ncsl03

logo

 

Home

Program

Search Proceedings

Author Index

About NCSLI

Contact NCSLI

CD Tech Support

 

 

 

 

Session:

10B - AC Metrology 

Date & Time:

Thursday August 10, 10:45 AM - 12:15 PM

Paper Title:

Verifying the Wideband Input of an AC Measurement Standard 

 

 

Speaker:

David Deaver, Metrology Manager  

CoAuthors:

Neil Faulkner & David Deaver 

Speaker Info

Company:

Fluke Corporation

Department:

Primary Standards Lab

Address:

M/S 169G , POB 9090
Everett, WA, 98290, United States

Phone:

425-446-6434

Fax:

425-446-5649

Email:

David.Deaver@Fluke.com 

 

 

Abstract:

This paper describes the test method used by the manufacturer of a high accuracy AC Measurement Standard to verify the flatness and gain error of its wideband input. This input has a 50? input impedance and measures voltage from 700 µV to 7 V, 10 Hz to 30 MHz on eight voltage ranges. The Wideband input has specifications for flatness over the entire frequency range and for absolute voltage up to 500 kHz. Verifying these specifications is done in two steps. The first step is to measure the absolute error at mid scale on each range at 1 kHz. This is the gain error measurement. Next the flatness error is measured, again at mid scale. This is the change in reading at any given frequency from what the unit read at 1 kHz using the same input voltage. These two errors are combined to get the absolute error. The Wideband input is intended to be used in a 50? system so it is calibrated to read correctly when using a 50? source. This is done even though the input impedance of the Wideband input is not exactly 50? and changes significantly with frequency. The calibration process corrects for most of the errors this shift causes. Verifying this calibration is done with a characterized 50? source. But the output impedance of this source is not 50? and changes substantially with frequency. It is characterized in such a way as to correct for the errors this shift causes. The source is in turn characterized at 1 kHz on each range using a nearly perfect 50? load and an AC Voltmeter. For flatness it is characterized at 3.2 V using a characterized 50? Thermal Voltage Converter (TVC). The input impedance of the TVC is not exactly 50? and changes with frequency but again the characterization corrects for this shift. The characterized source is used to verify the Wideband input at 3.2 V on the 7V range. To verify the next range down, a 10 dB attenuator is added to reduce the voltage to about 1V. More attenuators are used for each range until a total of 70 dB of attenuation is used to verify 1 mV on the 2.2 mV range. Each attenuator used is characterized for its change in loss and shift in impedance with frequency. This paper starts by describing how sources and detectors whose impedances are not 50? and change significantly with frequency can be calibrated to source or measure voltage correctly in a 50? system. Then it describes the process used to characterize the attenuators and the 50? TVC used to calibrate the source. Finally it describes how the characterized source is used to verify the Wideband input of the AC Voltage Measurement Standard.  

 

 

 

 

NCSL International © 2006 All Rights Reserved
2995 Wilderness Place, Suite 107, Boulder, Colorado 80301-5404
Tel: (303) 440-3339 Fax: (303) 440-3384
info@ncsli.org

 

CD-ROM Produced by X-CD Technologies