Fluke 731B Repair/Modifications

[branadic]

I came across this



a while ago, which I found pretty interesting and watched it recently again, as someone left his F731B for modification and characterization on my desk.

Modification:
Added some small connector on the back

https://www.binder-connector.com/de/produkte/subminiatur-steckverbinder/snap-in-steckverbinder-ip-40

and installed a 5k NTC to the lid of the reference.

Characterization:
I currently have it in my thermal chamber to measure its t.c., so I wonder if someone can share results for comparison? In the slope from 23°C to 15°C I can see some -3.5ppm/-8K = 0.4375ppm/K, but profile is not yet finished.

-branadic-

[bdunham7]

Characterization:[/b][/u]
I currently have it in my thermal chamber to measure its t.c., so I wonder if someone can share results for comparison? In the slope from 23°C to 15°C I can see some -3.5ppm/8K = -0.4375ppm/K, but profile is not yet finished.

I have a late-model one like that with a white face and I've cleaned it up and made a battery pack for it.  I don't have the means to accurately characterize it, but it does have a noticeable tempco and it is negative and appears to be on the same order as yours, although I have no accurate data for you.  I'd be interested in seeing how yours is from 20C to 26C.  I can keep my room +/- 3C from the standard cal temp of 23C, but not +/- 1C.  I considered putting a hole for a temp probe in it, but the way you've shown it looks good.  One issue is that it isn't in one large, heavy aluminum box so there could easily be temperature gradients--so I don't know how well the units temperature variance will be measured by the probe.

[branadic]

Profile finished overnight. Attached are the results I obtained. Not the shape and curve I expected. By the looks of it zero t.c. is at ~28°C, while the parabola is quite narrow. It proves that t.c. for this unit is <0.5ppm/K for temperatures <28°C.

The position of the NTC on top of the lid seems to be appropriate.

-branadic-

[bdunham7]

The position of the NTC on top of the lid seems to be appropriate.

It sure looks like it--the voltage follows the NTC at least 2X closer than the ambient.  Very nice!

Now all you need is a small heater to keep it at 28C. 

[branadic]

Comparison with two F7000 give stable and consistant readings. So the owner of this F731B can be lucky to have a stable and low noise reference, although the t.c. is not perfect. But with the NTC upgrade he can compensate for that.

+	-
F731B	F7000#1
F7000#1	F731B
F7000#2	F731B
F731B	F7000#2
F7000#1	F7000#2
F7000#2	F7000#1

-branadic-

[TERRA Operative]

That video is mine, hopefully I didn't make too many mistakes.

It has been on continuously since that video and is currently stable at 10.000054V @ 23^oC.


Maybe you'll find something interesting in my thread:
https://www.eevblog.com/forum/projects/fluke-731b-restoration-and-mods/

I replaced all the carbon composite resistors with 50ppm metal film, they were all basically out of spec and their temp coefficient is horrible.
low-ish temp coefficient metal film resistors are cheap nowdays, and every little bit helps I guess.