[volt-nuts] Drifting 3458As

[Randy Evans]

I am running the ACAL to check cal constant drift over several times a day
and over a 7 day period (per SN18).  The varying times of the day are to
see how temperature sensitivity affects the cal constant drift since the
room temperature is not constant through the day but is relatively constant
at the same time of day each day; i.e., the furnace is off during the night
so mornings are about 5 degrees C cooler than the afternoon.  Even after
ACAL, I am seeing up to an absolute difference of 3.5 uV on the 732A
voltage reading between the HP and Agilent 3458As, depending on the time of
day.  I suspect the A9 board in the Agilent is the reason since the HP
3458A reads much closer to the 732A value than the Agilent 3458A over
temperature.  Based on long term comparisons with other 732As and a 732B, I
am confident the 732A is stable within 0.1uV.

I am using an ebay A9 board in the Agilent 3458A for the current testing
since I suspect the original A9 board in the Agilent 3458A.  I am
separately testing the original A9 board since I am not completely
confident in the ebay A9 board.  I do not have enough data to draw a
conclusion yet about the original A9 board.  My current concern is to see
if there is agreement that the A9 board is likely the culprit.

Thanks,

Randy Evans

On Fri, Dec 1, 2017 at 1:58 AM, <acbern at gmx.de> wrote:

> What you describe (A9 drift) would be the explanation. However question is
> how much you see. The A9 should not drift much in the intervals you talk
> about (days).
> Also, you should run the ACAL to determine the cal constant not several
> times a day but with several days inbetween, and then divide by number of
> days to determin drift
> Frequent ACAL may not give you good results (random fluctuations such as
> noise, temp... and their impact during cal gets higher the shorter the time
> between ACALs)
>
>
>
> > Gesendet: Donnerstag, 30. November 2017 um 07:40 Uhr
> > Von: "Randy Evans" <randyevans2688 at gmail.com>
> > An: "Discussion of precise voltage measurement" <volt-nuts at febo.com>
> > Betreff: [volt-nuts] Drifting 3458As
> >
> > I have been testing two 3458As against a known good Fluke 732A. Each
> 3458A
> > was calibrated for DCV before the start of the measurements so they both
> > started at the same point.   I have been running a series of tests
> > consisting of measuring the Cal Constant as detailed in Service Note 18
> for
> > each meter several times a day (to calculate the drift per the procedure
> in
> > SN 18).  I also measure the 732A voltage in each meter each time using
> NLPC
> > 100 and NRDGS 100 and then recording the STDEV, MEAN, MAX, and MIN
> values.
> > What I have observed is that the Cal Constant is acceptably low but the
> > MEAN value per measurement  is drifting up in one meter and drifting down
> > in the other.   The unit drifting down has a new A3 board installed and
> the
> > unit drifting upward is an Agilent 3458A only a few years old so would
> not
> > be expected to have a drifting A3 board.  It was only calibrated 1 time
> per
> > the internal REV number, so would have likely been re-calibrated if the
> A3
> > board was replaced.
> >
> >
> >  My question is what is the likely cause of the drift in the MEAN voltage
> > reading if the Cal Constant value is relatively constant?  SN 18 says the
> > drift rate of the Cal Constant is an indication of drift in the A3 AD
> > board, but I believe it assumes the A9 voltage ref board has a constant
> > value over time.  If this is not true (e;g., the voltage ref is
> drifting),
> > I think this would explain the drift in the voltage reading even though
> the
> > Cal Constant is relatively constant.  Any opinions on this?
> >
> >
> > Thanks,
> >
> >
> > Randy Evans
> > _______________________________________________
> > volt-nuts mailing list -- volt-nuts at febo.com
> > To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/volt-nuts
> > and follow the instructions there.
> >
> _______________________________________________
> volt-nuts mailing list -- volt-nuts at febo.com
> To unsubscribe, go to https://www.febo.com/cgi-bin/
> mailman/listinfo/volt-nuts
> and follow the instructions there.
>