ARS Altmann

Sitemap & Help
You are here: Technical Support -> Frequently Asked Questions

Frequently Asked Questions (FAQ)

5. DGA distortion due to vacuum treatment of the transformer oil

A potential distortion of a DGA reading due to the on-line (long-term) vacuum treatment of (not-hermetized) transformers is undoubtedly the most frequent objection against this treatment.

These objections look very reasonable at the first sight :

But already a slight examination of present DGA methods immediately shows at least two very serious and fundamental discrepancies related to this measuring method.

The fundamental question is – can be the DGA at all regarded as something sacred, a sangrail, which can be irreversibly destroyed e.g. by the radicall degasing.

The present one-shot DGA analysis is a-priori based on the following presumptions:

  1. 1. the fault in the transformer produces a typical gas, the gas dilutes in the oil of the main tank, the DGA exactly determines the gas in the sampled oil
  2. 2. the absolute level of the given gas in the oil corresponds to the extent of the fault
  3. 3. the variation of the measured level of the fault gas corresponds to the variation of the extent of the fault

Only the first presumption is correct. The second and the third presumption is fundamentally wrong, because the actual content of the fault gas (or its variation) in the oil inventory does not depend only on the extent (or a variation) of the fault .

On the contrary, a fault gas content in the oil in the main tank always depends on two factors:

The weakest point of the present DGA is, that this method directly connects the reading of the absolute gas content level with the gas input (from the fault) only.

From the poinf of fluid dynamic this approach is definitely wrong. The main tank of a transformer can be regarded only as an open system where inflows- outflows relations determine the quasi-stable gas level in the oil inventory. Any change in the inflow (fault) or outflow (escape in surroundings) of given gas will therefore inevitably change its DGA reading.

For real DGA supporters this simple conclusion is very hard, because of:

the present DGA does not represent an exact measuring method and a precise diagnostic tool at all.

The absolute level of the measured gas in the oil filling of the transformer depends not only on the extent of its internal fault but strongly depends also on boundary conditions which controll entry / discharge relations between the transformer and its surroundings – for the experimental verification of this general statement See \ News\ Oxidation Ageing Alarm.

The recapitulation:

For another and a more problem-oriented analysis See\ News\ Impact of Vaccum Treatment on DGA.