Discussion of "Measurement of eddy-current loss coefficient P/sub EC-R/, derating of single phase transformers, and comparison with K-factor approach" [Closure to discussion]

Abstract

1331per unit length and D the diameter of the conductor and finally f the frequency of vibration.Using this formula (with a friction coefficient of = 0:2), a gross-slip amplitude of usip = 2:7 mm has been calculated for f = 34 Hz, which seems to coincide well with a pronounced change in the slope of the curve shown in Fig. 7 of the paper. 1 Can the authors report, whether they have detected a similar effect for the other frequencies and/or conductors where they have carried out self-damping measurements? G. Diana et al.We thank Chuck B. Rawlins and Konstantin O. Papailiou for their kind discussions.Considering Dr. Rawlins interesting remarks, we consider separately the three points in which the discussion can divided. The first point is about the need for measuring two points close to the nodes or elsewhere: in our experience if the cable self damping alone is considered, we can say that the different '(x)=u functions (Figs. 2 and 3 of the paper 1 ) depending on the non dimensional damping h, are well separated only close to the nodes (in the considered case a few centimeters). Far from the nodes the curves tend to collapse independent from the cable self damping and they cannot give any reliable information to identify the cable self damping. We agree that when a more important power flows happen, as when dampers are on the cable, the different curves keep rather well separated also not so close to the nodes.Concerning the second point, we did not check the mentioned variation of the testing procedure based on the vector product of the conductor slope and the transverse vibration velocity, coming out from the closely spaced vibration sensors: our aim was just to evaluate the node amplitude of vibration, which, in our opinion, gives raise to the most robust identification procedure.The third point is about a different representation of the self damping, as the fraction " of the maximum bending energy stored in the cable that is lost to self damping each cycle. We made use of h because it is a universal well known parameter, giving the energy losses in mechanical systems, such as cables.Manuscript received 1999.

Harmonic Models of Transformers

“…Three-phase transformers have similar derating properties as single-phase transformers [7,12,73,75,76,77]. The maximum error of the directly measured losses is about 15%, which compares favorably with the maximum error of more than 60% [7,73] for loss measurement based on the difference between input and output powers as applied to high-efficiency (7/> 98%) transformers.…”

“…It extends the measurement approach of single-phase transformers [7,73,12,75,76,77] to three-phase transformers [75,76,78]. The apparent power (kVA) derating (Eq.…”