11th International Symposium on H.V. Engineering - ISH'99--London, UK

DR. STRAUSS

Abstract
The paper describes the hardware of a new generation of digital recorders, designed and optimized to fulfil all requirements for measuring of full and chopped lightning impulses as also switching and current impulses.as required for national metrological laboratories and accredited laboratories.
The results of the official calibration performed by the Physikalisch-Technische Bundesanstalt (PTB) in Braunschweig on the first TR-AS 200-12 digital recorder available now as Reference Standard of Measurement in our DKD Calibration Laboratory for electrical impulse parameter are shown.
Introduction
The requirements for individual errors and the overall uncertainty of reference standard and approved measurement systems for measurement during high voltage and high current impulse tests are specified in the Standards IEC 1083-1 and IEC 1083-2.
The user in Accredited Laboratories are obliged to use measuring equipment according to these standards. To document the compliance this equipment must be calibrated accordingly with traceability to national standards.
For this task the German Calibration Service (DKD) installs calibration laboratories which are accredited and supervised by the PTB.
The first calibration laboratory in Europe for electrical impulse parameter was founded in 1994 in Germany by DR. STRAUSS GMBH, the accreditation was performed by the Physikalisch-Technische Bundesanstalt (PTB) represented in the Deutschen AkkreditierungsRat (DAR) /6/
The National Metrological Laboratories are also obliged to use measuring equipment according to these standards or better and need precise and long term stable digital recorders, too.
For this task the new generation of precise digital recorders from 100 MS/s with 10 and 12 Bit resolution up to 200 MS/s and 12 Bit resolution were developed to represent the first TR-AS 200-12 digital recorder as a Reference Standard of Measurement.

Hardware Description

The prooved Impulse Voltage Measuring Systems TR-AS, including the digital recorders from our own development and manufacturing, were consequently further developed and show now improved sampling rates up to 200 MS/s in realtime with 12 Bit resolution simultaneously in up to 16 measuring channels.
A careful design using sophisticated programmable logic and SMT optimizes the contrary

requirements like high sampling rate, small risetime, minimized noise, short settling time, small measuring uncertainty and good long-term accuracy.
The high sampling rate of 200 MS/s and the high bandwidth of 50 MHz allow now measurement of steep impulse voltages with a short time-to-chopping or superimposed oscillations with increased accuracy.
The resolution increased to 12 Bit  results in nearly smooth shapes without visible quantization steps and improves the results of diagnosis using difference method and transferfunction.

ring data minimizes the access times. The time duration for evaluating and storing of measuring data are considerably decreased.
The measuring electronic is specially designed considering the application in the h.v. testfield with extremly high electromagnetic distortions. The individual measuring channels comprising of input divider, measuring amplifier, A/D-converter and recorder memory are constructed modular and independently from each other, so easy assembly of 1 up to 4 channels is possible including a later upgrade depending on the application.
The 1600 or 2000 V input divider as the critical component is designed for high performance and long term stability /16/ including an overvoltage protection tested with 7 kV peak voltage. The analogue bandwidth and risetime of the input divider as also of the direct input could be increased and amounts to 75 MHz and 5 ns respectively, so the analogue bandwidth and risetime overall for both corresponds to 50 MHz and 7 ns.
The direct input feeds a fully programmable measuring amplifier comprises 23 stages, which allows a very fine adaption in 80%-steps according to a factor of 1.26 to the amplitude values to be measured. In this respect, signal amplitudes can be measured with optimal deflection from 2000 V downto 12 V using the input divider while signal amplitudes from 10 V downto 65 mV can be measured using the direct input.
The EMV-oriented design of the A/D-converter and memory system determines beside the sampling rate and resolution the overall performance of the digital recorder with respect to ripple, noise and interference.
The digital impulse voltage measuring systems TR-AS can therefore mostly be used without an additional measuring cabine also for testing of vacuum and SF6-insulation in the high-voltage testfield /10/.
For the calculation of the impulse parameters of standard and non-standard lightning impulses, switching impulses, oscillating impulses and impulse currents an evaluating program is implemented that determinates the respective values, e.g. for a lightning impulse the peak value, the front time and the time-to-halve-value and for chopped impulses the time-to-chopping, too. The calculation is done automatically for each shot, whereby the results can be displayed on screen together with the impulse shape. Additionally a normalized amplitude grid with respect to the actual peak value can be displayed. Furthermore measuring lines for front time, time-to-halfe-value or time-to-chopping can be displayed as also a time grid normalized to virtual origin.

