Siemens B2086 Decade Resistor: Calibration

I briefly mentioned the Siemens B2086 Decade Resistor in the previous post. In this post I want to provide additional information on this interesting “vintage” unit, which seems to be geared towards AC application, not just DC.

I couldn’t find a date code on it. However, back in the day they published proper manuals including part lists and schematics (yes!). Those are from 1975 to 1981, so it could very well be 30+ years old.

Here is an excerpt of the electrical characteristics specified in the manual:

All decades use metal film resistors, five of the six decades (probably the upper ones) have hermetically sealed resistors. Siemens used pretty much the same topology as my programmable decade resistor. It does look way more complicated with the rotary switches though. For more information on that I refer to my detailed description of the programmable decade resistor. (I actually didn’t realize that until I had a look at the schematics for the first time while writing this post…)

Also worth mentioning is the fact that the two funny looking “holes” in the top-left and top-right corners of the unit are “4/13 coaxial connectors”. To be honest, I never heard of those before. Apparently one could buy different adapters, including some inserts for a BNC connection. Since I have two of those “4/13 coaxial” cables I could build my own adapters if I ever felt the need to use the decade resistor for AC signals.

Now, let’s see whether this decade is in spec three decades later (pun intended). For this test I use an Agilent 34401A’s 6.5 digit multimeter. Its cal string states “16 AUG 2006 23.6°C”. More than 15 years later and at 18.8°C I measured my 1k precision resistor at 1.000,008,4k (+8.4ppm). This should suffice. (Note that there are quite large tolerances even for the 24h spec of the 34401A. This is just a hobbyist measuring an old decade box with a not quite as old 6.5 digit multimeter with effectively unknown cal history.)

All measurements are done at full resolution and with NPLC=100, triggered and recorded via SCPI. Setting the B2086’s resistance value, triggering and performing the measurement for 56 test points took me just under 10 minutes. Here are the results:

Especially for the values above and not including \(100\Omega\) the decade resistor does a really nice job. The switches might contribute to the deviation in the lower ranges despite their “dust-free design” and “self-cleaning” properties. But they still feel very solid and the readings were stable, so for me that’s fine.