prototyping my at-home resistance standards

[Marcus_S]

When starting playing around with multimeters last summer I was looking for an at-home-resistance standard... This resulted in...


Resistor #1

I ordered a VISHAY S102C 10k 0,005% from Bürklin in Germany. I got a resistor labeled with Bedek. Bedek seems to be a distributor of VISHAY foil resistors but it was somewhat surprising to find the resistor not labeled as VISHAY.

For this resistor I made a casing to protect it against mechanical damage. The leads were bent carefully and were soldered (avoiding any mechanical stress) onto large pads of thin pcb material while cooling the leads. The pcb was glued to the casing. The terminals were contacted using my special very-ugly-wiring technique. The casing is milled from aluminium - not very spectacular. A hole is bored into the casing to allow temperature measurement under the resistor.
(Bedek_1.jpg, Bedek_2.jpg, Bedek_3.jpg)


Resistor #2

some times later I thought about an alternative as a second resistor. Since I like to make some chips I took a block of aluminium and a milling machine and suddenly there was another casing.

A milled pocket for the resistor, an o-ring sealing of fluorinated rubber in the lid. The terminals were turned from pure copper and pressed into the aluminium body being isolated by PTFE-bushings. The contact zone of the copper to the PTFE is about 20 mm in length to achieve a good thermal contact. The terminals were washed with a solution of 1H-Benzotriazol (which is a corrosion inhibitor for non-ferrous metals, especially copper) in isopropyl alcohole. We will see how this performs in future. With this very tight press-fit and the o-ring I tried to obtain a semi-hermetic casing. Additional blind holes were made for inserting a temperature sensor and for contacting a shielding lead.

Buying a hermetic sealed foil resistor is not exciting - so I experimented with a wirewound power resistor (Dale RH25 10k, 1%). These resistors are cheap and easy to assemble and robust (I hope) and I found the temperature coefficient to be surprisingly low (about 3 to 5 ppm/K depending on the specimen) around room temperature.

By this I tried to get a simple at-home-standard which might be driven with more than 1 mA if necessary. And with its mass of 1 kg it sits on the table and will not be irritated by the cables...
(number2_1.jpg, number2_2.jpg, number2_3.jpg, number2_4.jpg)


Characterization

Using a small setup made from a fan and a peltier element and some aluminium I estimated the overall temperature coefficient of both resisitors in the casing (linear regression). The picture (tc_setup.jpg) of the setup is somewhat arranged, during the measurement the resistor was isolated with a towel, the temperature was measured in the borings in the casing using a Pt100. The measurement was done using a 3456A, 4w OCOMP, 100 NPLC and the resistors were heated very slowly. By this I estimated a t.c. of + 3.6 ppm/K (Dale) and - 1.3 ppm/K
(Bedek) (R_vs_temperature.jpg)

Then the resistors were measured from time to time for two months using the well warmed-up 3456A (4w, 100 NPLC, fixed 10 k-range). As raw data resistance, temperature of the resistor and relative humidity of the air (measured with a cheap hair hygrometer) were recorded. The "real" resistance was calculated using the estimated t.c. (R_vs_time.jpg, relative humidity is related to the right axis).


The results are a bit surprising to me. As expected the influence of temperature to #1 can be corrected easily and is therefore less important. But this design shows large variations of about 8 ppm which seem to depend on humidity of the surrounding and which I find not as easy to correct. The cause is not clear. I must have made a mistake in the design of the casing, creating some leakage paths at the connectors or the pcb due to humidity or whatever... Or are such resistors so sensitive towards humidity?

On the other hand resistor #2 can be corrected for temperature influence and seems to be undisturbed from humidity and shows only some small fluctuation in resistance value. Therefore either the casing does its job or the resistor is stable. Or both. The cause of these variations of about +/- 2 ppm is somewhat unclear to me. It might be influenced by the temperature measurement and correction calculation as 0.1 °C (resolution of temperature measurement) gives 0.36 ppm change in resistance. With some drift of the meter this gives 1 digit easily.

But it is interesting to see how stable the old 3456A seems to be. Looking at this data I wonder how to interpret the specifications of the meter and how this stability is achieved.


This short-time study is finished now as I opened #2 to take some pictures and I adjusted the 3456A to a calibrated Tek DMM4050 (3456A was approx. 20 ppm off). But the performance of the casing seems to be satisfying.


As next steps I think about improving the temperature measurements and the comparison of the resistors, determining the t.c. more carefully and I plan to duplicate or triplicate #2 and to fill it with the S102C or another Dale RH25 10k (or a RH25 1k as another value because some 1 k seem to have a t.c. around 1.5 ppm/K...). And then I will start a new measurement series.


