A new solution of high precision wideband time-division wattmeter

Abstract

A new solution of the time &vision multiplier of the Goldberg type applied in the high precision wattmeter is presented. The solution is based on the method of compensating the systematic frequency error leading thus to the hgh performances of the multiplier even in the case when the conversion frequency is only 10 times the frequency of the signals to be measured.

A method of compensation of the frequency dependent systematic error of the time-division watt-converter using two multipliers

“…This new method is based on two identical multipliers which works on two different frequencies. All advantages mentioned above about method [7] are also valid for the method based on two multipliers which is presented in this paper. There are some other additional advantages of this method comparing to the method [7] related to the practical realization which will be described in the paper.…”

“…It has been shown that for practical realization of the method there is no need to duplicate all functional of the two TDMs, but some simplifications are possible. Unlike the compensation method [7] this method enables possibility to couple secondary coil of the current transformer directly to analog multiplexer of the pulse amplitude modulating circuit as in [2]. A watt converter based on this method has been developed and results of measurements show that at the unity power factor precision of +/-50x10".…”

“…In this manner the frequency dependent systematic error inherent in TDM has been reduced but still remains finite. Paper [7] presents for the first time a method which compensates for the systematic error of TDM of the Goldberg type in a relatively wide frequency range by using corrective frequency dependent circuit on both inputs of the multiplier, enabling easier functioning of the multiplier at lower conversion frequencies at which the imperfections of the components have not a significant influence. Using this method of compensation there are no requirements for the conversion frequency to be two or three orders of magnitude higher than input signal frequency as in solution [5].…”