DGA confirmation

[cshro]

if a DGA analyse indicate a fault (partial discharge, overheating, arcing, celulose) witch are the measurements I can do to have a confirmation of this fault)?

 

[electricpete]

This is a transformer question I assume.

DGA is usually a pretty good indication, particularly if you have an increasing trend. Maybe if you tell the gases and concentrations there might be some more response.

I can't think of many ways to confirm. On-line you can check temperature, moisture, loading, voltage, possibly infrared. Electrical testing with the transformer out of service will identify some more but not all problems. Sometimes internal visual inspection is performed.

 

[cshro]

I will be more clearly
- for celulose decomposition: I measure H2O content, acidity ......
can you tell me what must to measure for
arcing; overheating; PD.

 

[electricpete]

Umm.. for cellulose you might want to look at CO2 and CO.

Arcing key gas is acetlyene

PD I think would be hydrogen but hydrogen has many other sources.

Local overheating from hot metal - key gas is ethylene.

Overheating from running an FOA transformer under load without cooling - H2, CH4 (been there, done that!)

http://www.usbr.gov/power/data/fist/fist3~30/3~30_con.htm

http://www.nttworldwide.com/tech2102.htm

 

[cshro]

My enhlish is very poor
I know very good how fault gases apear.
I refer to other measurements

 

[electricpete]

I think you are asking what you can look at IN AN OIL sample besides the fault gases.

You are correct for cellulose decomposition you would look at H20, both as a byproduct of decomposition and as a cause of decomposition. Likewise you would look at 02 as a cause of cellulose decomposition.

For partial discharge look for H20 and dielectric strength and possibly metal-in-oil as a possible cause of the problem.

That's all I can think of. If you have not posted in the power engineering forum you can try posting there.

 

[cshro]

Thank you for your answer. I think that is very hard to 'talk' with a person who don't know English

for have a confirmation of a hot point ther is another metod then termovision
how can I have the confitmation of a arcing

 

[electricpete]

Thermalvision = infrared may not be able to detect hot spots deep within the transformer. Much better for hot spots near the tank wall.

I don't know of any other well-established means to detect internal arcing other than dga. I have heard you can tune an am radio to 100khz and use the antenna as a probe to identify arcing/discharge. But the results are not quantitative and there may be interference from overhead lines. Also there are some acoustic methods for monitoring transformers to detect internal arcing (Triple-5 Technologies), but that is not common.

 

[patr]

Arcing is an intermittent high-voltage discharge
that occurs without high current. Arcing is
characterized by increasing levels of hydrogen,
methane, and ethane without a concurrent increase in
acetylene in the results of DGA.

Other than de-energizing and disassembly, corona detection using ultrasonic instrument is only other solution to detect arcing while energized. Look to UE systems webpage for examples of this technology.

Thermovision will very rarely detect arcing type of faults internal to a transformer mainly due to inability to gain line-of-sight. Be wary of anyone who claims to be able to detect using thermovision

 

[jbartos]

Suggestion: Visit

http://www.state.la.us/orm/lpequip.pdf

(for DGA and others)

 

[electricpete]

patr - acetylene will increase.