Precision Resistors - General Thread

[leighcorrigall]

Hello Everyone,

Let's talk about precision resistors! This is an open invitation for members to contribute to a list of resistor sources and provide a discussion about their own experiences with resistors. 

- manufacturers
- series/models
- specifications
- types
- applications

Seeing as there is interest in building environmental test enclosures https://www.eevblog.com/forum/metrology/diy-large-thermal-chamber-for-metrology-tm-testing/?all, maybe we can also combine our efforts to measure their performance specifications to either confirm or disprove claims made by the manufacturers. Personally, I am very interested in confirming voltage and temperature coefficients. My Advantest R6581T is being sent to the TransCat calibration centre in Texas, so it might be a while before I can contribute to that venture...

So far, I have collected a list of my own (MegaOhms to TeraOhms range) and maybe we can combine our knowledge to benefit the community. Here is a link to my Google Sheets work-in-progress:

https://docs.google.com/spreadsheets/d/1HSdXGIUwK_Is3uJIlgVGw9Qx-jLj2GVnV8qsp52-f8w/edit?usp=sharing

EDIT: The above spreadsheet outlines the best-known commercial resistors that I am aware of for a specific decade. The original intention of the spreadsheet was to help me decide on resistors for a precision decade box.

I have made the document viewable and you can comment on details if I have made mistakes or you want something else included. Preferably, one should discuss and comment at the EEVBlog so that the information is accessible.

Looking forward to your responses!   

That's it for now.

[TiN]

Good idea to start with methodology, on how we represent temperature and voltage coefficient. It is common piece of specmanship from vendors as they can use generic word "temperature coefficient" while showing only specific typical data or showing "averaged" tempco out of large resistor lot, rather than more difficult guaranteed maximum specification.

I often test resistors and parts in smaller chamber with slooow and smooth temperature ramp from low to high temperature and back. Doing it in cyclic manner help to reveals possible hysteresis issues (often common aspect of physically larger resistors and wirewound resistors). Temperature range I commonly run is 18-50C to include metrology grade 18-28C range and wider common PCBA temperature.

My dataset from few years back, based on collected and measured resistors is already published on my site, but these articles cry for major overhaul...I also tested hundreds more resistors since then.

 https://xdevs.com/article/tcr_test/
https://xdevs.com/article/caddock_news/

I also tried combining multiple 1 Ohm resistors into one combined resistor but failed miserably due to large error from copper wiring TCR.

https://xdevs.com/article/1r_tests/

Also bonus, little Fluke 742A factory speclist database and minor repair of few 742A I had hands on here : https://xdevs.com/article/f742a/ .

[veedub565]

Maybe a little OT but I'd be interested to hear peoples thoughts on this source of lab standard precision resistors

https://www.ebay.co.uk/itm/331951651645?hash=item4d49dc3b3d:g:VUwAAOSwLfNa2EXb

[leighcorrigall]

Maybe a little OT but I'd be interested to hear peoples thoughts on this source of lab standard precision resistors

You could also try https://www.eevblog.com/forum/metrology/teardown-standard-resistors/?all.

[veedub565]

Thanks for the link, there is a lot of detailed information in that thread.

[manganin]

Maybe a little OT but I'd be interested to hear peoples thoughts on this source of lab standard precision resistors
https://www.ebay.co.uk/itm/331951651645?hash=item4d49dc3b3d:g:VUwAAOSwLfNa2EXb

Those Krasnodar type standard resistors can be very good, but the quality varies a lot. You may need to buy ten to get one decent. And in real life the seal is far from hermetic, so nowadays it is hard to find anything not ruined by bad storage.

The primary resistor set of the Estonian national laboratory used to be Krasnodar, but that was at least ten years ago and they may have upgraded. Of course the examples found in high level labs are carefully selected.

[Grandchuck]

Here is my reference resistor for the 3458A DMM and other applications.  It is mounted in a Bud Industries enclosure and uses a Vishay VHP101T metal foil resistor.  I would like to have it calibrated without  spending a lot.