Best regards

Marcus

[alanambrose]

Well that is super interesting. Can I ask - what is the corr. line in the last graph - is it the residuals? Also do you have the stats for the regression - and did I miss the coefficients for rh? Very nice build btw.

Alan

[Andreas]

Bedek seems to be a distributor of VISHAY foil resistors but it was somewhat surprising to find the resistor not labeled as VISHAY.

Buying a hermetic sealed foil resistor is not exciting - so I experimented with a wirewound power resistor (Dale RH25 10k, 1%).

The results are a bit surprising to me. As expected the influence of temperature to #1 can be corrected easily and is therefore less important. But this design shows large variations of about 8 ppm which seem to depend on humidity of the surrounding and which I find not as easy to correct. The cause is not clear. I must have made a mistake in the design of the casing, creating some leakage paths at the connectors or the pcb due to humidity or whatever... Or are such resistors so sensitive towards humidity?

Hello,

Impressive results. Thank you for sharing.

Bedek is a precision center -> they trim (manufacture) some of the Vishay resistors to your needs under licence from Vishay.
Other types are traded.

I never thought that their T.C. is such low. (I remember the spec of around 80 ppm/K).

Humidity is a problem of the epoxy package.
The epoxy takes rH from the air and swells.
The mechanical dimension change creates forces on the resistive element and changes the resistance.
The largest problem with humidity is the large time constant of typical 3-7 days.
(for DIP8 packages, of course depending on thickness of the epoxy).

The RH resistors usually have also a epoxy packaged resistor within the aluminium (at least for the lower wattages).
Perhaps you should wait longer time for the RH25 package to see a rH influence.

With best regards

Andreas

[Marcus_S]

Thanks for the comments.

Alan, the lines with the small triangles are the resistance value as measured. The large triangles (labeled with "corr" = corrected) are the resistance values after correcting the measured values for t.c. using the indicated value (-1.3 ppm/K for S102C and +3,6 ppm/K).
The turquoise line (right axis) represents the relative humidity of the air and this is the same curve in both graphs.   
And I like to see how good this t.c.-correction works (for RH25 at least).

Andreas, yes, I was surprised too. And I found the t.c. of the RH25 1k to be even lower. But with the 100R the t.c. is much higher again. The best values seem to be the RH25 1 k and 10 k. But The RH50 1k, 10k and 100 k are very interesting, too... It depends a little from the resistor batch.


Best regards

Marcus

[lowimpedance]

Very impressive metal work. Clearly some careful thought went into it , and quality effort  .

[zlymex]

Casing and mounting, superb! Nice design, good milling, and also incorporate a seal ring
Resistor elements, not so good, especially compared to the effort above.
Adjustment/Compensation: N/A
Connectors, good enough for 10k resistor.

Temperature regulator, seems to be quite nice
Test equipment, not so good. A standard resistor and the test should be better than the meter.
Post test analysis, great.

[quarks]

I totally missed this.
That is indeed very well made with nice attention to details

Posible improvements are maybe good binding posts (Pomona 3770, MC PK4-TS or even lowthermal 2758) to be able to connect crimped spade lug or plane copper wire and a temp sensor inside the case.

Thanks a lot for sharing.

[Marcus_S]

Thank you very much for all the feedback and the input.
I thought a lot about the comments and that causes one question, please help me with that:

What is the real advantage using binding posts (Pomona, whatever...) compared to massive contacts made from pure copper (like in "my" design) and using small 4 mm connectors (like Multi Contact LS4). I think the contact resistance could be neglected when using four-wire measurement, therefore I see the only improvement in fresh metal surfaces which cause less errors due to thermal voltages. But as far as I understand these errors could be minimized by cleaning the contacts and by minimizing the temperature differences between the contacts?


The idea to use home-made copper contact-feedthroughs is to achieve a semi-hemetic casing. I assume that standard binding posts will not be that gas-tight. Nevertheless it is not complicated to design another type of PTFE-copper-feedtrough-connector which allows 4 mm-plugs and 6 mm spade lugs and make a new casing or exchange the connectors of an existing casing.

Regarding machinability of copper: I found no problems during machining when using sharp tools (good fresh drillbits, fresh polished carbide inserts for turning). Therefore I assume that the usage of other copper(-tellurium) alloys will be interesting for real production scale but I think it is not necessary for home-built devices.

And I would like to add a remark regarding the temperature sensor: I decided not to incorporate a temperature sensor into the chamber in order to avoid any additional feedthroughs into the casing. Therefore I made the boring into the casing which ends directly under the resistor. By this I can measure the temperature in the neighbourhood of the resistor in a distance of approximately 2 to 3 mm, separated by aluminium. And by this I can use the same (at-home-calibrated) temperature sensor for different measurement setups and different resistors.


