Teardown: Standard Resistors

[zlymex]

I've been fascinated by standard resistors of their performance, cannot stop thinking of what's on earth inside.

Before teardown, let's look at some of the important factors of a standard resistor, such as specified by esi SR104:


 - Stability
Specified as ppm/year, It is a long time factor that most important for a standard resistor.
Every Chinese and Japanese resistor has a CLASS, they classify them according to annual drift.

 - Temperature Coefficient, alpha23
This is the slop of the temperature chart at 23 deg C, in ppm/C. Because the temperature chart is curved for most resistors, alpha varies with temperature. That's why it has a subscript specifying at what temperature the alpha is measured. Also, for standard resistors that often used in lab environment, they don't specify or test in large temperature range. Normally 18 to 28 degree C is suffice.

 - Temperature Coefficient, beta
The unit of beta is ppm/C2(per sq deg C), not ppm/C as above spec. The temperature chart can be roughly represented by a quadratic curve, this is especially true for standard resistors where the temperature range is limited. This beta is the coefficient(constant) of the second order equation. The larger the beta, the more the curvature. Most of the metals and resistors have negative beta(parabolic shaped). If a WW made of manganin, the beta is usually -0.6ppm/C2, If made of evanohm the beta usually around -0.028ppm/C2. Foil resistors often have even smaller beta.

 - Power Coefficient
Represent in ppm/Watt. This parameter depends on how heat dissipated and also related to tempco.

 - Adjustment to nominal value, ppm
Its the deviation from nominal value at the time of being made, or may call it initial error, initial accuracy.

One thing also worth noting is that when people measure a standard resistor by a DMM, the standard resistor is superior in most of the cases even the DMM is a 8.5 digits. Most of the imperfection/noise/instability in the test results are probably from the DMM rather than the standard resistor.

ADVANTEST 10k, find no spec, 1 hermetically sealed WW resistor inside.



Chinese BZ3, 0.001 Ohm, class 0.01%, manganin.




Remarks: This is one of the nine BZ3 resistor set, 0.001 Ohm, max current 33A, everything is heavy from binding post, thick manganin stripe. There is a small compensation resistor in parallel with the main.


esi SR104, 10k, stability 0.5ppm/a. Oil filled, hermetically sealed evanohm WW in a stainless steel container.





What can I say, the best 10k resistor ever made. Pity that no further teardown.

[zlymex]

esi SR1, 100 ohm, 50ppm/year


Normal wire on mica, the favorite way of esi.


Fluke 742A, 10k







Stability 4ppm/year, tempco is very small. Four 39992 Ohm hermetically sealed evanohm WW resistors in parallel, a 32 ohm adjust resistor is in series with one before parallel. Fluke is probably the largest hermetic WW manufacturer.


Fluke 742A, 1 Ohm



Stability specified as 8ppm/year. Twenty 20.01 Ohm hermetically sealed evanohm WW resistors in parallel, parallel another small adjust resistor of about 2k.
The inside photo was by Leicamr of 38hot.net

[zlymex]

IET SRL 1 Ohm
Same specification as Fluke 742A 1 Ohm but different internal construction.



The small resistor on the top is probably the compensation resistor.

IET SRX, 1Ok
I bought this one from IET 11.5 years ago, specified as 10ppm/year, 1ppm/C. One hermetically sealed WW resistor with some compensation/adjustment





MI 9331, 100 Ohm, Specified as 5ppm/year, 0.1ppm/C




Transmille RS3000


http://www.transmille.com/references-and-standards/precision-references-and-standards/3000rs

Although this is not an actual teardown but I know what's inside.
When this UK standard was first released in 2009, I saw something interesting on their website that must be mistakenly put by someone the manuscript of the resistors they used in the standard. A few days later it's gone but I saved it.



Another interesting thing about the script is that they described the tempco of  VSC232Z as 0.05ppm/C, apparently mislead by Visahy's datasheet. Later, they modified this to 0.5ppm/C

[zlymex]

Wekomm RS9010A, 10k




More photo and info see Dave's blog #834 and xdevs.com
Specified as 1ppm/year, 0.3ppm/C, one hermetically sealed Vishay foil resistor


I must say this is not up to German standard as the actual tempco of -0.47ppm/C is larger than specified. Wekomm may have misinterpret the datasheet from Vishay like most of us did, and they didn't throw away or compensate ones with large tempco.
Also they are contradictory on those max voltage and max current specs. How can you apply 10mA max current on a 10k without exceed the 20V max voltage. Well, they may say they are on 'which is smaller' basis, but why 10mA max for 10k in the first place?
Another thing is that the resistor is off by as much as 7.72ppm, but the spec says it should be within 2ppm.

Inside VHA518-7, there are seven foil resistor elements.



Yokogawa, 1k, class 50ppm, not hermetically sealed






Lastly, let's teardown some hermetically sealed WW and see how they are made.






Apart from usual bobbin/teflon insulation etc, the special thing about the hermetic resistor is the seal. They have those pre-made glass/ceramic disk with a small tube in the middle and thin layer of metal ring outside to be later soldered to bass tube case.
Although brass has different temp expansion co-efficient with glass, it's flexible and will expand and extract with the hard glass as temperature vary. If the casing is thick, they cut the edge part to make thinner. The inner metal on the hand must be Kovar which expand the same with glass, otherwise, the not-so-easy-compressed 'rod' will soon destroy the seal or even break the glass especially in high temperature when soldering the leads. But, Kovar is notorious for its large thermal EMF that should not be used in precision WW. They solved this dilemma by first implant a Kovar tube(instead of Kovar wire) that bond with glass well, and pass the copper lead through the hole and fill the gap by soldering. Solder is soft that act as cushion if temperature varies and still keep the seal.

The end of the teardown. Sorry for taking too long and split, I had difficulty in uploading images that often stopped in the middle.

[Vgkid]

Looking forward to the rest. I have a 1k SR1 resistor(I tore it down too).

[TiN]

Very nice summary Hope will can update it as it evolve.

[quarks]

Great post

here are some pictures from my Leeds & Northrup 4210 "Thomas Type One Ohm Standard" resistor.
This was advertised by L&N as the "Prime standard for use in maintaining value of the ohm".

[quantumvolt]

Very interesting.

I would like to know how the 4 resistors and the trimpot on the compensation board of IET SRX, 1Ok are connected and what values they have (at least for me, they cannot all be seen clearly). Maybe it would be possible to infer if it is all resistance value adjustment, or there also are some temperature coefficient elimination (reduction).

Thanks for a very informative post.

[nidlaX]

Wow! Nice stuff you have access to!

[daqq]

Thanks a lot! Great images!

[tszaboo]

Very good stuff.

Another interesting thing about the script is that they described the tempco of  VSC232Z as 0.05ppm/C, apparently mislead by Visahy's datasheet. Later, they modified this to 0.5ppm/C

Yes, the VishayPG datasheets are very very misleading. They claim insane typical drift on the top of the page, and then you scroll down, and see that while the typical is 0.0005 PPM the actual worst case drift is 0.0005+/-2 ppm. And since it is in the datasheet, it is not even a lie.

If you want to see something stupid, order a few thousand of their precision resistor. Once I saw that they delivered a big order, where the SMD resistors were in trays, not reels. I guess the bending would destroy the accuracy. But that is not the funny part. The tray was tiny, containing 25 resistor. Way to go, production ready!

[manganin]

The Chinese standard resistor is a direct copy of the Soviet version down to the smallest details.

Of course the Soviet design wasn't their own either...

Chinese BZ3, 0.001 Ohm, class 0.01%, manganin.

[zlymex]

The Chinese standard resistor is a direct copy of the Soviet version down to the smallest details.

Of course the Soviet design wasn't their own either...

You are right, I have a few standards of the Soviet as well.

[zlymex]

Very good stuff.

Another interesting thing about the script is that they described the tempco of  VSC232Z as 0.05ppm/C, apparently mislead by Visahy's datasheet. Later, they modified this to 0.5ppm/C

Yes, the VishayPG datasheets are very very misleading. They claim insane typical drift on the top of the page, and then you scroll down, and see that while the typical is 0.0005 PPM the actual worst case drift is 0.0005+/-2 ppm. And since it is in the datasheet, it is not even a lie.

If you want to see something stupid, order a few thousand of their precision resistor. Once I saw that they delivered a big order, where the SMD resistors were in trays, not reels. I guess the bending would destroy the accuracy. But that is not the funny part. The tray was tiny, containing 25 resistor. Way to go, production ready!

I knew it's true, they have mislead me before. Although didn't  order a few thousand, but a few hundred is more than enough. When my first order of 50 VHP101 came, it arrived in one plastic bag.

[zlymex]

Very interesting.

I would like to know how the 4 resistors and the trimpot on the compensation board of IET SRX, 1Ok are connected and what values they have (at least for me, they cannot all be seen clearly). Maybe it would be possible to infer if it is all resistance value adjustment, or there also are some temperature coefficient elimination (reduction).

Thanks for a very informative post.

I've never care about the adjustment part before since I think it is the usual way of trim a resistor for exact value like those in Fluke 720A.
Since you ask, I just teardown it again and it is. There is no tempco related compensation what so ever. If you need something for temco compensation, I myself developed a way of doing that: compensate the alpha tempco and trim the value at the same time. Also, there are ways to compensate the beta as well.
I've update it with the circuit above, there is a 39.2 Ohm resistor on the board but shorted.

[manganin]

IET SRL 1 Ohm. Same specification as Fluke 742A 1 Ohm but different internal construction.

The internal construction is actually the same. Multiple resistors connected in parallel with the copper equalizing plates. But in the IET version the whole package was resin potted which allows to use much cheaper non-hermetic resistors. The solution is not perfect but another layer of protection seems to make the humidity time constant long enough.

The "wings" in the equalizing plates are a clever idea, because they allow soldering the adjustment resistor after the potting.

[zlymex]

Great post

here are some pictures from my Leeds & Northrup 4210 "Thomas Type One Ohm Standard" resistor.
This was advertised by L&N as the "Prime standard for use in maintaining value of the ohm".

