Is my "new" Fluke 731B stabilizing or going crazy ?

[bingo600]

Measuring on my "new" Fluke 731B.

Inside pics here
https://www.eevblog.com/forum/testgear/linux-gpib-and-python-hp3458a-hp3457a-logging-a-fluke-731b-10v/msg775169/#msg775169



*bay text indicated that it has been off for 9 years :

Here is a Laboratory Reference DC Voltage standard. It was taken out of service July 1996 and within calibration at that time. It works as can be seen from the pictures. This item will be professionally packaged by myself. Check my 100 % feedback since 1998. This auction does not include the DVM in the pictures. Thanks for looking. 

I'm using a 3457A & 3458A , see attached graphs in zip
Poweron was 48'hrs before taking any measurements, no adjustments made yet.
I gave it a new set of 2 x 7.2v NiCd batteries , thst's all.

And no 3458A autocal , as i'm just after the stability of the 731B for now.


I think it's looking quite nice in stability for $165, even very "nice".

But have noticed that it has "moved" the voltage "up" during the 5 days i have measured , even with the same temperature surroundings.

Using this for both meters , the 3457 is prob. downgrading to 7digits , and i'm using HIRES of the 3457A, that's why it's "dancing around" (3457A & 3458A graphs in .zip file).

command(HP3458,"DCV 10;AZERO ON;NPLC 100;NDIG 8;TRIG HOLD")

Looking at the DMMxxxxx-s.png graphs (3458A only measurements) , the 731B is now showing above 10v at the same temp , where it was below 10v when i started the measurements.

Is that because it's setteling in

Any tips/hints appreciated.

/Bingo

[Kleinstein]

When the reference is used after a long time, the circuit will be warmer than the surrounding air. So it will most likely start to loose some humidity, as equilibrium hunidity levels will be likely lower than in the cold unpowered state before. This can take quite a long time, much longer than the simple thermal equilibrium.

Drift does not look that high.

[edpalmer42]

Maybe you're expecting too much from the 731B.  After all, it's at the lower end of the Fluke references.

Your graphs show a LSD of 0.5uv or less, i.e. < 0.05 PPM.  But the specs for the 731B 10V output are much higher in the PDF manual that I have.

30 day accuracy:  +- 10 PPM
Temp coefficient:  1 PPM
Ripple & Noise:    < 70 uv RMS

Ed

[bingo600]

@Ed

I'm in no way dissatisfied, as it seems to exceed specs big time.

Just curious about the "slow" >10v change @ same temp

/Bingo

[Bill158]

Maybe you're expecting too much from the 731B.  After all, it's at the lower end of the Fluke references.

Your graphs show a LSD of 0.5uv or less, i.e. < 0.05 PPM.  But the specs for the 731B 10V output are much higher in the PDF manual that I have.

30 day accuracy:  +- 10 PPM
Temp coefficient:  1 PPM
Ripple & Noise:    < 70 uv RMS

Ed

bingo:

You are expecting too much from both your 3458A and the 731B.  Every week I do a inter-comparison between 4ea. 732A vrefs (and between my 731B and one of the most stable 732A VR, just for fun).  I do this by "bucking" the 10 volt outputs and measuring the microvolt differences between them.  I have an Excel plot going back for years on this data.  I have a constant variation of around 2 uv over that period of time.  Remember that 1 uv = 0.1 ppm at 10 volts (10 uv = 1.0 ppm).  Your chart "dmm-2015-10-17_00.01.01-s" shows that you are displaying the measured data over a range of 0.5 ppm.  This also includes the normal temperature and noise variations of both the 3458A and the 731B.  That the changes are below 0.5 ppm over 24 hours for both units is extremely good over that period of time!  The change in output of the 731B from turn on after being "cold" for 9 years is to be expected.  This is why the 731B and the 732A have battery backup for stability.  The Fluke "Reference Amplifier" being not powered up for so long, along with the critical associated resistors, can easily account for this minor change in output of the 731B.  It still looks like less than 1.5 ppm so far.  You may also have more changes in the next few weeks as everything settles out.  You have a "keeper" there!  It also looks like it has kept it's last adjustment extremely well.

congratulations,

Bill

[bingo600]

@Bill
Thank you for the detailed explanation. And the reassurance of my 731B not being "crazy" , you actually just "made my day" 

I got my 3458A used last year , and the info is this :

HP 3458A - Option 002
----------------------
SN: 2823A xxxxx

A5-board shows 9018 , and Roms 9014 - as datecode.

I was actually buying it "while waiting" for a "decent" priced 732A (not $1000+) , but now it seems like i can afford to wait for a "longer time" then i expected.

Funny enough i have seen a 732A go for $700 last week , and there's another decent priced out there now.

I'll prob. play with the 731B for a while.

/Bingo
Volt-Nut beginner

[bingo600]

Seems like my 732B is stabilizing a bit high :-(

Well ....  The temperature is rising also (sunshine on my lab-wall, and central heating turned on)

I'll prob turn off my 3458 , and follow it on the 3457 for a wile , until it has settled

/Bingo

[HighVoltage]

Bingo,

You have long term data now on your 732B, almost a year later?
Would be nice to see how stable it was over a longer period of time.

[bingo600]

@HV

Nope 
I have had it powered on constantly , but not monitored.

/Bingo