Electricity: measuring and testing – Measuring – testing – or sensing electricity – per se – Frequency of cyclic current or voltage
Reexamination Certificate
1998-06-10
2001-06-05
Metjahic, Safet (Department: 2858)
Electricity: measuring and testing
Measuring, testing, or sensing electricity, per se
Frequency of cyclic current or voltage
C324S076790, C324S520000
Reexamination Certificate
active
06242900
ABSTRACT:
FIELD OF INVENTION
The invention relates to a system using digital peak detection for measuring partial discharge.
BACKGROUND OF THE INVENTION
Partial discharge measurement relies on detecting the small perturbations on the voltage applied to a sample under test. The voltage perturbations are caused by charge released by small breakdowns in the insulating material of the sample. The magnitude and pattern of these breakdowns provide an important tool for the evaluation, quality control and fault diagnostics of insulation systems. In a partial discharge measurement system, a sample under test is connected between a power source and a coupling impedance to isolate voltage pulses caused by the charge injected into the high voltage signal that is energizing the sample. The magnitude of the high voltage signal is many orders of magnitude higher than the voltage pulses caused by partial discharge. The height of the pulses is proportional to the charge released in the discharge events that gave rise to the pulses. It is this proportional relationship that forms the basis of partial discharge measurement. Since the magnitude of the pulses, in many cases, is in the milli-volt region, amplification is used before the pulses can be measured.
A band-limited amplifier is typically used to increase the magnitude of the pulses to the level at which they can be measured. The use of a band-limited amplifier assists in further eliminating the energizing frequencies that could affect pulse measurement. The amplifier also assists in limiting the effect of Radio Frequency Interference (RFI) that may be acquired by the sample, which can also affect pulse measurement. The use of switched filters and attentuators in the amplifier allow its characteristics to be optimized for the individual tests being performed.
Once the pulses have been amplified to a suitable level, their height is measured to get the discharge magnitude. Traditionally, this measurement has been performed using an analog peak detection system. With reference to
FIG. 1
, an analog peak detector
2
uses a comparator
4
to charge a capacitor
6
through a diode
8
. As long as the voltage on the capacitor
6
is less than the voltage of the pulse, the capacitor
6
is charged by the comparator
4
. Once the voltage on the capacitor
6
exceeds that of the pulse, charging is stopped. The output voltage of the analog peak detector
2
at this point is equal to the highest voltage that has occurred on the input
3
. When a data acquisition system is used to detect and measure individual pulses, the detector
2
must be reset by discharging the capacitor
6
after a pulse has been measured in order to be ready for the next pulse.
It is difficult to configure an analog peak detector of this type to be accurate, particularly when operating at the frequencies used for partial discharge measurement. The analog peak detector relies on a non-linear feedback loop, which is dependent very heavily on the characteristics of the peak detection system. The delay around the peak detection system, particularly through the comparator
4
and an output buffer, causes the voltage of the output of the analog peak detector to lag fractionally with respect to the input, resulting in an overshoot on the output. The magnitude of this overshoot tends to be non-linear with voltage, thereby limiting the accuracy of the system.
It is also necessary with an analog peak detector to compromise on the capacitor
6
used for peak detection. The voltage on the capacitor
6
tends to “droop” once the comparator
4
has stopped charging the capacitor because of leakage currents in the system. This introduces an uncertainty in the measurement, since the voltage decreases by some amount before the magnitude is measured. This effect can be limited by using a larger capacitance. A larger capacitance, however, requires more current for charging, resulting in a lower rate of change of voltage and limiting the maximum frequency that can be used for the amplifier. Thus, loop delays in the peak detection system are increased. A further complication is introduced when a reset is required on the analog peak detector. Charge injection from a reset switch can result in offsets on the output of the analog peak detector
2
, which further limits the accuracy of the partial discharge measurement system.
It is possible to control the characteristics of the amplifier and the coupling impedance of a partial discharge measurement system very well, which makes the analog peak detector the limiting factor on the performance of the partial discharge measurement system. The limit on the accuracy of the measurements that can be made is the non-linearity of the analog peak detector, which increases at the extremes of the measurement range due to the factors that have been discussed above. If it is possible to eliminate the analog peak detector, a significant improvement in the accuracy of the partial discharge measurement system can be made.
SUMMARY OF THE INVENTION
In accordance with an aspect of the present invention, a partial discharge measurement system is provided which comprises a computer and a digital peak detection circuit configured to perform digital peak measurement on signals.
In accordance with another aspect of the present invention, the partial discharge measurement system detects both positive and negative slope on voltage signals and controls the shape of the pulse capture window in accordance with different modes of operation. Modes of operation can include a general purpose measurement and pulse display mode, a pulse capture and analysis mode and a time-dependent pulse capture mode. In the general measurement mode, pulses are is captured and stored in a capture memory device in accordance with their respective positions in a voltage signal cycle. The general measurement mode provides a realistic display but does not guarantee resolution of all pulses (e.g., resolution of more than one pulse occurring in the same capture window). The pulse capture and analysis mode takes pulse polarity into consideration and re-triggers the peak detection circuit at zero-crossings for advantageous statistical processing of pulse information. The time-dependent capture mode stores successive samples for fault location in cables and does not measure pulse peaks or discharge magnitude.
In accordance with yet another aspect of the present invention, a partial discharge measurement system comprises two peak detector circuits for positive and negative polarity signals, respectively. The output signals of the peak detector circuits are compared and the largest absolute peak magnitude of the two output signals is supplied to the computer.
In accordance with still yet another aspect of the present invention, a peak detection circuit in the partial discharge measurement system comprises a two-stage pipeline of three registers and two comparators for comparing successive pulse signal samples and storing the larger of two samples as a peak value. A state machine is provided for selectively gating the registers and the comparators and other components (e.g., latches) in accordance with operation mode, the desired shape of the pulse capture window and pulse signal slope.
In accordance with another aspect of the present invention, the digital peak detection circuit in the partial discharge measurement system comprises an amplifier and a digitizer for processing signals received from a sample being tested. The sampling rate is selected in accordance with a maximum acceptable error associated with amplifier characteristics and the desired resolution of the partial discharge measurement system.
REFERENCES:
patent: 3991364 (1976-11-01), Wiznerowicz
patent: 4574234 (1986-03-01), Inbar
patent: 4651105 (1987-03-01), Inbar
patent: 4887041 (1989-12-01), Mashikian et al.
patent: 5247258 (1993-09-01), Tripier et al.
patent: 5315527 (1994-05-01), Beckwith
patent: 5416418 (1995-05-01), Maureira et al.
patent: 5532944 (1996-07-01), Battista
patent: 5534675 (1996-07-01), Kaneko et al.
patent: 5544064 (1996-08-01), Beck
Fawcett Timothy J.
Fore Neil S.
Hollington Jermele M.
Hubble Incorporated
Longanecker Stacey J.
Metjahic Safet
Presson Jerry M.
LandOfFree
System for measuring partial discharge using digital peak... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with System for measuring partial discharge using digital peak..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and System for measuring partial discharge using digital peak... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2508045