Thanks quarks, 4210 is a great resistor. Is that on the bottom red paint? Or there is solder underneath for the double wall to hermetically seal?

[quarks]

Although didn't  order a few thousand, but a few hundred is more than enough. When my first order of 50 VHP101 came, it arrived in one plastic bag.

I did order even less than you and received individually packaged resistors, some even in plastic cases (see pic.)

[zlymex]

IET SRL 1 Ohm. Same specification as Fluke 742A 1 Ohm but different internal construction.

The internal construction is actually the same. Multiple resistors connected in parallel with the copper equalizing plates. But in the IET version the whole package was resin potted which allows to use much cheaper non-hermetic resistors. The solution is not perfect but another layer of protection seems to make the humidity time constant long enough.

The "wings" in the equalizing plates are a clever idea, because they allow soldering the adjustment resistor after the potting.

Great info. I'm curiosity about why the spec of the old IET SRL series are exactly the same as Fluke 742A.

[manganin]

Is that on the bottom red paint? Or there is solder underneath for the double wall to hermetically seal?

Solder which is just "tamper sealed" by the red paint.

The construction is very simple. The edges of the inner cylinder are bent so that it tighly fits inside the outer cylinder. The coil is wound on the inner cylinder. After the heat treatment the inner cylinder is slided inside the outer one and soldered together at both ends.

[zlymex]

Is that on the bottom red paint? Or there is solder underneath for the double wall to hermetically seal?

Solder which is just "tamper sealed" by the red paint.

The construction is very simple. The edges of the inner cylinder are bent so that it tighly fits inside the outer cylinder. The coil is wound on the inner cylinder. After the heat treatment the inner cylinder is slided inside the outer one and soldered together at both ends.

Got it. I'm asking this because I saw another teardown of a Cambridge standard where there is those red paint as well, but there is no solder inside the paint.

[manganin]

The construction is very simple. The edges of the inner cylinder are bent so that it tighly fits inside the outer cylinder. The coil is wound on the inner cylinder. After the heat treatment the inner cylinder is slided inside the outer one and soldered together at both ends.

By the way, the Soviet 1 ohm (and higher) standard resistors share the same contruction. Except that the the edge of the outer cylinder bended at the bottom and the inner cylinder at the top. The Chinese version is probably similar.

Unfortunately the stability is not even close to the Thomas. The mechanical construction was easy to copy, but not the art of the heat treatment.

I saw another teardown of a Cambridge standard where there is those red paint as well, but there is no solder inside the paint.

The Cambridge was also sealed by soldering but there is a separate bottom plate between the inner and outer cylinders. Please see the picture attached.

[zlymex]

The construction is very simple. The edges of the inner cylinder are bent so that it tighly fits inside the outer cylinder. The coil is wound on the inner cylinder. After the heat treatment the inner cylinder is slided inside the outer one and soldered together at both ends.

By the way, the Soviet 1 ohm (and higher) standard resistors share the same contruction. Except that the the edge of the outer cylinder bended at the bottom and the inner cylinder at the top. The Chinese version is probably similar.

Unfortunately the stability is not even close to the Thomas. The mechanical construction was easy to copy, but not the art of the heat treatment.

I saw another teardown of a Cambridge standard where there is those red paint as well, but there is no solder inside the paint.

The Cambridge was also sealed by soldering but there is a separate bottom plate between the inner and outer cylinders. Please see the picture attached.

I also teardown the Chinese 1 Ohm, top and bottom all soldered. However, there were many factors that they did not do it correctly as I have pointed out at 38hot times ago, so I did not include the teardown here. As for the average aging of those Chinese resistors, it varies. Here are the distribution of 24 samples, some are good but others not.


The bottom of the Cambridge is soldered, what about the top?

[manganin]

The bottom of the Cambridge is soldered, what about the top?

The top is soldered. But the connection bar seals are far from hermetic. The Cambridge type is oil filled, not air, so there is also an oil screw at the top plate.

But the ceramic feedthrough of the Soviet type is not perfect either. Many resistors which have spent decades in an oil bath have changed from air filled to oil filled...

As for the average aging of those Chinese resistors, it varies. Here are the distribution of 24 samples, some are good but others not.

Very interesting data of the Chinese BZ3. Large variation (poor quality control) is typical for the Soviet type too.

How did you eliminate the temperature coefficient in your measurements? The TC of the Soviet type is so high that an oil bath is a necessity. And the hysteresis is remarkable which can lead to wrong assumptions if the resistors are not stored in constant temperature between the measurements.

[doktor pyta]

Good job, zlymex, Thanks!

[zlymex]

How did you eliminate the temperature coefficient in your measurements? The TC of the Soviet type is so high that an oil bath is a necessity. And the hysteresis is remarkable which can lead to wrong assumptions if the resistors are not stored in constant temperature between the measurements.

I put a mercury thermometer to measure  the temperature and make corrections based on the alpha and beta tempco which are on the certificate.
Those standards was manufactured 25 to 40 years ago, need not measure the value very precisely to determine the average annual drift.
Most of beta are at around -0.6ppm/C2,which are similar for all manganin resistors, right? In China/Soviet, reference temperature is 20 deg C, 3 deg lower than 23C. Since most alpha20 is positive, if changed to alpha23, they will go done by 1.8ppm/C and appear smaller, magic! That is to say, alpha20=5ppm/C is the same as alpha23=3.2ppm/C when beta=-0.6ppm/C2.  I measured some Chinese BZ3 and they show little hysteresis.

[Vgkid]

Im intrigued to see how the trim the srx to 10k, as the vishay is 9.995k. I will need to look at the pics,for the possible series resistor.

[quantumvolt]

Very interesting.

I would like to know how the 4 resistors and the trimpot on the compensation board of IET SRX, 1Ok are connected and what values they have (at least for me, they cannot all be seen clearly). Maybe it would be possible to infer if it is all resistance value adjustment, or there also are some temperature coefficient elimination (reduction).

Thanks for a very informative post.

I've never care about the adjustment part before since I think it is the usual way of trim a resistor for exact value like those in Fluke 720A.
Since you ask, I just teardown it again and it is. There is no tempco related compensation what so ever. If you need something for temco compensation, I myself developed a way of doing that: compensate the alpha tempco and trim the value at the same time. Also, there are ways to compensate the beta as well.
I've update it with the circuit above, there is a 39.2 Ohm resistor on the board but shorted.

Thank you very much.

Again - a very interesting thread.

[Macbeth]

Great thread. On the opposite end of the scale this is my standard resistor 



A military version of the S102 I think.

[splin]

I knew it's true, they have mislead me before. Although didn't  order a few thousand, but a few hundred is more than enough. When my first order of 50 VHP101 came, it arrived in one plastic bag.

Ohh please do tell - what's your experience with respect to the 'typical' 2ppm/6 years claim? Even a few weeks worth of data could be very interesting.

[zlymex]

Im intrigued to see how the trim the srx to 10k, as the vishay is 9.995k. I will need to look at the pics,for the possible series resistor.

I updated to include the schematic of the trim part.

I knew it's true, they have mislead me before. Although didn't  order a few thousand, but a few hundred is more than enough. When my first order of 50 VHP101 came, it arrived in one plastic bag.

Ohh please do tell - what's your experience with respect to the 'typical' 2ppm/6 years claim? Even a few weeks worth of data could be very interesting.

I didn't test the aging in a systematic way for all, but I made a standard resistor out of them in 2008 and the result is superb. Have a look at photos below, the value is virtually unchanged in five and half years time.

[quarks]

Hello zlymex,

it is really a pleasure to see, read and learn from your great experience and knowledge!
Your DIY 10k seems to be extremly stable.
I wonder which combination of resistors (parallel, series or comb.) gives the very best resulting TC and stability.
Also I am interest to know how to meassure and use the alpha and beta values.
Thanks a lot for sharing

Bye
quarks

[zlymex]

Hello zlymex,

it is really a pleasure to see, read and learn from your great experience and knowledge!
Your DIY 10k seems to be extremly stable.
I wonder which combination of resistors (parallel, series or comb.) gives the very best resulting TC and stability.
Also I am interest to know how to meassure and use the alpha and beta values.
Thanks a lot for sharing

Bye
quarks

Thanks quarks,
I can open a new thread for DIY resistors - I made some before and this one is probably my best.
This particular one, made from selected 4 Vishay VHP101 hermetic resistors from a batch of 50, 9.999k, 0.01%.
I first tested those resistors for values and tempco for sometimes and type in the data in Excel, selected 4 best matched base on alpha and beta, then its not necessary to use compensation late, though I do have a spread-sheet way of compensation for tempco and value, which have been used in other of my DIYs.
Connect 2 in series first, and then parallel. This formed something like a bridge, those orange wires are for 'null detector' output.
As far as the performance is concerned, parallel first or series first is exactly the same, provided they are all equally weighted(same power). I believe that 1k to 10k range is the best for Vishay foils resistor as the value is concerned. I may have ordered the value at 2.499k but that will eliminate the option for single-resistor use at 10k.

I didn't design, select or control the stability, it mainly depends on the resistors themselves, just lucky to be this stable.
I did some small range temperature cycle before hand of -18 deg C in a fridge and about 80 deg C hot water bath, just for mental comfort.
Made sure not to introduce any unstable elements into it by some considerations like long-lead, less stress. When soldering use thick tweezers with heat conducting paste to prevent it reaching the body.
Well, the leads is copper, with about +4200ppm/K tempco, had been taken into account during selection and measurement.
No heatsink, thermal equalizer or grease were used because I was thinking those may do more harm than good for stability, and because it only dissipates 10mW, it can be handled well with four elements.
I do use Teflon board and dual case, the inner case is electrically isolated and also is the guard output.
Those small resistor along side were German brand, Czeck made Vitrohm of 0.1%, and also very small in value, in series with VHP101 to facilitate the adjustment to final value. I ordered those VHP101 at slightly less than 10k for that purpose. I do sensitive analysis for each components to make sure they don't affect the end results in an unexpected way.
The binding posts was taken out of my Dataproof scanner(164A), very low thermal EMF according to them.
This particular standard have been traveled in China for about two years among my Ohm-nuts friends and return to me recently.

