 deg. C but the name plate does not refer to that. I have a fan and an oil thermometer installed and it is pole mounted and I own the hydro plant.

I want to overload this safely. the only load is a motor with a power factor of 0.87 and i have a 5kva capacitor bank that I can trim back to the correct value. the compressor cycles between 25hp and 45hp.

I would like to target a capacitance to give me a 1.0 Pf at the offloaded 25Hp if this is safe for the equipment.

Following is an interesting comment from the pump forum from where the this thread began.

waross (Electrical)
29 Mar 09 19:36
Transformer capacity is based on self heating at the maximum rated ambient conditions. If you have an ambient significantly lower than the design ambient, (and many installations do), the actual overload based on temperature rise may be less than 15%. When a step increase in load is applied to a transformer it takes about 15 minutes for the transformer to respond thermally (reach a new, higher temperature). Transformers will often accept an overload with only a reduction in service life, but the reduced life may be many years.
Many transformers may be re-rated with the addition of external cooling.
The best approach may be to contact the transformer manufacturer and ask for recommendations for increasing the rating with increased cooling. This will avoid code violations.
An alternate is to compare the actual ambient to the rated ambient to see if the transformer is in fact overloaded. I would take this question to the electric power forum. Have accurate figures fr the rated ambient of the transformer and the actual ambient. Note, this approach may result in a code violation even if the transformer is safe.

 

[alehman]

Oil filled transformers built to industry standards can tolerate a fair amount of overload. 150% load for 10 minutes periodically probably will do it no harm, especially if you watch the temperature. I've seen small pad-mounted transformers run at 150% continuously for months without failing. It probably would violate any electrical codes.

Alan
----
"It’s always fun to do the impossible." - Walt Disney

 

[7anoter4]

If we shall use the approximate relation DeltaToil%=Iload%*(1-exp(-time/TC)) and for cooling Tcooled=Tbefore cooling*(exp(-time/TC)) where:
DeltaToil =maximum top oil temperature above the ambient
TC= thermal constant [2-6 hours]
According to ANSI the maximum top oil temperature has to be less then 65+30=95oC[65 oC above the ambient] for Iload=Irated. TC=2*60=120 minutes.
For initial 21.4 kVA load [76.4%] DeltaToil will be 50oC.
Then, at the first cycle of 38 KVA within 10 minutes this difference will reach 56.4 oC.
Within following 10 minutes the oil is cooled up to DeltaToil=51.9 oC as the load is now only 76.4%.
In the second cycle -of 38 KVA and 10 minutes DeltaToil=59 oC and after cooling DeltaToil=54.2oC.
The 3rd cycle [and the last] DeltaToil=61.3 oC and will decrease in a few hours up to 50oC if the compressor will continue at 21.4 kVA as initially.
So if the ambient will be 30oC then the top oil temperature will be 61.3+30=91.3[less than 95oC as permitted by ANSI] and according to ANSI the transformer total life has to be 30 years.

 

[jghrist]

The problem is that the transformer was built by a third world manufacturer. It was not designed to meet the ANSI loading requirements. Nobody except maybe the manufacturer can tell how much overload can be put on the tranformer for a short time.

 

[burnt2x]

Try a load run instead. There are some "3rd world" manufacturers who sometimes exceed ANSI specs! Ask a third party to verify temp rise of your trafo then load it accordingly.
My 2 cents.

 

[racobb]

You can probably overload the transformer by 100 percent regularly and it will last for at least a year. After all we are talking about a 28 kVA transformer. How much money can you afford to spend on testing to determine how long it will last? Load it up and see. Otherwise pick your size and mfg. more carefully next time.

Alan

Democracy is two wolves and a sheep deciding what to have for dinner. Liberty is a well armed sheep!
Ben Franklin

 

[retiredbum]

Just found this site and this is my first posting, so I hope I get it right without stepping on toes.

Utility owned pole mounted (lots of circulating air around the transformer) oil filled transformers are routinely loaded to 150%. The magic number to increase the size of the transformer is 185%. The loading on the transformer is based on a 15 minute average. A utility owned pad mounted oil filled transformer is changed at 110% and not allowed to exceed 120%. Different utilities may have different standards for changing out overloaded transformers.

As mentioned above, a standard utility "trick" to reduce the temperature on an overloaded transformer is to add cooling fans.

Having worked in third world countries, I would also be worried about the quality of the transformer. Quality control and safety are "foreign" concepts to many third world countries.

Now let me switch sides. Have you ever seen a transformer blow up? Hot oil goes everywhere. Not to mention fire hazard. Oh yes, what type of oil does it contain? Is it PCB oil? Coming from a third world country, it may very well have PCB or PCB contaminated oil in it. If so, under no circumstances do you want an oil spill. In fact, if you do not know if it is PCB oil or not, I would suggest you have it tested before use. Utilities have "quick test" kits.

I agree with your approach to monitor the temp. Since you own the plant, if you understand the risk, if the transformer is located in a "safe" place, if it does not contain PCB oil or is PCB contaminated, and the transformer does not represent a significant investment to you, the owner, then,......

 

[racobb]

Being in the states, did not think about the PCB issue this day and time. If it is PCB, I would not consider using it at a hydro plant. Even if it was contained somehow, the containment would probably cost more than the transformer.

Alan

Democracy is two wolves and a sheep deciding what to have for dinner. Liberty is a well armed sheep!
Ben Franklin

 

[jghrist]

If you're not in the USA where there is an irrational fear of all things PCB, don't worry about it. There is no real environmental hazard from PCB's, only a legal hazard in the USA.

The reason for the PCB phobia is that back in the 70's, an askarel (PCB) filled transformer caught fire. Normally non-flammable askarel will burn if exposed to a high enough temperature arc. When it burns, it produces dioxins, which were a suspected carcinogen. Anyway, the dioxins got all through a building causing a lot of concern and expense.

There are not many substances that have as much evidence of no long term health effects as PCBs. A major contamination of cooking oil in Japan resulted in thousands of people eating high concentrations of PCB with only some skin rash. A study of Westinghouse employees who worked for years with their arms immersed in PCB showed no increase in cancers.

 

[burnt2x]

jghrist,
You're right! Sometimes it's the attorneys who ponder and wonder about things they don't really understand much, and they did lots of regs to prevent "perceived" effects of lots of substances.