In the meantime I observed some miraculous multiplication... Now there are three additional casings (being only slightly modified):

#2: the "old" prototype (equipped with Dale RH25-10k)
#7: equipped with Vishay (Bedek) S102C 10k 0,01% with a small bag of desiccant (aprox. 1 gram. With this setup I should get a feeling for the performance of the sealing and the feedtroughs as the resistor is very sensitive towards humidity. Perhaps somewhat higher stability vs. time compared to the RH-resistors?)
#8: equipped with Dale RH25-10k 1% (as a comparison to #2)
#6: equipped with Dale RH50-1k 1% (as a comparison with another size and value and to "make" 1 mA or 10 mA more easily)

I did not ignore your comments and I thought about the comments carefully. I keep them in mind for the future but now I was too impatient to have a look on the performance of this design. I am especially interested in the behaviour of the Dale-resistors and the performance of the feedthroughs and the O-ring-sealing. While observing this I will think about the "improved next generation" of resistor casings. And: it is no big effort to substitute a RH25 by a better resistor if I find one.


Temperature coefficient:
For characterization I measured the temperature coefficient of the resistors (3456A, 4W, OCOMP, 100NPLC, temperature measurement using R&S UDL-45 with a home-calibrated Pt-100) as described above. And I like the results:

#2: Dale RH25-10k      t.c. = +3,85 ppm/K
#7: Vishay S102C 10k   t.c. = -1,91 ppm/K
#8: Dale RH25-10k 1%   t.c. = +3,87 ppm/K
#6: Dale RH50-1k 1%   t.c. = -0,73 ppm/K


Thermal voltage:
As a rough estimation (not a real measurement!) to study the influence of temperature towards thermal voltage I made a small quick-and-dirty test setup:
- setting the temperature with the peltier-device,
- isolation of the resistor using a towel wrapped around the resistor,
- measuring the temperature in the temperature-sensor-boring (as described above) with a home-calibrated Pt-100 with the R&S UDL-45,
- control of temperature stability with a thermistor connected to the 3456A obtaining a resolution of 1 mK,
- measurement of thermal voltage at the sense connectors of the resistors using a Keithley 2182 nanovoltmeter connected using a home-made cable (home-made connector, Cordial CMTOP222 low-noise microphone cable, Multi Contact LS4 plugs, soldered with Sn-Ag)

The results: The values are the values as indicated at the display of the meter. They are not corrected in any way (e. g. influence of the input bias current (spec: < 50 pA)).
All resistors show a "large" thermal voltage of up to 0,5 to 0,6 µV during heating. When reaching a "thermal equilibrium" (observed reading the display of the 3456A, being "constant" for some seconds) the thermal voltage reaches a "constant" value very quickly (within approx. 20 s).
#8 shows large noise in thermal voltage (and #8 shows a large standard deviation during resistance measurement, so this is a good candidate to be substituted by a better resistor).
But all resistors show a surprisingly low thermal voltage effect (what went wrong?). Therefore I assume that the design of the connectors seems to be not that bad.


The next steps:
I think I can (re)start now to monitor the resistors as good as I can. At the moment I have no better equipment than some 6,5 digit multimeters and no stable current/voltage source unfortunately.  So I have to improve my equipment and I have to learn how to measure and compare these resistors. But this will take some time.


Best regards

Marcus

[alanambrose]

Well again - all very interesting. I think you're onto something. I admire the quality of the machining. Did you consider incorporating a heater to keep them a bit above ambient? Looking at the TCs of the various builds, I wonder whether incorporating the right number of +ve and -ve TC resistors to give approximately zero TC would be an interesting strategy?

Alan

[Marcus_S]

Alan, thank you very much! When measuring the t.c. of the 1k-resistor I thought a nanosecound about making "my-ultimate-10k-resistor" by combining 10*1k in series. But the casing will be ... large...

I did not think about a heater - the sealing has to do the job. The idea was to get a cheap and small and simple at-home-standard as good as I can. And therefore I did not try to achieve the exact 10k (but if I would combine 10*1k... They could be assembled onto the outer of a medium-sized pentagon for good thermal contact... and then being put in a bin... I have to think about...).

Combining different resistors to get a better t.c. will be a good idea. But then it seems to be much easier to use other (better) resistors. And then I have to think about a new type of casing (as mentioned above). I think the casing should not be made much larger as the difficulties to make a good sealing at the lid will grow. I assume that for my purpose a good characterized t.c. of 1 to 5 ppm/K is good enough at present.