For tempco measurement of alpha and beta, put them in a car-fridge and run a temperature cycle, record value and temperature. The measurement is always done by my own scanner/switch, where I connect my two SR104 as a reference, data was feed to 3458A direct or indirect(by a Warshawsky bridge), and against one of my SR104, to eliminate the drift of 3458A.
Then, plot the temperature-value scatter diagram in Excel like this:


This is an example, but is the actual plot of a BZ3 1k standard resistor, also see the attached spread sheet for how to plot and calculate.
Then make a trend line of second-order polynomial(red line, by right click the data), this is the least-squares regression of the best fit of the scattered data.
Also, there is this 'display equation' and 'R square' option of this trend line in the form of
y = c + bx + ax^2
where y is the resistor value and x is the temperature
the R^2 is the measurement of the fit quality and should be close to 1. It is also the indication for hysteresis.
beta is immediately available: beta = a/c
actually alpha0=b/c but this alpha is at zero degree C, we normally use alpha at 23 degree C, so a little conversion has to be done in spread sheet(enclosed) to transform the equation to the form of
y = c1 + b1*(x-23) + a*(x-23)^2
and finally alpha23=b1/c1
 
In order to make the good result of alpha and beta, use slow temperature cycle to reduce hysteresis, use a good temperature sensor close mounted with the resistor. I use DS18B20 for sensor(part of the data collection system) and checked it by ice-water and mercury thermal meter of the Chinese 2nd grade(to 0.1 degree C).
SR104 is regularly calibrated. Recent is done at Hong Kong HKAS two months ago. I also monitor the thermometer and made small corrections if necessary. As for the measurement of alpha and beta, absolute value of the resistor reference is not critical at all.
I measured alpha/beta of one of my SR104 this way and the procedure/result were on 38hot.net
http://bbs.38hot.net/thread-1352-1-1.html
(Sorry its in Chinese, may be you can use google translate)

To use the alpha and beta, just measure the temperature and use the formula y = c1 + b1*(x-23) + a*(x-23)^2 to obtain the exact resistance at that temperature. Or better make a temperature chart(like attached) to get the resistance quickly.
The tempr chart in the inner case of SR104 is not good because it does not show the deviation explicitly and is error prone when add it up.

[altaic]

What a great look at resistance standards! I'm curious-- how did you select the resistors for the one you made? Any special method (Wheatstone bridge or something), or just use a precision ohmmeter?

[quarks]

Hello zlymex,

now I will sure need some time to work through all your very good information and try to make use of it.
Many thanks again and please start a new threat on DIY resistors.

Bye
quarks

[zlymex]

What a great look at resistance standards! I'm curious-- how did you select the resistors for the one you made? Any special method (Wheatstone bridge or something), or just use a precision ohmmeter?

I tested all the 50 VHP101 for alpha and beta according to above method.
I also make rough measurement for values by 3458A but that is not very important, they will be adjust at final stage anyway.
The selection was done on paper(spread sheet) for matched alpha and beta - just chose those that added up to zero.
The final adjustment is by use of different small value resistors in series and in parallel, value aided by spread sheet.
The final measurement of value that the adjustment is based on was performed by this:

Center part is the Warshawsky bridge specifically for 10k, capable of compare two 10k standard resistors at 0.05ppm uncertainty.
3458A here is merely as a sensitive voltmeter/null detector, no calibration needed. I use it because it hook up to the system most of the time.
The absolute value of the measurement depend on the reference resistor which is a SR104 in my case.
I use automatic substitute technique by employing a 4-pole double throw switch:
https://www.eevblog.com/forum/projects/diy-low-thermal-emf-switchscanner-for-comparisons-of-voltage-and-resistor-stand/msg610755/
 

[zlymex]

Hello zlymex,

now I will sure need some time to work through all your very good information and try to make use of it.
Many thanks again and please start a new threat on DIY resistors.

Bye
quarks

Sure, may be next weekend.
The reason I bought VHP101 is because of 3458A. Although I didn't find a lot of precision parts when I opened the case the day it arrived, I did find two things, one is VHP101, the other is LTZ1000A, and that start my DIY journey.
At that time, Vishay specified VHP101 as 2ppm/10years.

[altaic]

I use automatic substitute technique by employing a 4-pole double throw switch:
https://www.eevblog.com/forum/projects/diy-low-thermal-emf-switchscanner-for-comparisons-of-voltage-and-resistor-stand/msg610755/

Your setup is similar to one I came up with, but what is the automatic substitute technique you mentioned? Is this a way to attach lots of resistors to a test rig, and have it test all permutations of resistors to find a set which matches your desired specification? You mentioned measuring the alpha and beta values and using a spreadsheet to pair resistors, so I'm confused about the automatic substitute w/ the 4-pole double throw switch.

Hello zlymex,

now I will sure need some time to work through all your very good information and try to make use of it.
Many thanks again and please start a new threat on DIY resistors.

Bye
quarks

Sure, may be next weekend.
The reason I bought VHP101 is because of 3458A. Although I didn't find a lot of precision parts when I opened the case the day it arrived, I did find two things, one is VHP101, the other is LTZ1000A, and that start my DIY journey.
At that time, Vishay specified VHP101 as 2ppm/10years.

That would be great! I'm looking to bootstrap some metrology equipment, hopefully to high precision. I'm hoping to figure out how one can get high precision out of low precision equipment, though I suppose one always has to have some sort of standard-- voltage standard, or some such. Some day, room temperature quantum hall devices (graphene looks promising) will be easily available, giving cheap resistance standards to 1ppb, but currently we need some sort of cryogenic cooler (there are tabletop ones [https://en.wikipedia.org/wiki/Cryocooler] which aren't too expensive, all things considered).

Anyway, more on topic, do you know of any good references for types of resistors which can be used to balance the tempo (or if one must just tediously measure it for each one), and techniques for aging resistors? Seems like a lot of resistors only have to be aged for 10-20 days to get them to be fairly level.

[TiN]

Some time ago I got pair of Guildline 9330 resistors, with one being broken (readings jump all over the place).



It's very small, compared to most of other resistors of this type, designed for oil bath operation



Inside is just spool of wire, no bifilar winding, no support, just flying in the space. 



Problem is with thinner wire spool, which is most of 100K resistance.



Schematics of resistor network:



Hi-res photos and test results here in the article.

And since it's just a wire element, I'm not sure of repair approach on this one, except replacing wire with new manganin one of correct resistance. Tried to clean wire with IPA bath, did not help.

[Vgkid]

^^^
Seeing your post, I should tear down my Otto Wolff 10k resistor.
Time to either get medieval, or break out the blowtorch.

[zlymex]

Your setup is similar to one I came up with, but what is the automatic substitute technique you mentioned? Is this a way to attach lots of resistors to a test rig, and have it test all permutations of resistors to find a set which matches your desired specification? You mentioned measuring the alpha and beta values and using a spreadsheet to pair resistors, so I'm confused about the automatic substitute w/ the 4-pole double throw switch.

Well, substitution is a more general term than a specific technique that close related to the term 'transfer'. It may be used together with many existing measuring methods, whether on purpose or unintentional, to achieve a better result with less uncertainty. For example:
1. you have a standard resistor with a calculated deviation of +1.5ppm
2. you measure this standard with a specific method(4-wire), reading is +12.0ppm (which means 10000.120 Ohm if the nominal value is 10k)
3. you substitute your standard with the unknown resistor, either manually by detaching those 4 wires to the standard and attach them to the unknown, or automatically by a scanner with 4-poles to switch to the unknown from the standard for you.
4. then you measure you unknown resistor, with other setup unchanged, reading is +8.0ppm   
5. you calculate the unknows as +1.5ppm +8.0ppm -12.0ppm = -2.5ppm. That is to say, you unknown will be 9999.975 Ohm if the nominal value is 10k, and the uncertainty of you unknown would be the uncertainty of your standard 'plus' the uncertainty of your substitution measurement, and the later depends on what kind of specific method and the equipment used.
6. The above calculation is approximate, I have replaced the multiplications and divisions with additions and subtractions, but it will be accurate enough if all those ppms are small. This means that all the standards, unknowns, the equipment etc should be all close to nominal value in order for this simple calculation to be valid, and that's one of the reasons why we always have this 'close to nominal' desire, we can safely ignore those second order variables.

By meaning of 'a calculated deviation of +1.5ppm', this is how it calculated:
1. your standard resistor is calibrated 6 months ago with deviation of +1.4ppm @23 deg C
2. your standard resistor has a predictable aging rate of +0.4ppm per 12 months
3. your standard resistor has a tempco of +0.5ppm and you do your measurement @ 18 deg C, this will give +0.5/(18-23)=-0.1ppm correction
(for simplicity, I just ignored the beta tempco)
4, then you can calculate the deviation as +1.1ppm +0.4ppm/12*6 -0.1ppm = 1.5ppm

Substitution may be performed in one to one manner as exampled above, or may be performed in one to many manner as I often do by using a 4-pole 17-position scanner.

I measure those resistors first(for value, alpha, beta), then I match them in spread sheet.

  I'm looking to bootstrap some metrology equipment, hopefully to high precision. I'm hoping to figure out how one can get high precision out of low precision equipment,

That's exactly what I'm talking about. I, as an amateur, and I guess many of us here are, keen on to find cheap ways to measures resistance to high degree of uncertainty. I think I've got something, I may start a new thread on this and analysis the uncertainty associated.

Anyway, more on topic, do you know of any good references for types of resistors which can be used to balance the tempo (or if one must just tediously measure it for each one), and techniques for aging resistors? Seems like a lot of resistors only have to be aged for 10-20 days to get them to be fairly level.

