TRANSFOMRER TEMPERATURE READING VS TEMPERATURE RISE 2

[arvinolga]

Can anybody have some additional explanation, comments or confirmation regarding the issue below? Thanks in advance..

Transformer: Unindo by Altom (Poyry) T&D
Design Standard: ANSI C57
Power Capacity: 1000kVA, 3 phase, 60Hz
Nominal Voltage: 13.8/0.415kV
Temperature Rise Data:
Winding - 65 deg C
Oil - 60 deg C
Type of Oil - Mineral
Transformer Protection - DGPT
Year of Manufacture: 2000

Actual Data (though limited to 2 days data only)
Peak Load - 548kW
Temperature reading - 70 deg C

The issue is the interpretation of the temperature gauge reading taken, which is 70 deg C with approximately 68-70% loading.

From what I understand, the temperature rise (winding or oil) should not be directly compared to the temperature gauge reading. The temperature gauge reading already includes the ambient or in this case, the room temperature. So if the room temperature is say 28 deg C, the reading of internal transformer temperature is only 42 deg C (70 minus 28).

Considering the above, it is therefore safe to say that there should be no cause for concern reference to the actual temperature reading right?

Notwithstanding, when should the panic alarm be sounded for temperature readings?

Lastly, if in case there is a problem with the temperature, is it technically correct to just provide a radiator fan close to the fins to assist in the cooling process? Does anybody have an idea on how many fans and what size should be considered? Can this be connected to the contacts of the DGPT so that we can set when to start and stop the fans?

 

[electricpete]

I think you are right that you have not yet exceeded your nameplate rises.

So if the room temperature is say 28 deg C, the reading of internal transformer temperature is only 42 deg C (70 minus 28).

I wouldn't say it that way. The oil temperature (or winding temperature... not sure which one you gave us) is 70C. The rise is 70 - 28=42. That is less than nameplate rise of 60 or 65C.

Whether or not you expect to exceed your nameplate rise at full load... that's another question.

When to sound the panic alarm...I would not sound it at the nameplate rise. For one thing if you're not at 40C ambient, you still have some margin. For another thing, thermal aging is a slow process.

Supplementary fans are a viable solution to reduce transformer temperature. There are many companies that sell fans along with mounting and controls specifically for this purpose.

=====================================
Eng-tips forums: The best place on the web for engineering discussions.

 

[arvinolga]

Thank you very much electricpete for the comments/inputs

 

[wareagle]

Transformer temperature rise is defined as the average temperature rise of the windings above the ambient (surrounding) temperature, when the transformer is loaded at its nameplate rating.

Check here

http://www.copper.org/applications/electrical/energy/trans_efficiency.html

 

[arvinolga]

Thank you also for the inputs wareagle... have looked in it and the data provided was very helpful.

 

[dewone]

In most cases the alarm contacts are set @ 75C and shut off at 85C

 

[elektroeng]

When a transformer is specified it is calculated using an average ambient temperature (depending on the region in which it is installed) where I live we use 35°C. Taking a temp rise of 60°C oil and 65°C winding this would make the trip setting 95°C for the oil temperature and an alarm setting at 85°C.

You should check in your data sheets or with your manufacturer for wich max ambient temperature the transformer was designed.

Regards
elektroeng

 

[arvinolga]

ok thanks guys, I have actually e-mailed the manufacturer already and currently awaiting confirmation.

relevant to this, is there any standard computation as to the design of a transformer room ventilation, say for both normal ventilation and forsed air? I'm particularly interested in the computation of air flow for a forced air ventilation