Tan Delta Correction Factor

Poor scooby!!!....no answer yet!!!

Well. What kind of part, and what operating condition, we'd like to ask.

Tan delta, or D = 1/Q, is the tangent of the loss angle. Speaking for example of capacitors, there are two kinds of loss, parallel leakage loss, and series resistance. Usually one of these dominates in causing the loss. For example, with AC rectifier filter capacitors, it's generally the series resistance, or esr, that dominates the loss. But instead of tan-delta, this loss effect is usually expressed through the ripple-current rating of the capacitor (the designer having chosen a sufficiently high voltage rating to minimize any leakage-current losses).

Accordingly, the manufacturer's specs may be more concerned with a ripple-current rating, than with tan delta. Now, as it happens, esr and ripple-current loss go down with increased temperature. By contrast, leakage current loss goes up with temperature. It'd be nice to have curves for both of these effects, vs temperature. But some manufacturers may instead give us reliability vs temp graphs.

As far as how a parameter changes with temperature, so you can make a spec correction, that's another issue specific to the part. For example, a tantalum capacitor may have a logarithmic leakage scale vs temp plot, so you can't apply a linear formula.

What if your interest isn't capacitors? For example, it may be high-voltage cable insulation, which is often tested with tan-delta tests, looking for water trees, etc. As it happens, polar dielectrics have a funny D vs temp plot, going up to a peak and then back down again. Aha, but then usually one doesn't have the luxury of temperature cycling a half-mile of high-voltage cable.

Thanks Mr. Hill,

I am talking about motors, generators and transformers.

If we measure tan delta at 35oC and I would like to correct it to 27oC to get a reference value so that future readings taken too are corrected to 27oC so that we can get a comparative value. This is same as we do for insulation resistance and winding resistance.

I would like to know the formula for correcting tan delta value from measured temperature to reference temperature.

A paper on the subject says that IEEE method of temp. correction does not give a correct picture as it has been revealed by DFR test results that tan delta of winding will be dependent on moisture content of solid insulation, i.e. different temp correction factors for transformers with different moisture content of solid insulstion.