Balance the temco? I use a copper resistor if the tempco of the main resistor is negative, as describe in my new thread Yesterday:
https://www.eevblog.com/forum/metrology/spread-sheet-aided-design-of-compensation-for-7v-to-10v-step-up-resistor-set/
If the tempco is positive, then it will be troublesome to find the suitable negatively going balancer. I bought some NTC resistors for this purpose but have not tried them out yet, I was worrying about the stability and none linear curve, although the curve is concave and is perfect to balance the beta of most of WW.

As for techniques for aging resistors, I believe there are many discussions already on this forum. Personally, I rely on the design/construction/materiel of the resistor rather than accelerated aging by myself.

To summarize, if the tempco not balanced but known, and there exist aging but predictable, then you can deduct those small values by calculation, therefore no worry about balancing or adjustment.

[zlymex]

Some time ago I got pair of Guildline 9330 resistors, with one being broken (readings jump all over the place).
........

Superb photo!
Is that 'no support' thing only for the small WW? or also for part of the 99.998185k wire?
Any idea for the maker of those two foil-liked hermetic?

^^^
Seeing your post, I should tear down my Otto Wolff 10k resistor.
Time to either get medieval, or break out the blowtorch.

That would be very nice. But hopefully only to the degree of able to reassemble in original precision. I've only seen some photo of Otto Wolff shunt.

[Vgkid]

^^^
Seeing your post, I should tear down my Otto Wolff 10k resistor.
Time to either get medieval, or break out the blowtorch.

That would be very nice. But hopefully only to the degree of able to reassemble in original precision. I've only seen some photo of Otto Wolff shunt.

Unfortunately it was dropped, so it never went back to its original stability. Wildly drifty,even in a short span.

[zlymex]

^^^
Seeing your post, I should tear down my Otto Wolff 10k resistor.
Time to either get medieval, or break out the blowtorch.

That would be very nice. But hopefully only to the degree of able to reassemble in original precision. I've only seen some photo of Otto Wolff shunt.

Unfortunately it was dropped, so it never went back to its original stability. Wildly drifty,even in a short span.

That‘s sad news, but not too bad from teardown point of view, or may be repairable.

[TiN]

Is that 'no support' thing only for the small WW? or also for part of the 99.998185k wire?
Any idea for the maker of those two foil-liked hermetic?

More meant for wire (both 99K and 1K). It's just in there, not fixed to anything.
Hermetics more or less sturdy hanging with their thick legs. They have no labels except written resistance on them, and can is bit taller than VHPs. Probably Guildline inhouse resistors?

[zlymex]

More meant for wire (both 99K and 1K). It's just in there, not fixed to anything.
Hermetics more or less sturdy hanging with their thick legs. They have no labels except written resistance on them, and can is bit taller than VHPs. Probably Guildline inhouse resistors?

Well, I have some photo of AE and TDK hermertics with taller cans, inside, they are plastic foils, not bare chips like VHPs.

[manganin]

Some time ago I got pair of Guildline 9330 resistors, with one being broken (readings jump all over the place).

After resin removed, resistive EVANOHM® or Manganin wired elements were revealed. Shiny yellow coil with thinner wire is 99x K? element, and smaller dark wire coil is 1.1K? element.

The main resistor wire is manganin. The smaller one is another alloy with much higher TC. The 9330 relies on compensation instead of adjusting the TC of the main resistor by annealing process. Measuring the TC of the 1K wire would help to deduce the type of it.

The bulk metal foil resistors are used for fine adjustment after the container has been sealed. The 9330 resistors that I have seen had only one BMF adjustment resistor.

And since it's just a wire element, I'm not sure of repair approach on this one, except replacing wire with new manganin one of correct resistance.

That will ruin the temperature compensation made for that specific wire.

Sounds like a bad solder joint or a short between the adjacent turns (or the metal container). The manganin wire itself doesn't behave like that. Carefully rewinding the wire on a large piece of cardboard (mica card resistor style so that the turns don't touch each other) could help to make it stable again. That also helps to locate the possibly damaged wire or insulation using a microscope.

Inside is just spool of wire, no bifilar winding, no support, just flying in the space.

That allows almost free thermal expansion, which much reduces the hysteresis and improves the stability in varying temperature environment. The drawback is that a mechanical shock can cause permanent change.

The bifilar construction would have been easy to do by simply double folding the wire before the winding. Are you sure it wasn't made that way?

[Vgkid]

That‘s sad news, but not too bad from teardown point of view, or may be repairable.

Was tempted to not tear it down,since it was kind of stable last night 20ppm p-p over 8 degree delta. Slid the outer case off to see how deep the dent was, and it dropped 9 ohms .
What offset voltages should one get for these types of resistors?  I had 8uV last night, and about that, after cleaning.
I will tear it down.

[Vgkid]

My Otto Wolff 10K resistor, Heat didn't even melt the solder, needed force.
I could not determine what caused the resistance value to changeby 10 ohms...
The top of the unit.

Wires joining the copper terminal block to the resistor. They appear to be silver. I sanded off the paint.

Getting medieval. Used a chisel, and a large screwdriver to flex the entire cylinder(breaking the solder seal).

Resistor

close ups

This almost looks like a metal wire, under the right light.

Currently hooked up to my 3456A, and will run it over night. These resistors really do not like offset compensated ohms mode(pulse DC), my others resistors don't mind it.

[manganin]

Heat didn't even melt the solder, needed force.

The soldered seam of the container can be opened without brute force by using a heated wire. Sand paper needed first to reveal the bare metal surface.

[manganin]

Information about the Otto Wolff NG series:

The standard resistor NG 100 consists of a wire coil which is hermetically sealed between two brass tubes. A nickel-plated brass tube of diameter of 102mm and height of 170mm serves as a protecting device which is locked at the top by an ebonite cover. The sealed coil is connected to the outside terminal through glass. The ebonite cover as well as the terminals on it are of exactly the same design as those of the N 100. The escape of the heat being poorer with the model NG 100, the admissible load is lower than the case of the model N100 and corresponds to the values of the N 85. The model NG 100-2a has an accurary of 0.005 percent; the other models have an accurary of 0.002 percent. Loading in air must be restricted to 0.1 Watt.

[Vgkid]

@ Manganin how did you find this information, I looked a few years ago, and found nothing. Were these oil filled, or just air filled? When I opened it up, it smelled of stale gasoline, and noticed that the internal walls of the brass conainer appeared to have corrosion on them.

[Vgkid]

For those curious, if I killed it. Nope
Ran it on the 3456 over a day, didn't note the temps.
Min: 10.00055k, max:10.0076k, avg: 10.00068, var 4.21988E^-3.
i need to look into a cheap datalogging thermometer.

[manganin]

Manganin how did you find this information, I looked a few years ago, and found nothing.

I have a full Otto Wolff binder somewhere, but all I could find now was a poor one page photocopy which had that short description.

Were these oil filled, or just air filled? When I opened it up, it smelled of stale gasoline, and noticed that the internal walls of the brass conainer appeared to have corrosion on them.

Air filled. But a few decades in an oil bath...

The sealed version is quite rare, probably because it was more expensive and didn't offer much advantage over the open version. The brass cylinder was less rigid than the ceramic and the closed container actually amplified rather than attenuated the pressure coefficient. Same thing with the temperature hysteresis.

I took a photo of one of my N 100 resistors (100 ohm) wound on a ceramic cylinder. No hard laquer, but elastic wax instead which doesn't stress the wire as much.

[quarks]

does anyone know what is inside this "Zero Ohm Standard"





http://www.ohm-labs.com/resistance-standards/100-0-standard.html

and explain the
    BALANCED TETRAJUNCTION DESIGN
    A TRUE ZERO-OHM RESISTOR
   

[manganin]

does anyone know what is inside this "Zero Ohm Standard"

and explain the BALANCED TETRAJUNCTION DESIGN A TRUE ZERO-OHM RESISTOR

http://www.google.com/patents/US3252091

[zlymex]

does anyone know what is inside this "Zero Ohm Standard"





http://www.ohm-labs.com/resistance-standards/100-0-standard.html

and explain the
    BALANCED TETRAJUNCTION DESIGN
    A TRUE ZERO-OHM RESISTOR
   

It's better than those in patent US3252091, all of which are not fully symmetrical. That is, when any two terminals passing thru a current, the other two might have none-zero potential.
A regular tetrahedron will do the trick but is difficult to implement. Another one I know is attached.
 

[manganin]

It's better than those in patent US3252091, all of which are not fully symmetrical. That is, when any two terminals passing thru a current, the other two might have none-zero potential. A regular tetrahedron will do the trick but is difficult to implement. Another one I know is attached.

In real life the simplified versions developed by ESI provide excellent performance. With practical current levels and reasonable copper thickness the connector thermal EMFs are a bigger problem than the error caused by the current distribution. And the shape can be fine tuned afterwards based on the measurements to compensate the limited metalwork accurary.

[quarks]

http://www.google.com/patents/US3252091

Looks very much like inside my ESI SR1010s

[manganin]

Looks very much like inside my ESI SR1010s

And if I remember correctly, there is even a brief junction error analysis in the SR1010 manual.

[babysitter]

Yes, there is. Giving approx. numbers for the junctions, iirc in the 50-150µOhm range depending on the used "ports".

[zlymex]

It's better than those in patent US3252091, all of which are not fully symmetrical. That is, when any two terminals passing thru a current, the other two might have none-zero potential. A regular tetrahedron will do the trick but is difficult to implement. Another one I know is attached.

In real life the simplified versions developed by ESI provide excellent performance. With practical current levels and reasonable copper thickness the connector thermal EMFs are a bigger problem than the error caused by the current distribution. And the shape can be fine tuned afterwards based on the measurements to compensate the limited metalwork accurary.

I've never in doubt the performance of those ESI simplified versions within their Hamons. It is the universal zero ohm that people may connect 4 terminals in anyway which requires the use of full symmetrical version.

A 2D fully symmetrical version(electrical sense) can be made like this.

[quarks]

@zlymex,

the 4 hole symmetry makes very much sense, but I wonder why they made all posts inline and not just use the same pattern outside of the box. At least that would avoid cabeling inside.

