Attaining confidence

[RobK_NL]

So, I have a Fluke 732A of unknown providence that has been standing around (powered) in my lab for the past six years or so.

I'm looking to sell it (as some of you already noticed  ), but the question that keeps popping up is: Is it stable? (dental drills optional)

How best to go about in getting some data?

I don't have an HP3458, but I do have two Fluke 5440's (which I also want to sell), so the idea is to hook all three up to my HP3457 and periodically measure the differences between them (A-B, A-C, and B-C). Throw in a temperature measurement as well, so as to have all relevant data.
I have an old laptop I can dedicate to this, so I was thinking about a two week run, with measurements ever 15 minutes.

Any comments or suggestions (especially w.r.t. the number crunching) welcome.

[Echo88]

Is the 732a able to be shipped hot? I dont think so, since the battery life is specified with 12h at 23°C. Maybe with an additional battery pack which powers the 732A, its possible to ship it powered to someone with better equipment? If thats possible i suggest you PM the user plesa and ask if he can help you. 

Another possibility: A few users in Germany own 3458As, so maybe its possible to compare the 732A to own of those and get an indication whether your 732A drifts.

I dont have a clue about the number crunching 732A against 5440s.

Maybe users with more metrological knowledge like Dr. Frank or Kleinstein can suggest better ways to check your 732A.

Anyway: Id be interested in getting a 732A/B and maybe one of those 5440. Goddamnit, the voltnut-rabbithole is really deep. 

[TiN]

3458A alone is not stable enough (4.15ppm/90 days) to verify properly operating 732A (1.5ppm/90days).

5440B has better stability spec, so perhaps your best would be connecting 3457 on lowest range in null-meter mode between 5440B and 732A to obtain differential voltage between them.
Given that both of your 5440B's are stable, you can verify setup operation first by using two 5440's in null-meter mode. That's best way you can do, without additional investment.

[RobK_NL]

connecting 3457 on lowest range in null-meter mode between 5440B and 732A to obtain differential voltage between them.

Yes, that's exactly what I had in mind. My assumption is that, by measuring all three instruments against one and other, I will end up knowing something about the stability of each.

[Andreas]

I have an old laptop I can dedicate to this, so I was thinking about a two week run, with measurements ever 15 minutes.

Any comments or suggestions (especially w.r.t. the number crunching) welcome.

Hello,
I would start with around 24 hrs continouosly logging of the difference voltages. (all 15 minutes gives too low information about noise).
By calculation of the Allan Deviation (e.g. "plotter" software from Ulrich Bangert) from the logs you get a quick overview of the stability of each pairing.
From this you can decide on the further tests.

with best regards

Andreas

[Edwin G. Pettis]

With the sole assumption that your 732A is in good operating health, it is the standard by which all the other equipment you mentioned is calibrated from, comparing them to the 732A will get you a fairly high accurate set of readings, a freshly calibrated 732A from Fluke sports a 0±0.5PPM accuracy and a very low drift rate.  Since your 732A has been 'sitting' for around six years without calibration, you will likely have an uncertainty of a bit less than 1 PPM drift per year as is typical for an older 732A, on average a good estimate would be 0.7 PPM to 0.8 PPM/year.  As far as stability goes, a 'golden' 3458A will be in approximately the same drift range per year.  A single Vref is not considered best practice when calibrating high performance instruments such as you have listed, those require a 734A with constant calibration cycling to reach the necessary accuracy/uncertainty.

Your 732A should show you what the approximate drift rates of your other instruments are compared to the 732A but you will have considerable limitations on how accurate they are, they will only be relative to the 732A and its uncertainty.  If you want to know accurately what you've got, it has to go in for a full calibration at Fluke and I hear that they are only doing Z540 cal on them now, you'll get all the dope on it.

[Awesome14]

You could wire the 732A in series with another 10 V reference. Then you'd be reading the meter on the 100mV scale, which is greater resolution of the 10V scale. Do a search on eBay for "10 V voltage reference standard".

[The Soulman]

You could wire the 732A in series with another 10 V reference. Then you'd be reading the meter on the 100mV scale, which is greater resolution of the 10V scale. Do a search on eBay for "10 V voltage reference standard".

Measure 20V on the 100mV range??
Or did you mean connect reference "a" ground to reference "b" ground en use the 100mV range to read the voltage differential between the two positive terminals?
That is what is discussed here from the start.
Do a search on google for "null-meter".

[try]

Hi kaysert,

we don't want to measure a known characterized standard like the 732A against your volt but against the volt at NIST, PTB and others.
Where is your traceability documentation?

Buying some used drifty 732A as you did is not enough.

Best regards,
try

[try]

Hi RobK_NL,

you can compare your 732A with my references. Mine are freshly calibrated with PTB at the maker fair in Hannover.
You can reach me sometimes not too far away from Venlo.

Best regards
try

[HighVoltage]

Hello RobK_NL,

If you ever come to northern Germany, we can compare your 732A to my 3458A and my other 7 1/2 digit meters. Also, I will have my LTZ reference checked very soon in a good calibration lab, so that one is also available for comparison.

[Awesome14]

Hi kaysert,

we don't want to measure a known characterized standard like the 732A against your volt but against the volt at NIST, PTB and others.
Where is your traceability documentation?

Buying some used drifty 732A as you did is not enough.

Best regards,
try

I have a cert from Fluke Calibration for one of our 732Bs, from June, 2017. But if you think about it, the OP only wants to determine stability. Therefore, the only critical factors are a 10V output and stability. If one or both of the standards are a bit off of 10V exactly, it really doesn't matter for what the OP wants to do.

Anyone is welcome to come to our facility, and view our traceability documents and calibration procedures. Unlike you, I actually try to help others. And when you say "we", who is the other person? I think it's just you and your subjective, uninformed opinion!

You can cut me down and interfere all you want, but you'll always be incorrect, because you don't want the truth. Your arrogance causes indignation in others. When the proud are cast down the simple learn wisdom. The just man falls seven times, but he always rises again.

Trouble-makers stumble to their ruin! You are a trouble-maker. I don't why you make unfounded criticisms of myself and my product, but I can only assume it's jealousy. You're ill-mannered and unwise.

[Echo88]

While were at it: Are 732As actually able to be shipped hot, despite their 12h battery-spec in the datasheet or how are they calibrated?

[Awesome14]

While were at it: Are 732As actually able to be shipped hot, despite their 12h battery-spec in the datasheet or how are they calibrated?

The 732A has an external battery pack option. But you could just as well wire (2) 12V 7Ah batteries in serial, and wire that to the external battery connector. Then, if you shipped it overnight, it would still be live. I tested the battery life in a 732A, and it was still going after 15 hours.