[TiN]

Send it over, I can do +/-0.4ppm uncertainty for you on it 

[Dr. Frank]

Here is my reference resistor for the 3458A DMM and other applications.  It is mounted in a Bud Industries enclosure and uses a Vishay VHP101T metal foil resistor.  I would like to have it calibrated without  spending a lot.

Hello Grandchuck,
you have a solid 4W Kelvin connection, the case has a seal and it provides some thermal mass, but your assembly is lacking a proper thermal connection between the resistor and the case. And it's lacking a stable/precise thermometer, to first determine the T.C., and afterwards mathematically compensate the resistance value for its own T.C.,  compare the setup of the SR104.

Without that, it makes no sense to calibrate it to a certain value, because the T.C. might be as high as 0.3ppm/K (acc. to the datasheet of the VHP101).
If Illya offers 0.4ppm uncertainty, I would strongly suggest a proper setup first.

Frank

[manganin]

My suggestion would be fill the box with styrofoam with a small space in the middle for the resistor and a thermistor. (Check the insulation resistance of the material chosen of course.)

The idea is to make the thermal time constant from the outside world to the resistor as long as possible. Two advantages:

1. temperature stays steady during measurement

2. protects the resistor from temperature extremes during transport or at least makes the rate of change less violent.

[Dr. Frank]

My suggestion would be fill the box with styrofoam with a small space in the middle for the resistor and a thermistor. (Check the insulation resistance of the material chosen of course.)

The idea is to make the thermal time constant from the outside world to the resistor as long as possible. Two advantages:

1. temperature stays steady during measurement

2. protects the resistor from temperature extremes during transport or at least makes the rate of change less violent.

Hi, in this case it's much better to have a big thermal mass (aluminium or oil filled cavity) to stabilize / slow down temperature changes.
Isolation with styrofoam is usually done on self heated applications, like in OCXOs.

The thermometer and the resistor have to be tightly coupled, e.g. assemble them in the same aluminum block.

If you ship such a standard ( Ohm, Volt) over several days in cold weather, a styrofoam isolation does not really help, it just looks as if it would protect from excessive cooling down, but in reality it doesn't. We already made our experiences with "cold" shipped LTZ1000 inside a big isolation package, and they simply shifted anyhow.
Only way around is to monitor the shipment temperature inside the package to find out if the temperature fell below a certain critical threshold (+15°C in this case), or to ship the standard "hot", or to invent a rejuvenation process, to mitigate possible hysteresis shifts (similar to the Fluke 7000 voltage standard).

The bulk metal foil technology also shows hysteresis on big temperature changes. Don't know how much this is on the VHP101 chip, but the Z foil technolgy (VHP202Z) which I use, showed +5ppm on a -40°C trip. I could remove that hysterisis by thermal cycling.

Precision wire wound is much better in this aspect.

Frank     

[Kleinstein]

+1 to adding thermal mass. Part of this could be some silica gel, to also get a stable humidity. It is not that important with a hermetic resistor, but the temperature sensor may not be hermetic.

For the 4 wire connection there is no need to have very low resistance from the drive to the sense terminals it is more like aiming for a defined point where the 3 connections to the terminals and resistor meat.  With a 10 K resistor this is not that important though. So the design looks a bit sloppy but is OK for 10 K, but would be a problem for 10 Ohms.

[TiN]

+2 to thermal mass and temp sensor.

[TimFox]

To Grandchuck:
Is there a reason why you soldered the resistor wire leads to the bars, rather than clamp them to the bars with additional screws?

[leighcorrigall]

Here is my reference resistor for the 3458A DMM and other applications.  It is mounted in a Bud Industries enclosure and uses a Vishay VHP101T metal foil resistor.  I would like to have it calibrated without spending a lot.

Hi Grandchuck,

Can you be more specific about what you mean by a 'reference resistor'? Is this simply to periodically track the measurement drift of the 8.5-digit DMM over time to ensure it does not fall out of tolerance?

