Why measuring only 3 gases is enough and saves you money

The ECOSENSE® ACTIVE PART measures the concentration of three components: hydrogen, carbon monoxide and water. Other, more expensive devices offer measurements of five or more dissolved gasses. So which one should you buy? More gasses is better, right?

In this article the idea behind a cheaper, streamlined sensor like ECOSENSE® ACTIVE PART is explained. Hopefully, after reading it, you can choose for yourself if the premium of a multi-gas sensor is worth it to you.

Rationale

The main benefit of DGA sensors is in detecting and identifying transformer faults. Let's look at this in a bit more detail. The order of events often looks like this:

an undetected fault occurs
gasses are accumulating
the fault is detected by a DGA monitoring system
an oil sample is sent to a lab to confirm the result
the results are interpreted and a fault type is identified
measures are taken

It should be noted that this process is not dependent on the number of measured gasses. In most utilities and industry, a lab check is done anyway, even if a fancy sensor is installed. Therefore, the additional measured gasses of a more complex sensor aren't necessary for fault categorization. The lab results can be used instead.

In the end, the advantage of multi-gas DGA sensors boils down to a higher fraction of detected faults. Not all transformer faults can be detected using DGA. For instance, bushing faults usually have no interface to the transformer oil.

Risk Reduction

So the main advantage of multi-gas DGA sensors is a higher probability of detecting failures. This comes at a price, of course. Multi-gas DGA sensors are more expensive. Simpler sensors offer the bulk of the benefit at a smaller price. This is illustrated here.

Gas Choice

Since not every fault can be seen with every gas, it is especially important for less complex DGA sensors to pick meaningful ones. Why did we pick hydrogen, carbon monoxide and moisture?

Hydrogen

For hydrogen the answer is simple. It is the gas that is produced over the broadest range of fault conditions. This can be seen for instance in the following illustration, taken from IEEE Standard C57.104:

Carbon Monoxide

The main contributor to carbon monoxide production is aging of the paper insulation. This is directly coupled to the remaining lifetime of a transformer. Replacement of the paper insulation is too expensive in most cases, so if the aging has advanced too far, the transformer has to be scrapped.

Situations that lead to accelerated aging are therefore critical - not from a safety standpoint but from a financial standpoint.

Moisture

Transformer oil is a good electrical insulator, thats one of the main reasons for its use. This property is dependent on the relative moisture content of the oil, however. Even at 40% of relative moisture, the breakdown voltage can be half of what it could be. At 20 degrees celcius, this corresponds to only around 25 ppm.

Insulation breakdown can quickly lead to significant problems. It is therefore critical to detect moisture problems early.

Should you buy a multi-gas DGA sensor instead?

Multi-gas DGA sensors have a higher probability of catching faults, but they are more expensive and have more parts that can fail. The higher probability can be translated into an expected benefit which is proportional to the consequence cost of a transformer failure. If the additional value of the benefit is higher than the premium of a more complicated DGA-sensor, then it is worthwile to buy one. If not, is makes more sense to buy a less complex sensor instead. When conducting this calculation, also keep in mind that the internal complexity of multi- or nine-gas DGA-sensors usually results in a reduced lifespan.

So there is a tendency that multi-gas DGA sensors make more sense in larger transformers and/or for more critical applications.