UOA on a Power Transformer - Oil Leaks

[Shannow]

Worked in transmission for a year after Uni, and attended a lecture on transformer UOA.

Imagine a big steel box, and sitting inside it laminated iron cores, wrapped with copper conductors, some as thick as your thumb, and insulated with what's essentially craft paper...then vacuum filled with insulating oil, highly refined mineral oil, with some Anti Oxidant, and alkalinity additives.

UOA is regular (monthly), and tests for
* electrical resistivity (oil contamination and breakdown)
* reserve alkalinity
* moisture
* oxidation resistance
* furans (indicate that the cellulose is breaking down due to age/oxidation, or internal sparking)
* traces gasses (ratio of N2 to O2, CO, CO2, CH4 tell a lot about what's happening inside).

Anyway, the lecturer stated that he could tell a lot about the future working life of the Transformer oil, and the transformer itself before he took the sample, by visual inspection for oil leaks.

Pretty common in any switchyard, I've included the pic below, of a typical, and common oil stain from a gasket (and a bloke patching the leak).

Question is

Why did the transformer UOA expert state so emphatically that he could "tell a lot" about the future working life of a power transformer by the presence of gasket weeps ?

 

[440Magnum]

Maybe he gets a good idea that the oil is getting far into its usable life simply because new transformers aren't likely to leak, old ones are.

Therefore, if you see a leak its probably old-ish oil.

I actually don't know, but that would be my first guess.

My second guess- once a leak opens, it goes both ways. Oil comes out, but contaminants (oxygen, moisture, etc.) can get in.

 

[Shannow]

Originally Posted By: 440Magnum
My second guess- once a leak opens, it goes both ways. Oil comes out, but contaminants (oxygen, moisture, etc.) can get in.


Spot on...air and moisture permeates back into the transformer along the leak path.

Some transformers have an elevated, nitrogen blanketted head tank to prevent air ingress, but a weeping gasket below oil level will still let the oxygen in.

 

[turtlevette]

Originally Posted By: Shannow
Worked in transmission for a year after Uni, and attended a lecture on transformer UOA.



You worked in transmission too? What haven't you done? It's not really a UOA it's called a dissolved gas analysis and tells you if there is any internal arcing such as a turn to turn fault.

They make real time analyzers that connect to the tank that provides data to SCADA systems.

There should not be any moisture ingress as the tank should always have a positive pressure of nitrogen or dry air. Other systems have a conservator tank which keeps all gasses out.

When a transformer experiences a low grade fault, the arcing creates combustible gasses in the oil which causes a small pressure gradient over time. A sudden pressure relay will detect the pressure gradient and trip the transformer off line. Any gas creation from arcing below this threshold will need to be detected by dissolved gas analysis.

 

[Shannow]

Originally Posted By: turtlevette
A sudden pressure relay will detect the pressure gradient and trip the transformer off line.


Bucholtz relay does a little more than just that.

Originally Posted By: turtlevette
There should not be any moisture ingress as the tank should always have a positive pressure of nitrogen or dry air. Other systems have a conservator tank which keeps all gasses out.


The point of the post was that a leaking gasket will allow air and moisture in...same as oxygen leaks into a nitrogen filled tyre, due to the partial pressure difference between the outside and the inside...even if there's a conservator, a leak means oxygen ingress.

Originally Posted By: turtlevette
You worked in transmission too? What haven't you done?


Have covered more than a few bases in my career...never programmed a PLC.

 

[userfriendly]

Does the positive pressure of one gas, in this case nitrogen
prevent other gasses from entering where a leak has formed?

I would say "watch out for a rake in the grass".

 

[edhackett]

Originally Posted By: used_0il
Does the positive pressure of one gas, in this case nitrogen
prevent other gasses from entering where a leak has formed?


No. Diffusion takes place at the molecular level and a pressure gradient has no effect. It is driven by the partial pressures only. If you're having trouble with the concept of partial pressures, think of them as consentrations. Gases will seek to equalize their concentrations if they can. If there is a permeable barrier or a leak between two concentrations of a gas, the gas will move to equalize the concentration regardless of any pressure differential.

Diffusion against a 50,000 psi gradient has been demonstrated.

The gas delivery guy being "helpful" and cracking the valve on your bottle of 'seven nines' helium to blow the dust out of the threads results in measurable contamination of the bottle of helium.

Ed

 

[turtlevette]

Originally Posted By: Shannow
Have covered more than a few bases in my career...never programmed a PLC.


So you can't get along with anyone either. LOL

 

[userfriendly]

How much gas exchange occur in your lungs when you hold your breath?

 

[kschachn]

About the same as when you don't, at least at first.

Originally Posted By: used_0il
How much gas exchange occur in your lungs when you hold your breath?

 

[SHOZ]

Originally Posted By: Shannow


Originally Posted By: turtlevette
You worked in transmission too? What haven't you done?


Have covered more than a few bases in my career...never programmed a PLC.


Once you start programming PLCs then nobody trust you. For you work the magic.

 

[userfriendly]

My dad smoked in the car all the time when I was young.
(That is not all he did in the car with the windows rolled up).

About 75% of railway workers smoked when I started in 1974.
(Now about 75% don't smoke)

If the molecular rules of gas transfer hold true, I might as well
have been considered a smoker for the most part of my life.

 

[ac_tc]

he probably looks at the general standard and appereance of the site and the transformer..
- a good looking site with nice and Clean transformer are usually well maintained and the chans of finding faults is smaller.

 

[SHOZ]

Originally Posted By: used_0il
My dad smoked in the car all the time when I was young.
(That is not all he did in the car with the windows rolled up).

About 75% of railway workers smoked when I started in 1974.
(Now about 75% don't smoke)

If the molecular rules of gas transfer hold true, I might as well
have been considered a smoker for the most part of my life.


~500,000 people die each year in the US from smoking related illness.

 

[userfriendly]

That would make for an interesting bell-curve on the
baby boomer generation. 1944-1964, 51-71 years old
in 2015.