Also is there a reason to go for the triangle shape of the plate?
Im asking because a round disc with the same 4 hole symmetry pattern would be my first idea to build it

Thanks
quarks

[zlymex]

@zlymex,

the 4 hole symmetry makes very much sense, but I wonder why they made all posts inline and not just use the same pattern outside of the box. At least that would avoid cabeling inside.

Also is there a reason to go for the triangle shape of the plate?
Im asking because a round disc with the same 4 hole symmetry pattern would be my first idea to build it

Thanks
quarks

I didn't quite get the question, did you mean why they use a metal box?

It just because I draw the triangle easy, it's thinner than a round disc and will save material. The thinnest would be down to just 4 wires. Be careful though, a cross won't do.

[quarks]

I didn't quite get the question, did you mean why they use a metal box?

No my question was not about the metal box but about the arrangement of the connectors.

It just because I draw the triangle easy, it's thinner than a round disc and will save material.

if the triangle plate is not for a technical reason, I think with my limited lathe/mill skills, it should be easier to use round stock and achieve near perfect symmetry (also there are no plates in CuTe but only round stock afaik).

[Kleinstein]

The triangle shape should be a little less sensitive on how well the wire is connected / soldered. It could also make soldering more easy, as heat is not conducted away so fast.
Still I see the advantages in machining for the round form.

[quarks]

The triangle shape should be a little less sensitive on how well the wire is connected / soldered. It could also make soldering more easy, as heat is not conducted away so fast.

when I build it, I plan not to use any wire or solder and directly screw on the binding posts

[zlymex]

The triangle shape should be a little less sensitive on how well the wire is connected / soldered. It could also make soldering more easy, as heat is not conducted away so fast.

when I build it, I plan not to use any wire or solder and directly screw on the binding posts

I made this purely by hand: hand cut, hand drill, hand soldering, without any measurement of geometry when make, symmetry done by eyeball observation only, no adjustment/trim of any kind afterwards.
Thickness of the triangle copper is 1.5mm, length of each side is aprox. 21mm, diameter of the copper bar is 2.2mm, diameter of the wires is 0.8mm.
When I run thru a current of 6.2A into 2 wires of any of the 6 combinations(excluding reversal), I measure the other two wires with less than +-0.3uV voltage variation with my 34401A.
This means that I have achieved the kind of zero that OHM-LABS claimed in their specification just by this simple DIY.
 

[Vgkid]

Before reading through the last few post I assumed that either a sphere, or diamond would be needed to make a zeroe ohm resistor.

[quarks]

I made this purely by hand: hand cut, hand drill, hand soldering, without any measurement of geometry when make, symmetry done by eyeball observation only, no adjustment/trim of any kind afterwards.
Thickness of the triangle copper is 1.5mm, length of each side is aprox. 21mm, diameter of the copper bar is 2.2mm, diameter of the wires is 0.8mm.
When I run thru a current of 6.2A into 2 wires of any of the 6 combinations(excluding reversal), I measure the other two wires with less than +-0.3uV voltage variation with my 34401A.
This means that I have achieved the kind of zero that OHM-LABS claimed in their specification just by this simple DIY.

my guess was, that a perfect symmetry is the most importand part, but if it is that easy, I am even more interested to see what is inside the Ohmlabs product

[zlymex]

I made this purely by hand: hand cut, hand drill, hand soldering, without any measurement of geometry when make, symmetry done by eyeball observation only, no adjustment/trim of any kind afterwards.
Thickness of the triangle copper is 1.5mm, length of each side is aprox. 21mm, diameter of the copper bar is 2.2mm, diameter of the wires is 0.8mm.
When I run thru a current of 6.2A into 2 wires of any of the 6 combinations(excluding reversal), I measure the other two wires with less than +-0.3uV voltage variation with my 34401A.
This means that I have achieved the kind of zero that OHM-LABS claimed in their specification just by this simple DIY.

my guess was, that a perfect symmetry is the most importand part, but if it is that easy, I am even more interested to see what is inside the Ohmlabs product

Agreed that symmetry is the key. The bulky copper also plays an important role if not perfectly balanced. And you always have the option if trimming if all fails.

[lowimpedance]

Here is a PDF (pt1 and 2, split to fit the forums upload conditions!!) which you resistance 'tinkerers' might find an interesting read.
Its a scan from the ESI Engineering bulletins; this one being the Traceability of Resistance Measurements no.30.

[Vgkid]

Thanks for those, I wish IET Labs hosted the esi documents.

[quarks]

lowimpedance thanks very much for the "Traceability of Resistance Measurements no.30" document


I just looked through my ESI documents to see if I overlooked the details about 4-terminal junction like Figure 27 in your document


but only found this (which also shows patent from manganins post)


 

[zlymex]

IET SRL 1 Ohm. Same specification as Fluke 742A 1 Ohm but different internal construction.

The internal construction is actually the same. Multiple resistors connected in parallel with the copper equalizing plates. But in the IET version the whole package was resin potted which allows to use much cheaper non-hermetic resistors.

Here is a new photo showing the oil leakage from one of the SRL-1. It seems that the pot is oil filled.

[quarks]

just in case anyone is still interested, here are pictures of my prototype and the final product

[Dr. Frank]

Very nice, quarks!

That's physically worthwhile and esthetically appealing.

Did you already check its Symmetrie, i.e. its zero value to best possible resolution?

Frank

[lowimpedance]

Neat, any pics of the inside of final build ?.
 Interested at the mounting of the binding posts through the diecast lid then to the copper plate !. I assume there is suitable mechanical isolation from lid to the plate. (eg nut holding post to lid then a few turns down the thread a nut/washer either side of the copper plate ?... just guessing hence any inside photos!)

[quarks]

here is a inside picture

my construction is made with 5mm copper plate and Pomona 3770 posts directly screwed in, because I did not want any other metal involved
(not even the screw nuts provided with the pomona binding posts) 
my main focus was "perfect" symmetry
(and it came out very well: mechan. it is within about 0.01mm in any dimension and el. so far it shows less than 1µOhm change in any plug combination I tried)
only the screw thread gave me trouble, because the binding posts have more than expected tolerances
because of that, the screws are very close fitting (and one of my 3770 broke off), but becuase of very fast oxidation I like the fact that the screws are most probably gas tight and therefore hopefully no need to have it gold plated

[guenthert]

does anyone know what is inside this "Zero Ohm Standard"

and explain the BALANCED TETRAJUNCTION DESIGN A TRUE ZERO-OHM RESISTOR

http://www.google.com/patents/US3252091

I glanced over that patent, but need another hint on what's it's purpose.  I read through https://www.google.com/patents/US5867018, can't claim that I understood that, but believe I grasp the advantage of using a four junction terminal as part of haman resistors and similar.  But what on earth is a single 0 Ohm resistance 'standard' for?  That question was asked in https://www.eevblog.com/forum/chat/a-precision-zero-ohm-resistor/ before and there were (naturally) some more or less funny replies, but I didn't see an serious attempt to answer it.

[Macbeth]

guenthert, you will be saying it is nothing more than an AY-O fingerbox 



There are many many practical uses for a fingerbox, everyone knows that. But there are many fakes or poor designs too. Some you can't even waggle the full finger but just the knuckle finger. Very poor design, but it doesn't disprove that a perfect fingerbox is out there somewhere. 

[chickenHeadKnob]

my main focus was "perfect" symmetry
(and it came out very well: mechan. it is within about 0.01mm in any dimension and el. so far it shows less than 1µOhm change in any plug combination I tried)
only the screw thread gave me trouble, because the binding posts have more than expected tolerances
because of that, the screws are very close fitting (and one of my 3770 broke off), but becuase of very fast oxidation I like the fact that the screws are most probably gas tight and therefore hopefully no need to have it gold plated

A thread will never be air-tight by itself. OK, there are some exceptions like stainless steel on stainless steel which can  gall and self weld shut. With solid gold or completely annealed dead soft copper you will still have the spiral leak. However you might be able to seal by peening  the exits I wouldn't trust such a seal for long term integrity. Thin flash gold on gold plating in the thread might work better. I would be tempted to do the simplest and cheapest: assemble as you have with the threads as clean as possible and then coat or dip the outside with paraffin wax. Leaving  the contact areas of the pomonas uncoated of course.

[zlymex]

just in case anyone is still interested, here are pictures of my prototype and the final product

Very nice mechanical construction and finish. However, frankly speaking, I think the binding posts should be soldered to the copper triangle to ensure even electrical contact. Otherwise, the screw force cannot be guaranteed the uniform distribution of contact resistant on all directions especially when the binding post are used.

[CalMachine]

*CalMachine casts a resurrection spell*  

Over the past few months I have been accumulating some test gear to get my own home lab up and running.  A few of the things I purchased were some resistance standards.  I snagged the ESI SR1010 because it was pretty cheap and it was the first one that I had seen with cal trim pots.  I was amazed when I opened this guy up.  I wasn't sure exactly what to expect...   Each trimmer has got a set of wire wrapped resistors with it!  I took these pictures a week or so ago...  I should have gotten a better one of the back trim pot PCB.

[ManateeMafia]

Ahhh. It took me a while to find one of those. It looks like you are in your way to getting the full set plus ESI 242D/E. The 10k version is called out in the 242 manual and is initially adjusted to match a higher standard like the SR-104. Then you can build up/down to the other SR-1010 using the series-parallel adapters.

I have the full set but need to set aside time to step through the manual. The SR-1030 are also nice but are more expensive.

[TiN]

I thought it was you who snagged stash of recent 742A's. 

[CalMachine]

I thought it was you who snagged stash of recent 742A's. 

If you're referring to the big lot on Equipnet, unfortunately not.    Someone over in Europe swooped up $200k+ test equipment for 0.pennies on the dollar before I even got a chance at bidding.

Ahhh. It took me a while to find one of those. It looks like you are in your way to getting the full set plus ESI 242D/E. The 10k version is called out in the 242 manual and is initially adjusted to match a higher standard like the SR-104. Then you can build up/down to the other SR-1010 using the series-parallel adapters.

