Sensible approach to HP3456A check/calibration?

[guenthert]

[..]
Last thing, I have noticed some hum coming from 3456A, weak enough to be noticed only at night. Is that normal?

   I haven't noticed a significant hum on my 3456A, but my 3478A hums badly (more so and of course at lower frequency) since moving to Europe.  You might want to check for ripple on the supplies.  The acoustic hum is afaik just annoying.

[tatel]

I know DMM Check isn't enough for a 6-1/2 device, but it's the best voltage reference I have so far. That will change in the future but for the time being, it is what it is. I'll check for ripple, thank you very much.

[Wallace Gasiewicz]

I hope that this will not upset too many of the Europeans, but I just passed up a Wheatstone Bridge that had a connection for use as a resistance box at a Hamfest for $ 5. Yes, $ 5 US. It was a US Navy model and had a big dent in the Aluminum Case, but it worked.  Manual was in the box>   
Maybe this is the reason I saw so many Europeans and Japanese guys at the big Dayton Hamfest in the past.  They were buying stuff that they could fit into suitcase for travel and were lamenting the fact that they could not ship the big old radios like Collins. Of course they were Ham Operators and this was a vacation, or at least part of a vacation for them.
I have looked at the Euro Prices of some test gear and am quite surprised at the very high cost compared to US Prices.
C'est tout simplement terrible!!!!!  There are also some good sources for test equipment in Canada but I understand shipping is too expensive to EU. Sphere Research is a good company in Canada:         https://www.sphere.bc.ca/oldsite/index.html       

 As an update I managed to fix the 3456 that I had mentioned in a previous post. I had a repaired board that i swapped and now the unit "calibrates" well with my volt standard and the other 3456.Comment about using the resistance box feature: There is a minimal offset when using the resistance box feature. in my unit it appears to be .013 Ohms . Advertised as less than .02 Ohms.  Accuracy is .05%.

[alm]

Without having these resistors calibrated, I don't know if it's in spec or not, but at least it seems it's not jumping like a goat.   

Obviously this does not constitute a traceable calibration, but I'd say most of the measurements are consistent with both the resistors and DMM being in spec (note that the 3456A uncertainty at 1 GOhm is very large). Or maybe both have drifted an equal amount . Other than a traceable calibration, you could try to find someone with a calibrated 6.5+ digit DMM that could measure the resistors for you. Note that for the lower values how much of the leads are included may matter. Resistor manufacturers sometimes specify how far from the case resistance is measured.

I don't know what the linearity of the 3456A is like, but another thing you could do is measure the 100 Ohm resistor at the 100 Ohm and 1 kOhm range, 1 kOhm at 1 kOhm and 10 kOhm, etc. You would need to do the math to see if this would gain you any additional confidence in the linearity of the Ohms range. Anything further, like measuring ratios between the resistors, would get you deeper into metrology.

I was going to do Ohms Performance test but I'm not sure I'm getting it rigth. Please see attachmente. As I understand it, the red marked steps would be done in autorange, then the green ones would be done in manual 100K range?

I don't think it matters. I'd always use manual ranging during calibration, or at least verify it auto-ranged to the correct range for every point (which I'd argue is more effort than just using manual ranging). I guess they specified auto-ranging to speed up the calibration procedure, since back when the manual was written they may not have automated the process.

I hope that this will not upset too many of the Europeans, but I just passed up a Wheatstone Bridge that had a connection for use as a resistance box at a Hamfest for $ 5.

How does this solve the problem of traceability?

Sphere Research is a good company in Canada:         https://www.sphere.bc.ca/oldsite/index.html

Sphere has been winding down for a while since Walter died about a year ago. This is their current website.

[Kleinstein]

The list of testpoints rely on accurate resistors. So at the moment the tests listed by fluke can't be done. The point that can be done are a few test that don't need accurate resistors, like measuring the same resistor in different range. Another point could be checking a 100 K and maybe 1 M resistor with both AZ on and AZ off mode. The difference between the 2 cases should not be large. Similar a test in 2 wire and 4 wire mode could check if the sense Hi input has much leakage / low input impedance.

[tatel]

That hum worried me, so I just turned off the device. Hoping to have time to check ripple next days.

Thank you all