Transformer Oil Level Check

[mpeck1982]

EE's,

I have a question about checking oil level in transformers. We recently checked the oil level in all of our transformers. We connected a fitting to the drain valve and then connected tygon tubing and held the tubing to the height of the transformer. We cracked the drain valve open and visually could see if the oil level was just above the radiator fins. The end of the tygon tubing was to atmosphere. That method worked on a few transformers. When performing this method on GE Prolec transformers we had to connect the end of the tygon tube to a port on a pressure gauge (negative/positive psig) located at the top of the transformer to see the correct oil level. When the end of the tubing was to atmosphere the oil level in the tygon tubing would be well below the radiator fins like 16-in. or more. Can someone tell me why we had to perform the oil level test differently?

 

[waross]

Checking the oil level with the end of the tubing open to the atmosphere is only accurate when the transformer tank is open to the atmosphere. Many transformers have a two way pressure relief valve to limit changes of internal air and the ingress of moisture. Many others are sealed and have some method of allowing expansion and contraction of the oil without venting.
It is common for the tank pressure in many transformers to be either above or below atmospheric pressure depending on the oil temperature and the ambient atmospheric pressure. With one end of the tubing open to the atmosphere you are measuring a combination of the oil level and the internal pressure. The internal pressure may change from positive to negative as the oil temperature changes.

Bill
--------------------
"Why not the best?"
Jimmy Carter

 

[dpc]

If you try that on a transformer with a conservator tank, you'll get a big surprise. Otherwise, what Bill said.

 

[unclebob]

Don't you have Oil Level gages on your transformers? Or are they become unreadable?

 

[analogkid2digitalman]

Why not use an ultrasonic method to detect the backwall of the transformer?
No liquid at your probing height yields no backwall pulse reflection.

-AK2DM

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
"It's the questions that drive us"
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

 

[roydm]

Also be aware that the oil in the tubing is at a different temperature, this will mean the tubing will indicate slightly less than the actual by 1/2" or so.
Record the ambient and tank temperature

 

[mpeck1982]

Transformer was definately sealed and not open to atmosphere. So what am I actually seeing when I have the tube to atmosphere on a sealed transformer??

 

[electricpete]

It's well covered above, but I'll try to summarize.

The pressure due to fluid at penetration elevation must be same inside and outside the tank.

(Rho*g*h)inside + Papplied_inside = (Rho*g*h)outside + Papplied_outside
Where rho is density (depends on temperature as mentioned), g is acceleration of gravity, h is height, Papplied is pressure of gas above the fluid.

You can see that density difference or difference in pressure applied will cause the height outside to differ from height inside.

Also if there is flow in the neighborhood of the tap there may be dynamic effects not captured in above equation.


=====================================
(2B)+(2B)' ?

 

[unclebob]

Don't you have Pressure gages on your transformers?

 

[wbd]

I would add that if the tubing was connected to the pressure gauge port, that the pressure gauge is mounted above the oil level in the gas space. Therefore, no oil should be in the tube.

The nameplate should have on it what the correct oil level is. Another way, if you don't trust the level gauges, is to use an infrared camera. There should be a difference of temperature between the oil and air space.