High Ethane Concentration (256 ppm) In Starting Transformer

[powering2002]

Dear All,

I work At Steam Power Plant 2 x 100 MW. It has Two Generator Step Up Transformer , Two Auxiliary Transformer and One Starting Transformer. The Starting Transformer is used mainly to supply auxiliary power to Main Unit during Shutdown, desalination plant, chlorination plant, Water Treatment Plant, and Office Building. It was rated 13 MVA. This starting transformer is never OFF. It operate continuously because when both main unit is Online,the starting transformer is still supply power to the desalination plant, chlorination plant, Water Treatment Plant, and Office Building. This power plant is built in 1980. So the starting transformer is about 34 years old now.

We performed Predictive maintenance including DGA analysis for all of our transformer. Based on our DGA results for Starting Transformer in the last 6 months, it is found that the Ethane (C2H6) concentration is very high. It exceed 200 ppm. Since January 2014, we have conducted 4 DGA Test. And from that tests, the C2H6 concentration is 247, 241, 253, and 256 ppm. . But this C2H6 concentration increasing trend is not followed by the increasing of the other Dissolved Gas Concentration (Co2, CH4, etc).
Here is our Latest DGA Result (21st of July 2014) for our Starting Transformer :
H2 : 60 ppm
Water : 22 ppm
CO2: 1581 ppm
CO : 108 ppm
CH4 : 97 ppm
C2H2 : 0.5 ppm
C2H4 : 9 ppm
C2H6 : 256 ppm
TDCG : 531 ppm

Oil Visual Inspection Color : BROWN

Using Roger Ratio to analyze that gas composition, it was concluded that It is NORMAL
But, we are worried about the high concentration of C2H6
My questions are:
1. Based on our latest DGA test, is Our Starting Transformer NORMAL or ABNORMAL?
2. If it is ABNORMAL, could you help us give the cause of high Concentration of Ethane Concentration?
3. Could you give use solutions or recommendation to overcome the high ethane concentration.

Thanks for your attention

 

[prc]

This is a DGA pattern very rarily reported. The pattern of gases shows that it is low temperature heating (say due to leakage flux from winding or heavy current leads) without involving paper or press board.Did you notice this recently or the ethane was prominent always from the start?

Nothing to worry. The only damage may be quicker oil deterioration. What are the values of N2&O2 in DGA? Test oil for all parameters to check deterioration level.

 

[powering2002]

Dear prc,

The high Ethane Concentration occured since we conducted DGA Test For The first time (2005)
According to Our DGA Test Result (2013),
N2 Concentration : 78908 ppm
O2 Concentration : 14062 ppm

Then my next questions:
1. Could you explain/mention all paramaters which has to be checked during oil test you have suggested?
2. In your opinion, Can a hotspot on the winding or tap changer cause a high ethane concentration?
3. In your opinion, Can a poor transformer grounding system cause a high ethane concentration?

Thanks for your attention

 

[prc]

Does it mean the pattern was same since 2013 or the value of ethane steadily going up? Another observation O2 being eaten away in this local heating .Normally the ratio of N2 to O2 should be 2:1 assuming in air ratio is 4:1 and solubility of O2 in oil is double of N2.

Are you from IEC or ANSI world ? Based on that look at IEC 60422-2013 or IEEE C57.106-2006 Usually I will suggest acidity, DDF, IFT and sludge as minimum.
Q2 - Ruled out
Q3- There is a remote possibility. Check current from core earth to ground in service.Check core to earth, core to core clamp and core clampto earth insulation with 500 V megger.

 

[waross]

Has the ambient temperature or the transformer loading risen (anything that will raise the oil temperature) over the last 6 months. The tests may be influenced by the oil temperature at the time of the sample taking.

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

 

[FPelec]

Poor insulation between core, clamps and tank could cause moderate heating, as suggested by prc. Another possible cause of moderate overheating could be the flux shields: those are generally insulated from the tank and grounded (or connected to the tank) in only one point; if the insulation is lost stray currents could cause overheating.
Another hypotesis could be presence of unbalanced loads (or 0-sequence harmonics) causing stray flux and tank/magnetic shields overheating.
the last, less probable, hypotesis is that the core clamps desing form a loop; induced current in the loop could cause gas generation, especially upon transformer energization this happened once in one relatively old ATR, leading to a similar gas pattern.


Si duri puer ingeni videtur,
preconem facias vel architectum.

 

[powering2002]

Dear prc,
The value of ethane is slightly going up since 2013. We usually use ANSI/IEEE Standard. I almost forget. We have performed Liquid Screen Test (Oil Test) before. here is our latest result of the test:
1. Acidity : 0.014 mgKOH/g
2. IFT : 35.8 mN/m
3. Dielectric : 88 kV

Thanks prc. Any further suggestions?

 

[prc]

Oil values are quite OK. I think you can leave as such but repeat DGA every year, keeping an eye on the annual increase of thermal gases.

 

[squeeky]

Dear Powering2002, were your transformers correctly specified on purchase. If you look at IEC60067-1 of 2008, clause 8.3 requires that transformers connected directly to generators should have thier voltage ratings at 1.4 times rated for 5 seconds due to rejection problems. The switching may be causing small break downs in the insualtion or creepage and clearence distances of the insulated parts within the transformer.

I have no experience of this or what the remedy may be. I have been working with transformer specifications and discovered this three line clause. Surge suppressors? can they be graded so close to the nominal voltage?