I have the full set but need to set aside time to step through the manual. The SR-1030 are also nice but are more expensive.

It will probably be awhile before I accumulate an entire set and the bridge.  I just had to jump on this one because it was the first time I had seen one.

[TiN]

No, on ebay for 9k

[CalMachine]

No, on ebay for 9k

Ahh,  I must have missed that!

On Equipnet there was a lot of like 12 742As going for like 2 or 3k.  They had an 8508 going for 250, a 740A going for 1k, A40B shunt sent for 1k, 5720 for 7k, 5790 for 8k, 792 for 9k, 720A for 250... They had some triple point of water cells as well as some other pretty cool metrology gear.

[CalMachine]

Another goodie I acquired! This one is a Resistive voltage divider (or a Kelvin Varley Divider to be precise).  The ESI RV722! 

I'm sure pictures of the insides of one of these have been passed around already... well here are some more.

[Assafl]

That's a Mylar card millefeuille. Impressive.

[Pipelie]

 
Thanks for sharing, looks like the MTB have a oil tan? like his brother( SR1010 MTA).

[CalMachine]

Do you know the differences between the MTA and MTB?  I need to find a SB103 too.

[VintageNut]

@Calmachine   did you buy the KVD from eBay? I saw the 7-decade KVD for $500 or thereabouts. It was tempting, but I already have two Gen Res KVDs.

Do you have a lead compensator?  I bought a Fluke 721A to use with the KVD.

[CalMachine]

@Calmachine   did you buy the KVD from eBay? I saw the 7-decade KVD for $500 or thereabouts. It was tempting, but I already have two Gen Res KVDs.

Do you have a lead compensator?  I bought a Fluke 721A to use with the KVD.

I did!  I offered him $375 and he took it.  There was an RS925 up very recently that I was thinking about picking up.  I had my reservations as it was a non-A model.

I do not have a lead compensator yet.

[ManateeMafia]

If you go for a ESI 925, I would recommend the C or D models. They have the ability to trim the upper four ranges. Just look for the removable plate on the front of the case.

A few ESI 242 have popped up in the last couple of months. The D version is nice and I have one I assembled in pieces over a couple of years and mounted in a case. The E version is basically the same but the null meter is a Keithley 155 vs the HP design.
You can tell by the control knobs on the meter which model is installed.

[Henrik_V]

Hi,
I got my hand on a Leeds & Northrup Hammon Transfer Standard Resistor Type 4323. That 10x 1K  and according to
http://www.ohm-labs.com/resistance-standards/hamon-transfer-standards.html
it's good for 0.01ppm for a 1:100 transfer 

I got it because the former owner soiled it with mercury, so it's hazardous waste now 

However, I took some risc in the name of science ... well ... my curiosity...  and made a small tear down:

What looks like solder on the contacs and the bar is visible pure mercury, so nothing for my home lab.

[Henrik_V]

part II  some more pics
Nice to see the realization of the Zero Ohm junctions

[Vgkid]

Very Cool. The more I see old L&N gear, the more I like it.
It will safe to handle, just remember to wash your hands after touching those posts, after all it forms a mercury amalgam( they were used as fillings, for a very long time)
Here are the specs.

[guenthert]

[..]
I got it because the former owner soiled it with mercury, so it's hazardous waste now 

However, I took some risc in the name of science ... well ... my curiosity...  and made a small tear down:

What looks like solder on the contacs and the bar is visible pure mercury, so nothing for my home lab.

Actually Leeds&Northrup specifies for their standard resistors that the contacts need to be mercury wetted.  So such 'soiled' resistors shouldn't be all that rare.  And yes, thanks, but no thanks.  You can keep it.

[Vgkid]

I believe that up until the 80's standard resistors were supposed to be used with mercury wetted contacts. After all, they sold mercury cups/posts for that purpose.

[e61_phil]

Iih!

I was only afraid of cadmium solder until now...

[Henrik_V]

When you hunt for the next digit in metrology, physics don't care human biology
Cd in solder, Hg on contacts , my amatures are Be  ...   

[e61_phil]

When you hunt for the next digit in metrology, physics don't care human biology
Cd in solder, Hg on contacts , my amatures are Be  ...

Pure Be oder BeCu?

If you don't inhale Be dust it is quite safe, isn't it?

[Henrik_V]

I calibrate accelerometers and the amarture of the shaker is ceramic or mostly pure berillium. No problem unless you lick it for cleaning 
Even our workshop got used to it, because after some years you need to replolish the armature, and you don't what to inhale the dust.  (Primary calibration with laser doppler interferrometers) And these workshop guys are awesome.  They also polish the Si-'orange' for the new representation of the kilogram. Different story...

From Wikipedia: https://en.wikipedia.org/wiki/Ohm
The "international" ohm was defined as a mercury column 106.3 cm long of mass 14.4521 grams and 0 °C at the International Electrical Conference 1893 in Chicago.

[CalMachine]

Well... I'm back at it   This time the magical RS925D!   

This was purchased, for pretty cheap, in an attempt to get an ESI 242D system up and going.  I opened her up to reapply a new foam seal as well as make sure there are no obvious issues that needed to be addressed.  After some quick preliminary tests, it looked like some of the lower decades had slight a hysteresis.   The issue seemed to have disappeared after letting the decade, and the K2002 I was using, thermally stabilize together for over 24 hours.





Let's open her up!


100 kOhm decade


10 kOhm decade


1 kOhm decade


100 Ohm decade


Top and Insides of trim pot assembly PCBs and WW resistors for the 100 kOhm, 10 kOhm, 1 kOhm, and 100 Ohm decades











10 Ohm decade


1 Ohm Decade




100 mOhm decade



10 mOhm decade.  These units have a fixed minimum resistance of 10 mOhms.  Because of this implementation there is no 'zero ohm', similar to insertion loss, measurements need to be taken into consideration.  This allows for the displayed resistance value to actually be the absolute value.



100 µOhm Decade fine trim pot


More Goodies!

100 kOhm and 10 kOhm decades



10 Ohm and 1 Ohm decades



1k and 10 kOhm decades

[ManateeMafia]

Thanks for sharing. It looks like it in great shape.

[CalMachine]

Thanks for sharing. It looks like it in great shape.

You're are very welcome!  This is my first RS925D I've seen, so I wasn't 100% positive on the condition.  It seems as if there is a decent amount of flux residue still on some of the connections... is that normal? 

[TiN]

Auch, purple wire resistor on 10K decade looks damaged to me. I have some Fluke PWWs, and those edges are real easy to get damaged and open wires, due to very fine gauge.
Thanks for sharing.

[quarks]

here are pictures from my RS925A for comparison to the RS925D above

[manganin]

This time the magical RS925D!

That is a late one, seems to be from the mid 1980s.

Switches are the weak point, inferior compared with the General Radio counterparts. ESI solved the problem by paralleling several switch wafers. Which actually works fine, until dirty or totally worn out.

It would be a good idea to clean all the switches with a rough paper simulating the movement of the switch blade. You are ready when there is no dirt in the paper. Not easy to access and takes a lot of time, but worth the effort. Absolutely no switch cleaner chemicals!

Another thing worth checking are the screw connections of the slide wire. That is the 4-wire node so the possible change can only be seen in a 2-wire measurement. Open the screws, clean the surfaces and bend the spring washer teeth if required.

The trimmers are better that you would expect, but still the obvious reason for any high decade instability. That specific type is still available but very expensive. Turn back and forth a few times as a preventive measure.

[Assafl]

Why not use a switch cleaner (like Caig stuff)?

For electrometers and the like one has to ensure that any cleaner does not touch any isolators such as Teflon, but if done extremely sparingly (e.g. Just on the wiping metal surface) then it does do away with oxidation and the like.

Does it affect calibration if it does not touch insulation?

(By sparingly I mean with a syringe or a wipe, and watch oilers do work well to dispense tiny drops into leaf contacts. )

[CalMachine]

Auch, purple wire resistor on 10K decade looks damaged to me. I have some Fluke PWWs, and those edges are real easy to get damaged and open wires, due to very fine gauge.
Thanks for sharing.

I didn't really notice that at first, Good Eye!  Is the only coarse of action to repair said damage, to replace the resistor?

This time the magical RS925D!

That is a late one, seems to be from the mid 1980s.

Switches are the weak point, inferior compared with the General Radio counterparts. ESI solved the problem by paralleling several switch wafers. Which actually works fine, until dirty or totally worn out.

It would be a good idea to clean all the switches with a rough paper simulating the movement of the switch blade. You are ready when there is no dirt in the paper. Not easy to access and takes a lot of time, but worth the effort. Absolutely no switch cleaner chemicals!

Another thing worth checking are the screw connections of the slide wire. That is the 4-wire node so the possible change can only be seen in a 2-wire measurement. Open the screws, clean the surfaces and bend the spring washer teeth if required.

The trimmers are better that you would expect, but still the obvious reason for any high decade instability. That specific type is still available but very expensive. Turn back and forth a few times as a preventive measure.

Thank you for the advice!    I'll work on this here in the future.  Will probably post some before/after pictures, when I get to opening her back up.

[0.01C]

I love 925D ,it is a great resistance stanard

[Edwin G. Pettis]

That discoloration on the 10K may be an indication of overload or a chemical spill, the coating on these resistors are not very resistant, even isopropyl will soften it up and if you get any on it, it will have to dry out thoroughly but has no lasting effects.  You'll have to check the resistor with a high accuracy measurement to see if it is out of tolerance.  You can use a contact cleaner like deoxit but do not spray it on the resistors, use a lint-less wipe.  Yes there is always some rosin residue left on the solder joints, it is of no consequence.  Usually these units require very little maintenance.

I have both a RS925D (part of my 242D) and a RS925A (part of a 242B) which I am going through and giving it a once over.  The connections on the rheostat are very important and should be snug (do not over tighten), the resistance element/wiper can be gently redressed with crocus cloth and cleaned.  I have found some units with some kind of gooey grease on the rheostat, that attracts dirt, I prefer to keep them clean and dry.  These units are more or less sealed in their case, there should be little to no contamination getting in there.

