Dear all,

we need some inputs and advices.

how should we calculate IEC 60599 ratio method if the DGA results show some gas values below the analytical detection limit S?

for example (these data have been report by other third party):
H2 < 5 ppm
O2 = 25860 ppm
N2 = 53703 ppm
CH4 = 17 ppm
CO = 117 ppm
CO2 = 2465 ppm
C2H4 = 67 ppm
C2H6 = 5 ppm
C2H2 = 566 ppm

so, how to calculate ratio CH4/H2 or C2H2/H2 since the report only shows the H2 value as "< 5 ppm"?

thank you, we would like to hear from you guys,

anf

The ratio methods generally do not apply when you have gases in the undetectable range.

As you probably already know, the simple analysis for these results is arcing (based on very high C2H2 which is a strong indicator... cannot be formed at temperatures below where arcing occurs)

Also your oxygen seems abnormally high which to me suggests the transformer is not well sealed from the outside air... makes me wonder about the moisture content (was that tested?)

Thank you for the reply Electricpete,

I agree with you that C2H2 is strong evidence for arcing since this gas will be formed at least at temperature 800 degC – 1200 degC (IEC 60599- section 4.1), and this range of temperature will only achieve in arcing failure.

Water content for this transformer is 32 mg/kg.

Please correct me if I’am wrong, the ratio for O2/N2 for this transformer is about 0.48. So, as per IEC 60599- section 5.5 “O2/N2 ratio”, the abnormal oil heating/oxidation is not detected in this transformer since the ratio is above 0.3. I’m really sorry since I only depend on what I have been read in IEC 60599 as interpretation guide.
I will be appreciated if you would share your experience on this condition.

Hmm..i still confuse , if generally ratio method is not for the data with undetectable gase(s), why this condition was not mentioned in IEC 60599? (or perhaps I have missed this information in IEC 60599)

Ok, thanks a lot for your attention…

I am familiar with the US IEEE standards, but not the international IEC standards. Two comments:

1 – The ratio methods are not valid unless you have some minimum quantities of gas. Per IEEE C57-104 (1991), page 22 "Determining the validity of the ratio procedure: if at least one of the gases in each ratio R1, R2, R3, R4 exceeds limit L1, the ratio is valid, otherwise the ratios are not significant"

2 – Generally we expect a fair quantity of nitrogen based on the nitrogen blanket, but a smaller quantity of oxygen. Your oxygen is higher than I am used to seeing which could indicate air inleakage. You mentioned a threshhold of 0.3 for the ratio O2/N2 as if it were a minimum. It would make more sense to me that 0.3 should be a maximum rather than a minimum (there is nothing wrong with a low ratio). In this case, you are exceeding the max O2/N2.