I think for the next time I have to study the performance of the sealing (at least 1 year cycle). Other topics I have to care about are
- how to measure/compare (the) 10k-resistors, getting a good current source (and I am no electronics expert)
- how to generate 1 true ampere (or how to measure it)
It seems simple in theory but I have to do this at home as a little hobbyist.


Best regards

Marcus

[Conrad Hoffman]

With that level of metal work, I think you should consider winding your own resistor. It should be possible to get a spool of manganin and maybe do a copy of the Guildline resistor shown in the tear down thread. The wire in those isn't supported, but is loosely wound (to avoid strain) and sits in oil. You wind it a tad over value, then use a very large conventional resistor to pad it down, with a small trimpot if desired. There are several threads on the topic here if you search around a bit.

[Marcus_S]

Hello Conrad,

thank you very much. I had a look into the drawer and found some Isotan wire... I like the idea to make a self-supporting winding (I thought a lot about good supporting material before reading your comment but I had no idea). Isotan will no be the best material regarding thermal voltage but perhaps I should test this.

So I started to think about a new design... And this might be an example that life is impossible without a lathe and a milling machine...


Best regards

Marcus

[zlymex]

What is the real advantage using binding posts (Pomona, whatever...) compared to massive contacts made from pure copper (like in "my" design) and using small 4 mm connectors (like Multi Contact LS4).
......

For one, binding posts are universal(5 ways), which is convenient if you have the standard calibrated/measured elsewhere.
For two, Multi Contact LS4 is made from beryllium copper, which has large EMF with copper.

[ap]

Copper beryllium is perfectly fine, Fluke e.g. states they use it for their 8508 DMM inputs. Same with CuTe. Same with gold plating. Exception is measurements in the low nV range.

[zlymex]

Copper beryllium is perfectly fine, Fluke e.g. states they use it for their 8508 DMM inputs. Same with CuTe. Same with gold plating. Exception is measurements in the low nV range.

According to the information provided by quarks, beryllium is much worse than gold:
https://www.eevblog.com/forum/chat/help-wanted-low-and-high-ohm-measuremet/msg174908/#msg174908

[ap]

Neither have I seen differences between CuBe or CuTe baed cables (gold plated) in conjunction with use on the 3458A nor would such a high EMF (5uV/K) justify use of CuBe in the 8508A. But this topic has eben beaten to death anyway. I leave it there.

[zlymex]

Neither have I seen differences between CuBe or CuTe baed cables (gold plated) in conjunction with use on the 3458A nor would such a high EMF (5uV/K) justify use of CuBe in the 8508A. But this topic has eben beaten to death anyway. I leave it there.

There is a product cat. http://www.lowthermal.com/pdf/2758-series-data.pdf
" The use of tellurium copper instead of beryllium copper or brass, ...... ensure these posts exhibit the lowest contribution of thermal EMF. "
This is also a statement that beryllium copper is inferior.

[Conrad Hoffman]

I'm always wary of marketing claims, though a lot of what we want to know is only available from manufacturers. The choice of Nylon for an insulator surprises me as it's very hygroscopic.

[Marcus_S]

Hello zlymex,

thank your very much for your remarks.

To this day I did not think about "elsewhere" - about the idea to get my "at-home-standards" measured in a professional calibration lab. For this purpose 5-way-binding-posts will be useful. But such a binding post for contacting bare wires can easily be made with a lathe and some pure copper. I think there is no need to buy a binding post made from a whatever-alloy to contact this to a pure-copper wire?

And regarding the thermal voltage of the Cu-Be Multi Contact LS4: in my design I thought about trying to achieve a good "thermal symmetry" between the copper contacts of the resistor casings and I neglected the influence of thermal voltages caused by the different material(s) of the banana plugs. I have to re-think the design for the next generation of casings...
But perhaps I should do some additional measurements of thermal voltages of different contacts and materials in the next future.


Thanks again and best regards

Marcus

[VintageNut]

Pure copper tarnishes very fast. My DIY solid copper tetrajunction has to be cleaned every time it is measured.

[ManateeMafia]

@VintageNut

How about applying a thin coat of DeoxIT to the copper? That might buy you some cleaning time. The Keithley 262 called out Cramolin Red for the copper pcb but it looks like it isn't manufactured anymore.

[VintageNut]

I think that the only thing you can do with solid copper is to connect it "permanently" to other solid copper. The final connection to the outside world should be gold plated copper binding posts. This is how the front end of the KE260 nanovolt source is built.

You clean everything one time when the device or instrument is built. After that you just let it age.

As an aside but on point for this thread, I visited a calibration lab recently. The resistance standards are individual devices that are encased/sealed. They are all over 30 years old and are measured regularly. The more that is known about the aging and drift, the more valuable the resistors are.