These resistors are wound on a mica card and therefor is somewhat easily damaged, not only the fine wire but the mica so be very careful around them.

[CalMachine]

She's Alive!!!!   

[e61_phil]

Is there a Helmut Singer Sticker?

[CalMachine]

Is there a Helmut Singer Sticker?

Correct!

[Vgkid]

I remember seeing one on ebay for $1500 obo, it sat on there for years. I think it was a dealer based on the West Coast.

[ManateeMafia]

The black monolith is alive. Now the race is on to see which one of us can figure it out first. Mine keeps looking at me wondering when I will stop fixing stuff and start measuring my SR1010 and SR104 resistors.
I think there will be some projects to do soon involving resistance bridges.

[0.01C]

She's Alive!!!!   

How do you calibrate this black monolith? I have a 242C still sleeping on the floor.

[ManateeMafia]

According to the manual you need a 10k reference resistor with 1ppm calibration accuracy (SR-104) and a trimmable 10k/step transfer standard like the hard to find SR 1010 MTx.

[CalMachine]

The black monolith is alive. Now the race is on to see which one of us can figure it out first. Mine keeps looking at me wondering when I will stop fixing stuff and start measuring my SR1010 and SR104 resistors.
I think there will be some projects to do soon involving resistance bridges.

You're on!    I don't have any of the connectors or adapters so I ended up making my own.  We'll see how they turn out, as I have not put everything to the test yet.  I spent most of the day cleaning everything up on the outside, getting her looking good, as well as calibrating the 801.  The 801 performed very well and needed no adjustments coming from Helmut Singer. 

According to the manual you need a 10k reference resistor with 1ppm calibration accuracy (SR-104) and a trimmable 10k/step transfer standard like the hard to find SR 1010 MTx.

Done and done!   

[ManateeMafia]

Ok. Let's communicate offline and get some details worked out.

We can start a new thread and post any results if you want to do some detailed comparisons of portable standards.

[CalMachine]

Ok. Let's communicate offline and get some details worked out.

We can start a new thread and post any results if you want to do some detailed comparisons of portable standards.

Sounds good! And yes, I would like that very much.

[Vgkid]

I look forward to reading about your 242 endeavors.
I have a feeling E.G. Pettis will chime in.  

[Edwin G. Pettis]

The SR-1010 10K does not need to be trimmable, it is directly compared to the SR-104 for each step and the readings noted, the trimmable SR-1010 is optional.

Okay, I chimed in......

[CalMachine]

The SR1010 I picked up a few months ago is the MTS version, and I've already trimmed each step to our Ohm Labs 10k standard resistor.  If I remember correctly (I'm not at the lab), but I think our Ohm Labs 10k standard has drifted about 2ppm over the last 5 years.  So it's holding pretty strong!   I plan on shifting it to our working standard once I can get a SR104 in our hands with sufficient data to use as our lab's primary standard.

I tore it down, slightly, and posted pics on page 4 for those interested!   

[ManateeMafia]

I thought you partially tore down your Ohm Labs 10k standard. I was disappointed they weren't there.   

Which model do you have? Is it the 104 or 204?

[CalMachine]

I thought you partially tore down your Ohm Labs 10k standard. I was disappointed they weren't there.   

   No way I'm opening up our current, freshly cal'd, 10k standard.

Which model do you have? Is it the 104 or 204?

We have the 100 (1 amp version) and 104.  I wish we had the 200 series!  We were really small back then when they picked them up, before I was around. 

[ManateeMafia]

A couple of years ago I got a quote on the 200 1 ohm model and the price was $3400.00

[PartialDischarge]

Very interesting thread.

However I noticed that very few resistance standards use an oil bath ( the Leeds Northrup and another potted resistor If I recall did use it), and none have heated ovens, something common in voltage standards.
Why is that? any guesses?

[TiN]

Elevated temperature accelerate aging drift.
Same apply to voltage references, but with them it's "lesser evil" and in the end result is better. Not so much for passive resistor though.

[PartialDischarge]

Elevated temperature accelerate aging drift.
Same apply to voltage references, but with them it's "lesser evil" and in the end result is better. Not so much for passive resistor though.

Hmm, well since voltage standards use resistors too, to set currents, and as voltage dividers, heat would be evil too in that sense.
I've been reading for example about the HP 735A, its basically a selected and ovenized (at 80ºC) 6.3V zener, whose output is divided to 1V via resistors, ie no op-amp feedback or similar. Two
of those resistors are inside the oven. If those resistors drift fast its game over for it. But these little things were pretty stable over time.

More about it here https://www.eevblog.com/forum/testgear/hp-735a-dc-transfer-standard/

[PartialDischarge]

Earlier this year I was too busy and missed this one, how is it not referenced here??

https://www.eevblog.com/forum/metrology/fsl-project/
https://xdevs.com/article/fsl935/

[Echo88]

Well, its a prototype, not a standard resistor. Regarding the HP 735A-post: The necessary resistors have to be inside the ovenized compartment together with the voltage reference, otherwise the temperature difference between voltage ref and resistor junctions would create drifting parasitic thermal emf voltages.

[PartialDischarge]

Regarding the HP 735A-post: The necessary resistors have to be inside the ovenized compartment together with the voltage reference, otherwise the temperature difference between voltage ref and resistor junctions would create drifting parasitic thermal emf voltages.

Oh, the HP735A was just an example to see that resistors are heated in voltage standards. The Fluke 732B also does the same. Thus my question of why we see no resistance standards with heating, but well Fluke also thought of that in the SL935

[ap]

There absolutely are resistance standards that are heated. E.g. Measurements International offers them. Price is steep though.
You can also diy for your labs, using oild filled precision resistors and a precision heater device.

[schmitt trigger]

Well... I'm back at it   This time the magical RS925D!   

Watching these images, I'm completely in awe, no, actually speechless, at the top-level workmanship of these units.

I'm pretty sure that this level of craftsmanship was obtained in a shop which employed only a very few, very skilled workers.
Ran by managers which were technically very competent.

Back then when the label: MADE IN USA actually meant something special.

[Echo88]

Interesting ap! A few standard resistors to check your equipment? Are they more suitable for some jobs than your 5720A?

[ap]

Actually I use 4808s as sources (which, btw, in many areas have better uncertainty specs than a 5720), but resistors uncertainties of calibrators are usually pretty high. The 5720 10kOhms spec is e.g. 9ppm for 90 days, +/-5K and 3ppm at +/-1K (plus factory calibration). A hermetic, thermo-regulated resistor has <0.5ppm/a uncertainty (plus calibration uncertainty). The temperature stability can easily be made better than 0.1K.

[Vgkid]

There absolutely are resistance standards that are heated. E.g. Measurements International offers them. Price is steep though.
You can also diy for your labs, using oild filled precision resistors and a precision heater device.

Do you have a build thread for your resistance standard. I feel like I have seen it before.

[ap]

There was some mentioning earlier, but no real build thread, just one or two more pictures.

[TiN]

It was unavoidable, at rate I'm lurking and buying stuff of evilbay. RS925D ESI arrived today. Great deal at 600$ huh?

Hmm, what is that rattling sound... 





Ah... those are bits of coal from the switch... ... 

0.1 ohm decade is cooked to coal condition. Other decades "seem" to work, but I'd be very surprised if whole box is any good other than metal enclosure and binding posts.
Oh well, don't go shoppin unprepared, dear precision-nuts. 

P.S. was listed as "used".

[cellularmitosis]

Thanks for the cautionary tale!  That's a bummer.

[Vgkid]

Damn, interesting things always happen when TiN spends money...

[CalMachine]

That is most unfortunate =(  I was hoping the RS925 you picked up was going to be a good one, so we can get some international transfers going     

We will have to resort to other plans.

Perhaps, now, you might have a donor unit for fellow precision-nutt hobbyists who might have faulty parts. (I don't know.. perhaps $50 or so per dial?  ) 

How does the µOhm vernier feel as you make complete revolutions?

[alm]

P.S. was listed as "used".

At least that gives you a shot at returning and/or receiving a (partial) refund. Ebay's description of used is "fully functional, but possibly some sings of wear that are described in the pictures and/or item description". So based on that, any non-described defect is a reason to consider the object not as described. Severe undisclosed defects make it significantly not as described.

[ManateeMafia]

The initial accuracy of 20ppm on the untrimmed decades would be difficult to match. The 0.1 ohm decade on my RS925D has signs of filing/sanding where they adjusted each step by hand. Not sure about the 0.01 decade but it was probably done the same way.

[TheSteve]

It was unavoidable, at rate I'm lurking and buying stuff of evilbay. RS925D ESI arrived today. Great deal at 600$ huh?

Hmm, what is that rattling sound... 

Crispy sadness.

[plesa]

The initial accuracy of 20ppm on the untrimmed decades would be difficult to match. The 0.1 ohm decade on my RS925D has signs of filing/sanding where they adjusted each step by hand. Not sure about the 0.01 decade but it was probably done the same way.

In HARSX series the 0,01 and 0,1 Ohm ranges looks like trimmed by solder joints inside bend area. I can post some pictures. On TiN picture the solder has been evaporated :-)
I like the shadows on front panel. RIP

[manganin]

0.1 ohm decade is cooked to coal condition. Other decades "seem" to work, but I'd be very surprised if whole box is any good other than metal enclosure and binding posts.

Luckily the lowest decade resistive elements are identical to the ESI DB series decade boxes, which are very cheap. Especially the low resistance ones.

Note that the DB series has only two parallel switch sections while the 924D has four.

By extending the spindle you can parallel all six sections, there is plenty of space. Better than the original...

[Edwin G. Pettis]

Hello TiN,

It is really aggravating when someone abuses precision equipment like this, I hope they got fired for it, no excuse.  You could have a case for a refund given the condition it is in, definitely not in working condition by any definition.  That decade was severely overloaded, the resistors are definitely ruined, the switch might be recoverable hard to tell from the photo.  The blackening on the panel indicates severe stupidity.  If the switch could be recovered, it would be possible to rebuild the resistor chain since it is of Manganin alloy.  However, given the design of the 925D, the 0.1? decade has to calibrated to the 925D's 'zero' ohm value which is different for every unit, just 'plugging in' another decade will result in higher error.