Picture 1: Desktop System TR-AS 200-12
The integration of all components like screen, keyboard, printer and the measuring system with directly coupled personal computer, designed in 19" rackmount technic and built into an EMC-tested measuring rack with common line filter, yields an excellent shielding. The requested tests of electromagnetic interferences of digital recorders according to IEC 1083 were fulfilled without any additional measures /5/7/10/17/.
WinTRAS, the sophisticated multitasking measuring and analysis software for Windows95®, consequently developed on basis of the multi-purpose DOS-software TRAS, shows extensive functions including OLE-automation. The integrated dBase®-database stores all measuring results of all tests performed including the shapes (raw data) and allows userspecific evaluations and analysis for documentation and quality control.
Additional to the measuring rack MIRA 25, proofed in test fields with high electromagnetic interferences /10/, for mobil use further housings are available as also a desk variant (Picture 1).
To solve the problems of electromagnetic interferences and interfacing, the AT-compatible control- and evaluating computer was built into the well-shielded housing of the measuring system and coupled electrically directly. The direct computer access to the programmable registers of the digital recorder and to the recorded measu

For non-standard shapes an automatic, selfadaptive meancurve calculation can be selected, additionally a screen-oriented hand evaluation feature is available /15/.
Actual recorded or from harddisk reloaded measuring data and mathematical combinations of them, e.g. a difference curve or the transfer function during transformer tests, simultaneously can be displayed in up to 32 virtual channels and allow a fast and improved fault diagnostic.

PTB-Calibration Results

The following table show calibration results with full lightning impulse LI 0.84/60 and chopped lightning impulse LIC with a time to chopping of Tc=0.5 ms for the digital recorder type TR-AS 200-12 calibrated by PTB in 1998.
The calibration of full and chopped waves from 1000 V downto 50 V peak can be performed with the Reference Calibration Generator System KAL 1000.

Technical Data of Digital Recorder TR-AS

The following table shows the technical data of the new generation of digital recorders for h.v. impulse tests:
The dynamic behavier of the A/D-converters available today show a decreasing resolution for increasing signal steepness respectively signal frequencies. Therefore the determination of the differential dynamic non-linearity with triangular signals with 2,5 MHz as required in IEC 1083-1 shows a reduction of resolution from 12 Bit for impulse shapes similar to peak and tail of LI down to 10 Bit in the front of LI and LIC. For steep voltages a resolution of 10 Bit show together with the long term stable analogue input excellent calibration results. The resolution of 12 Bit improofes  a.c. and d.c. accuracy and comparative measurements advantageously.

Address of the author:
Dr.-Ing. Werner Strauss
DR. STRAUSS SYSTEM-ELEKTRONIK GMBH
Bodenackerstraße 6
D-96163 Gundelsheim/Bamberg
Telefon  +49-951-43063
TeleFax +49-951-43858
INFO http://www.strauss.de
email: dr.strauss@strauss.de

1) referenced to 8 Bit  2) referenced to 10 Bit (according to IEC 1083-1)

References

/1/ IEC Publication 1083-1: Digital recorders for measurements in high voltage impulse tests, Part 1 - Requirements for digital recorders, 1991
/2/ IEC Publication 60: High Voltage Test Techniques, Part 2: Measuring Systems, 1994
/3/ W. Strauss: Automation of i.v. tests online with microcomputers and digital recorders, thesis TU Berlin 1983.
/4/ W. Strauss: Calibration of digital recorders for h.v. impulse tests, 7th ISH Dresden 1991, paper 62.03.
/5/ H. Bachmann, et al: Field and EMC-tests of a digital h.v. impulse control and measuring system, 7th ISH Dresden 1991, paper 62.04.
/6/ W. Strauss: Calibration of digital recorders and impulse calibrators with traceability to national standards, 9th ISH Graz 1995, paper 4495.
/7/ W. Hauschild, et al.: A 4 MV outdoor impulse test system for the Korean 765 kV transmission project, 9th ISH Graz 1995, paper 4487.
/8/ W. Hauschild, et al.: Computer-aided performance tests and checks for high-voltage measuring systems, ERA-Report 94-0776, paper 3.3.
/9/ IEC Publication 1083-2: Digital recorders for measurements in high voltage impulse tests, Part 2 - Evaluation of software used for the determination of the parameters of impulse waveforms, 1996.
/10/ Type Test Report: Digital i.v. measuring system TR-AS 100-10, information brochure of DR. STRAUSS GmbH, 1993.
/11/ Type Test Report: Impulse calibrator system KAL 1000, information brochure of DR. STRAUSS GmbH, 1994.
/12/ Type Test Report: Control and evaluating software for digital recorder TR-AS, information brochure of DR. STRAUSS GmbH, 1996.
/13/ A. Marinescu, V. Onu: Accreditation by Relar of high voltage and high power laboratories of Craiova, ERA-Report 96, MILAN, paper 2.3.
/14/ W. Hauschild, et al.: Calibration of hv measuring systems for the mutual recognition of hv test results in eastern and western Europe, ERA-Report 96, MILAN, paper 3.3.
/15/ W. Strauss: Selfadaptive evaluation software for high-voltage impulse tests, ERA-Report 96, MILAN, paper 5.4.
/16/ M. Aro, et al.: Intercomparison of h.v. impulse measuring systems with digital recorder, 10th ISH Montreal 1997.
/17/ W. Strauss: Hardware, software and calibration results of digital recorders for h.v. impulse tests, 10th ISH Montreal 1997.
/18/ W. Strauss, et al.: Hardware, Software und Kalibrierung von Digital-Recordern für Stoßspannung, HIGHVOLT-Kolloquium 1997, Dresden