[volt-nuts] RE "new" 3458A

[Bill Gold]

Randy:

    Look at page #23 of the 3458A Calibration Manual.  Make yourself a "4
Terminal Short" as shown and put it in as shown.

    Before I did an ACAL I had around -000.00025 mVDC.  After ACAL I now
read +000.00002 mVDC.  I did the "CAL 0" myself so I would expect that the
meter should return to a low value, and it does.  I do get a variation of
+/- 30 nVDC using 100 PLC and just observing the variations.  As I remember
I have never seen a spec on the ZERO stability over temperature.

    If I turn on the MATH function and then do 40 measurements with 100 PLC
the statistics show:

Low reading        -70 nVDC
Mean reading       -28 nVDC
High reading        +3.5 nVDC
Total Variation    73 nVDC

    So that correlates with my visual observation of 60 nVDC.  After an hour
the room had gone up around 1 degree C.  Then I observed -000.00023 mVDC.
After another ACAL the reading was again +000.00002 mVDC.  This particular
meter has a negative tempco as the room temp goes up.

    Obviously do an ACAL before any precision measurements requiring low
nanovolts.

    Go to the Keysight website and go to "Technical Support" and choose
"Parts".  Then enter in the "Part Number"  "03458-66517" which is the
replacement "03458-66507" assembly and you will see the replacement part
number on the right hand side.  Click on that and you will get the
information about the exchange program and so on.  Looks like you can just
order this part online and pay for it with a credit card, but you have to
create or use an existing login account.

    I needed a new display a few years ago.  At that time you could order
just the display for around $80.  Being extremely good at removing and then
inserting and soldering I ordered the part.  The problem was that the
spacing from top to bottom of the pins had changed.  It went from around 1.3
inches to around 1.5 inches.  So I had to bend the pins to fit my display
board and then get something like 72 pins into the holes on the PC Board.
It took hours.  This change is probably why HP/Agilent/Keysight doesn't let
you just get the display anymore but wants you to get the whole PC Board
assembly.  It did work just fine once installed.

    Hope this helps your decision to keep or not.

Bill



----- Original Message ----- 
From: "Randy Evans" <randyevans2688 at gmail.com>
To: "Discussion of precise voltage measurement" <volt-nuts at febo.com>
Sent: Saturday, August 16, 2014 10:36 PM
Subject: Re: [volt-nuts] RE "new" 3458A


> Interesting note.  After the room cooled down from about 79F to 73F, and
> another ACAL, the meter now reads +000.00035 mVDC, a more reasonable
value,
> although it does bounce around a couple of tenths of a uV.
>
> Maybe that is OK?  If so, then the only issue would seem to be the
display
> has some faint pixels, which a new display should fix.
>
> Randy
>
>
> On Sat, Aug 16, 2014 at 9:59 PM, Randy Evans <randyevans2688 at gmail.com>
> wrote:
>
> >
> > The unit seems to be working so far except for one issue.  After doing
an
> > ACAL, and making sure the Auto Zero is ON,  I short the input leads with
a
> > copper wire shunt across the inputs and the reading is approximately
> >  -000.0023 mVDC.  That seems rather high.  I would expect the unit to
short
> > the input leads internally and force a zero reading during the ACAL.
> > Anyone have any comments on this reading?
> >
> > Thanks,
> >
> > Randy
> >
> >
> > On Sat, Aug 16, 2014 at 3:37 PM, Richard Moore <richiem5683 at gmail.com>
> > wrote:
> >
> >> Hi Randy -- sounds like your unit is in cal, based on your measurements
> >> of DCV and precision 10k resistor.
> >>
> >> Using autocal all is recommended before doing precision measurements,
and
> >> I do that if it's been more than a day or two since last use. The
autocal
> >> uses the internal Vref and an internal 10K resistor to do cal on
everything
> >> else, so that tells you what the basic cal procedure is. I just got my
3458
> >> back from Loveland, and that's what they did for me -- warmed it up,
then
> >> ran autocal, then measured everything against a Fluke 5700, aided by an
HP
> >> 3325, and another 3458.
> >>
> >> It has been 5 years since I replaced the display board (no "exchange"
> >> deal was available then AFAIK, so I don't know what's changed) and also
the
> >> NVRAM board, which was dead, with one with the Snap-cap RAM chips. I
did
> >> those replacements, then sent it home for cal, which was complete,
since
> >> all the RAM was new. Now after 5 years, the unit passed all incoming
> >> performance tests and was sent back to me without a cal process of any
> >> kind. This tells me that an old, well-aged Vref module is a good thing.
The
> >> 10VDC test had changed by a bit under 5ppm, or roughly 1ppm/year.
> >>
> >> They have a cal deal -- use code 1.090 -- press them for it -- and that
> >> saved me 30% off the normal price. I think this deal lasts until
> >> mid-September, so my recent "cal" ended up at just under $400 including
> >> shipping. I'm not sure the deal is available on new or first-time cals;
my
> >> unit was in their data bank.
> >>
> >> But this is a long way of saying I don't think you need to send it for
> >> cal -- just push Auto Cal and Enter and wait about 10 minutes and you
> >> should be good to go.
> >> _______________________________________________
> >> 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.