If drift monitoring is the case, then one should ensure that the drift from the resistor is significantly lower than the instrument so that the drift is representative. A calibrated resistor will not be important for this task but calibration would be necessary as a transfer standard.

Edwin Pettis sent me a link to a recent post he made about resistor terminology that may interest you. https://www.eevblog.com/forum/metrology/precision-resistor-specifications-and-definitions/msg1557151/#msg1557151

Have you made any resistance measurements with your 3458A over time and compared them with the specifications of the instrument?

Apologies for yet another round of criticism, but what is your reasoning behind the black washer underneath the nut and metal washer? If this is a permanent test fixture, why not purchase a 4-wire resistor and directly solder the leads onto the binding posts after wrapping them around the posts a few times?

For resistor box designs, I personally enjoy how zlymex has made his resistor box with dual shielding and I may borrow some of his ideas when I receive a VHA518-11Z (Y4773310K0000S4L) later this year. He is using VHP101 Series resistors like you and has claimed to have excellent low drift results (see enclosure labels). https://www.eevblog.com/forum/metrology/teardown-standard-resistors/msg892700/#msg892700

My Advantest R6581T has been sent to Texas by TRANSCAT for calibration. This amateur is leaving it up to the experts. Once I have an accredited instrument (adjustment with data and uncertainties), I plan on calibrating the rest of my test gear with it.

If you want a reliable resistance transfer standard, you could consider:
http://www.ohm-labs.com/resistance-standards/index.html
http://lowthermal.com/resistance-standards.php
https://guildline.com/primary-electrical-metrology/resistance-standards/precision-dc-air-standard-resistors
https://www.ab-precision.de/products/electrical-standards/ (ask if it is appropriate for an 8.5-digit DMM)
https://us.flukecal.com/products/electrical-calibration/electrical-standards
https://www.ietlabs.com/esi-sr1-calibration-resistor.html

Personally, I would look at the teardowns before investing ~2000 USD though. https://www.eevblog.com/forum/metrology/teardown-standard-resistors/?all

Regards.

[Grandchuck]

I am going to mount a solid chunk of aluminum in the bottom of the case (see photo).  What should be used to insure thermal conduction?  Is heat sink compound OK?
Do you recommend attaching a heat sensor to the aluminum chunk?  Is a thermistor OK?

A quick (crude) first measurement yields a negative temperature coefficient of about 0.15 ppm per degree C.  I will be making more careful measurements.

I soldered the resistor because it was easy.  How much trouble will this cause?  I used heat sinks on the leads, by the way.

The black washers are lock washers.  They were supplied with the banana jack assemblies.

I have been mostly Volt-nutting and felt the need to try Ohms for a while.  Also, having something available in the way of a resistance standard seemed like good idea.
I have several 10 volt references that I feel reasonably good about, so I guess this is the next step.  Having fun, in any case, and learning too!

TiN:  that is soooooooooo generous of you.  Might take you up on it!

Big thank you to all who posted.

[Dr. Frank]

I am going to mount a solid chunk of aluminum in the bottom of the case (see photo).  What should be used to insure thermal conduction?  Is heat sink compound OK?
Do you recommend attaching a heat sensor to the aluminum chunk?  Is a thermistor OK?

A quick (crude) first measurement yields a negative temperature coefficient of about 0.15 ppm per degree C.  I will be making more careful measurements.

Big thank you to all who posted.

Hi Grandchuck,

we already had a similar thread like this one, and there's my assembly, showing how the VHP, precision thermistor and case are all thermally connected together.
Most important is, that the NTC and the VHP are closely assembled into holes of the aluminium block. Yes, I simply used thermal grease for that purpose.
The oil fill of the VHP resistors gives a very good thermal conductivity between the resistive element and its case.
The lower the T.C., the less important is the accuracy of the temperature. Anyhow, the sensor needs to be stable over time only, as the absolute temperature does not play a role, only the initially measured relation between the resistance of the VHP and the temperature as given by the NTC.

