KVD comparison endeavor

[VintageNut]

I recently added a Fluke 721A Lead Compensator to my volt nut rack.

The rack has two KVD. A DV-4107 and a DV-4107D. It is now set up to compare one KVD against the other.
The working hypothesis is that the newer model should be the reference divider.

I have adjusted the 721A so that there is a zero reading on the null meter for LO and HI lead compensation.

The voltage applied to the test is 10V from a Fluke 731B.

Below are the results for the most significant decade of the KVDs where each comparison is a balanced null.

Question: Is there an algorithm for checking a certain number of settings to be confident of the maximum deviation of the two KVDs?

Any comments or suggestions are welcomed.

DV4107   DV4107D 999999X   999999X 8999998   899999X 7999997   799999X 6999995   699999X 5999995   599999X 4999995   499999X 3999994   399999X 2999997   299999X 1999997   199999X 0999996   099999X

[JS]

check all the... 10^7 combinations??  ok, maybe not.

Maybe the 9 combinations 1111111 counts apart from each other? Half up/half down looking for the max "INL" or deviation? Get a girl? Maybe she helps you?

JS

[mycroft]

Try one of these references:

  - Calibration of a Kelvin-Varley Voltage Divider http://ieeexplore.ieee.org/document/4313387/
  - Calibration of a Kelvin-Varley Standard Divider http://ieeexplore.ieee.org/document/5006924/

[VintageNut]

Try one of these references:

  - Calibration of a Kelvin-Varley Voltage Divider http://ieeexplore.ieee.org/document/4313387/
  - Calibration of a Kelvin-Varley Standard Divider http://ieeexplore.ieee.org/document/5006924/

Thanks for the references. It looks like you have to pay for these articles.
Anybody have a copy?

[VintageNut]

Try one of these references:

  - Calibration of a Kelvin-Varley Voltage Divider http://ieeexplore.ieee.org/document/4313387/
  - Calibration of a Kelvin-Varley Standard Divider http://ieeexplore.ieee.org/document/5006924/

pm sent
Thanks for the references. It looks like you have to pay for these articles.
Anybody have a copy?

PM me with your email address...