Fluke's First Voltage Reference IC - DH80417B - was a data sheet ever published?

[KJ-90]

I recently acquired a Fluke 8100A Nixie DMM that was in excellent physical condition for its age, and had the internal Ni-Cd battery pack option. The 8100A works perfectly ( no cal tests done yet though ), and I properly recycled the old leaking Ni-Cd cells, which I plan to replace ( and rebuild the pack ). However, I've found a voltage reference IC in the unit, called DH80417B. It's in a TO-39 metal can, and has a trademark on the top ( a right hand arrow point going through a lower case letter D ). In the Fluke 8100A schematic, and based upon it's position in the circuit, it's the DMM's voltage reference. I have read this IC was used in a number of other early Fluke multi meters and some early calibrators. However, I haven't been able to locate a data sheet for it. Was a data sheet ever published for this device? From what I know, data sheets were not published for the SZA263 and LTFLU devices, but what about this one?

Although the next point isn't a metrology topic, should the 8100A's tubular metallized foil capacitors made by TRW be replaced? We already know about replacing all the electrolytic capacitors, though. This 8100A has a lot of the Phillips "tropical fish" polyester capacitors, which don't need replacing from what I know.

[montemcguire]

Film capacitors don't wear out, they simply fail catastrophically, but only if the part was basically defective or applied improperly in the first place. Once infant failures have been weeded out, leave all working film capacitors in place.

[Dr. Frank]

That's obviously the same as in the later Fluke 332/335 calibrators.
See this teardown: http://eevblog.com/forum/testgear/fluke-332baf-in-the-slaughterhouse/msg393627/#msg393627

I assume, it's from T.I.

Frank

[Dr. Frank]

Since a long time, I always wanted to replace the Klixon oven assembly by a proper one.
Recently I had a brilliant idea, to simply copy the LTZ oven design upside down to the RefAmp circuit, which requires only 1 additional 1k resistor.
I also changed the output voltage to 10V.

Here's my simplified schematic, and the modified PCB. I also added the same 5 Andreas capacitors, so the circuits runs well and undisturbed. As the SZA263 is TC compensated to zero TC, the elevated temperature will barely change the reference voltage, i.e. no different calibration resistor is required.

For those who do not want to touch the original 15V divider, simply connect the divider for the oven in parallel to the +15V reference.
You need an Op07, a power transistor, and resistance wire. For the oven mass, I recommend a 25x25x25 mm aluminum block, drill an appropriate hole for the SZA263 and wind the wire onto the block. Done.

[Dr. Frank]

Here's another version, for LTFLU w/o an oven.

It's as well w/o the divider for the collector voltage, and I added all necessary Andreas capacitors.
low component count 

[Kleinstein]

With a larger dimension and thus slower oven one likely has to adjust the compensation part for the oven. So C5 in the FB part to set the I part would likely have to get much larger (more like 10 µF instead of 100 nF).