Only with such a solid assembly, one will be able to make repeatable measurements of R(T) and T.C.

https://www.eevblog.com/forum/metrology/t-c-measurements-on-precision-resistors/msg464413/#msg464413

Frank

[dietert1]

Not completely true.
I have been using a SR1010 (12x 1KOhm) with its wirewound resistors hanging free from the connector panel and a built-in temperature probe (PT1000) glued onto the inside of the connector panel and was able to characterize its TC to be 1.84 ppm/K with great precision, see here: https://www.eevblog.com/forum/metrology/esi-resitance-standard-sr1010-and-standard-resistor-sr104/msg2719638/#msg2719638.
This is the same SR1010 i brought to the metrology meeting in 2019 where we reproduced its original calibration from 1969 within one ppm or so. In the SR1010 the deviations by self heating that Mr. Frank is talking about, are small due to the size of the wirewound resistors. No solid aluminium blocks.

Regards, Dieter

[Grandchuck]

Finished the reference resistor.  It measured -0.5 ppm/degree C over a range from 16 C to 25 C.  Picture and thermistor characteristic curve attached.  Thanks again for the help!

[leighcorrigall]

Anyone else noticing a massive shortage of precision components as of late? A resistor order I had scheduled this month is now set for November 21st. Some of my other orders are now saying next year. There doesn't seem to be any stock. 

[leighcorrigall]

I just wanted to point out that Vishay sells secondary resistance standards with NIST traceable certificates:

http://www.vishaypg.com/docs/63223/fsr.pdf

They even offer custom resistance values. 

[ManateeMafia]

These are also available from Texas Components

http://webdirect.texascomponents.com/searchresults.asp?cat=1839

[mendip_discovery]

Out of curiosity what oil would be the best for submerging the resistors?

I have a couple of resistors here at home that I planned to make into a crude resistance box, some being low ohm ones. Nothing posh but more of an exercise in learning more about it but making mistakes so I can avoid them in the future. Plus I have several bits of kit at work that does get some use and I would like to make some spares.

[leighcorrigall]

Out of curiosity what oil would be the best for submerging the resistors?

I have a couple of resistors here at home that I planned to make into a crude resistance box, some being low ohm ones. Nothing posh but more of an exercise in learning more about it but making mistakes so I can avoid them in the future. Plus I have several bits of kit at work that does get some use and I would like to make some spares.

Mineral oil is used for the submersion of oil resistors.

Usually, you can find this information in a datasheet or manual for oil bath products. For example, Guildline makes a 5600 Series fluid bath (https://guildline.com/temperature-metrology/environmental-control/air-and-fluid-baths/5600-series-high-precision-variable-temperature-fluid-baths) and the manual (https://guildline.com/media/k2/attachments/OM5600-E1-00.pdf) states the following:


6.1.1. Model 5600 Oil Recommendation:

There are a number of different metrology grade mineral oils, from various manufacturers, that are suitable for a Guildline oil bath. The key specifications are:

Density at 15 Celsius, kg/m cubed = 850
Color, Saybolt = +30
Kinematic Viscosity, Cct@40 Celsius = 12.8

The original oil recommended and used by Guildline for oil baths was Marcol 72 or Marcol 7. However, this oil is no longer manufactured. Guildline now uses Total Nevastane Clear 15 White Mineral Oil. It is manufactured in Canada by Total Lubricants in Laval Quebec.
Other equivalent oils, with their supplier, are:

Petro-Canada - WO10
Sonneborn - Carnation White Mineral Oil
Penreco - Drakeol 7 and Parol 70
Whitaker Oil - 70 White Oil USP

There is also Total Finavestan A80B oil which has a slightly higher viscosity than we recommend but its use should not affect stability of a Guildline oil or fluid bath by more than 1 mK.

---

This is clearly a Canadian company that lists products from my country, there might be other options where you are from.

I hope this helps.

[TiN]

Higher resistance values and setups also can use silicon-based oil, which provides higher resistance insulation and more chemically stable (also more expensive).