[TiN]

Just got the message from seller, with promise to honor my partial refund requestGot the agreed partial refund, leaving me which what it is.
So eventually I'll be looking on ways to restore this unit, however at this point it likely to collect dust on the shelf, until I get decade element for replacement.

In the end whole thing must be calibrated, so I'd expect current "zero ohm value" doesn't mean anything anymore. What puzzles me a bit, how the other decades are not damaged (at least visually) as per my understanding they all in series? Maybe operator had all higher decades set to zero, and was trying to use RS925D as a 0.x ohm shunt?? That would be the only possible explanation.

[Edwin G. Pettis]

The .01? minimum value consists of the interconnect wiring between the decades (at 'zero' setting of course) plus the minimal error contributed by the vernier resistance which is not quite 'zero' either.  The 1st decade setting of the .01? bank is adjusted to correct (as best possible) for that total 'zero' resistance, the subsequent settings on the .01? decade are accordingly adjusted at each step and so on.  The 0.1? decade resistors are adjusted to the .01? bank.  The 1? and 10? banks are wound and either selected or adjusted to also account for the lower bank's error to stay within the required accuracy spec.

Since your 0.1? decade is destroyed and possibly the switch as well, the only compensation required to the total 'zero ohm' net resistance is if a new switch has a significantly different 'zero' resistance that the original switch if replaced.  If the resistance of the new switch is sufficiently different to throw the accuracy of the other low ohm banks off of spec, then you would indeed have to re-calibrate the lower decks.  This is a cascade failure in that one low end bank could cause the entire low end to require re-calibration and the upper adjustable banks need to be re-calibrated anyhow after traveling and obviously being mishandled, its accuracy/calibration is unknown.  I would also verify that the .01? resistors are okay, that black on them may be just blow back from the 0.1? resistors but then....maybe not.  There is no public copy of the calibration procedure for a newly made RSxxx standard.

When time permits, I would first remove the damaged resistors and then clean up the switch decks and see what condition they are in before doing anything else,  the question of whether or not the switch requires replacement is the first step, we already know the resistors are toast (literally!).

Personally I have never seen an RS925D is such bad condition, sometimes I've seen dec boxes with damaged resistors (DB series or even SR-1010s) but usually standards of the RS725 or RS925 series are treated with kid glove care, they are worth too much for such tomfoolery.

May I inquire what country this came from?

[TiN]

I've also got a word that decade from new IET RS925 are not compatible with this older ESI unit. It came from USA as most of my stuff

[Edwin G. Pettis]

Well, I'm not too surprised, IET has been busy trying to cut costs some as those older ESI switch decks were of very high quality and weren't too cheap back when ESI  was buying them.  I'd say if you want to try and save this unit, start by cleaning and checking the switch, if that is still in good shape, that is one less hurdle, next would be replacing the resistors and re-calibrating it, could be worth the trouble.

Argh, it appears that this website doesn't support standard special character coding, it changed my Omegas into '?'....tsk, tsk.

[TiN]

Will do, Edwin, thanks for suggestions. I might also first test higher decades by bypassing the lower side connection, just to check if whole thing worth the efforts.
Based on smoke marks it would look like there was open fire of spacer at 0.1 decade, which could very well caused PWW's above it to be cooked as well.
I'd expect resistance wire stability may be affected by that.

And yea, none of &Omega; &deg; &plusmn; stuff working here. Another reason why I maintain own site .

[alm]

\$ 1 \Omega \$ (Mathjax) works on the desktop version.

[sarel.wagner]

My lesson also learned on the DB788, burnt    but not quite as bad as Tin's unit.... Managed to get some NOS resistors to repair. Also talking to Edwin about that first decade resistors.

Rgrds

[manganin]

What puzzles me a bit, how the other decades are not damaged (at least visually) as per my understanding they all in series?

I = U/R and P = U*I

Low voltage / high current supply will damage the lower decades. In case of high voltage / low current supply the higher decades will suffer.

Higher decade switches are probably just fine. The switch current carrying capacity is much larger than the switching capacity.

The ESI 801 source is not able to damage the 925D, even if directly connected. Must have been something with output current much higher than 2A.

[manganin]

IET has been busy trying to cut costs some as those older ESI switch decks were of very high quality and weren't too cheap back when ESI was buying them.

GenRad 1433 by General Radio:




Genrad 1433 by IET Labs:




More IET photos:
http://www.edn.com/design/components-and-packaging/4421782/When-wires-become-components

[CalMachine]

Quality and craftsmanship has hit a rock bottom, it seems...  Thanks IET! 

[Macbeth]

Just look at the ludicrously overpriced relabeled IET DE-5000/6000 debacle compared with the Taiwan DER EE DE-5000 and all the bullshit that goes with it.

I've always wondered why Dave gave a glowing review and never used it since, he always uses his Agilent LCR like with the $7 AVR 328 Transistor Tester review that is so recent.

[cellularmitosis]

Just came across this powerpoint presentation about the history of the Ohm and resistance standards.  Great intro to the subject!

https://pdfs.semanticscholar.org/presentation/7395/d0272221da51ebe5b45970a3bb55c29d0f30.pdf

[ManateeMafia]

nice find. 

It is good to see some of my own standards covered and their history. There is an article out there that describes the method used to build a 4210 resistor and how the wire was annealed including temperatures. Have you come across that one?

[Dr. Frank]

Very nice presentation, thank you!

Resistance  seems to be a typical German affair: Georg Ohm, Werner von Siemens, Wilhelm Eduard Weber, Otto Wolff, Reichsanstalt Ohm, von Klitzing QHR... great evolution story.

Next year, 20th May, this story will come to a finish, as the Ohm will have a zero definition uncertainty, and an extremely small realization uncertainty, due to being a quantum standard, finally.

Frank

[Vgkid]

nice find. 

It is good to see some of my own standards covered and their history. There is an article out there that describes the method used to build a 4210 resistor and how the wire was annealed including temperatures. Have you come across that one?

Here you go
Stability of double-walled manganin resistors - Nvlpubs.nist.gov…
PDFhttps://nvlpubs.nist.gov › nistpubs › jres

Another good one:

A new design of precision resistance standard - Nvlpubs.nist.gov…
PDFnvlpubs.nist.gov › jresv5n2p295_A2b

[alm]

The above two articles as links:
Stability of double-walled manganin resistors
A new design of precision resistance standard

[Vgkid]

Thanks , my phone doesn't like to copy certain links for some reason.

[ian.ameline]

I have recently acquired a General Radio 1433-M in very clean condition. I've tested it (as well as I can with an HP34401A with kelvin probes, and using recommended metrology practises.) All of the values on all of the decades are within 0.01% +- 2mOhms, with the exception of 3 of the values in the 100 Ohm/div decade -- 500 Ohms is high by 0.012% and 600 and 700 are high by 0.011%. Several other values in this decade are at 0.01% high.

The other decades are all closer -- the 1k/div has all values within 0.001% !!!. The highest decade is about 0.005% low, and the low three decades are between 0.004% and 0.009% high, with the exception of the 100 ohm/div decade which is 0.009% to 0.012% high.

Other than cleaning the contacts with deoxit (as recommended by IET Labs) is there anything else to be done? I don't think I can say with any confidence that the unit is in or out of spec without something with more precision and less uncertainty than an HP34401A.

The IET documentation says to lubricate using Synco Super-Lube with PTFE -- would not the PTFE particles suspended in the synthetic grease increase contact resistance? This unit at 0 measures at under 2.5 mOhm using kelvin probes on my HP34401A. The General Radio specs for these switches was  < 0.5 milli-Ohms of contact resistance. With the exception of the three values above, it is so precise, I'm reluctant to fuck with it.

Any advice?

[Edwin G. Pettis]

According to original General Radio specs, the 1433-M accuracy is ±0.01% + 2m , so it appears that most of the resistors are within tolerance, with a few of them close to the limits and some of the 100 ohm units a bit on the high side possibly.  The newer IET versions are using significantly lower quality switches and they appear to have began replacing mica card resistors with film/foil resulting in lessor specifications.  This is probably due to their inability to manufacture the mica card resistors.  Given the accuracy of the 34401A on low ohms range, a fair number of your readings are on the 'ragged' edge of being out of tolerance.

In my opinion, the older genuine GR units are of higher value than the new units.

[Vgkid]

Thr new IET 1433's appear to use the same switches as the ESI DB(the old ones , and probably new as well) series boxes.

[ian.ameline]

According to original General Radio specs, the 1433-M accuracy is ±0.01% + 2m , so it appears that most of the resistors are within tolerance, with a few of them close to the limits and some of the 100 ohm units a bit on the high side possibly.  The newer IET versions are using significantly lower quality switches and they appear to have began replacing mica card resistors with film/foil resulting in lessor specifications.  This is probably due to their inability to manufacture the mica card resistors.  Given the accuracy of the 34401A on low ohms range, a fair number of your readings are on the 'ragged' edge of being out of tolerance.

In my opinion, the older genuine GR units are of higher value than the new units.

On the 1 Ohm/division all but one reading was within the 2 milliOhm uncertainty, and the single one that was more than 2 milliOhms was 2.2 milliOhms out. So I think those ranges are ok (within what a 34401 can say). Its the 100 Ohm/step decade that is on the "ragged edge".

I'm attaching the readings (the 0 offset has been subtracted from all readings on each scale of the 34401) The "corrected" column is the actual readings with up to 2 milliOhms of error subtracted. (The 34401 had been on and connected to the unit with kelvin probes for 24 hours no movement or touching other than turning the knobs and waiting for 30 seconds, temp is a constant 22C, with no gradients or breezes) The readings are very stable and very reproducible - to within 1 digit in the least place every time.

I agree that the older units seem to be much better built. I've ordered some deoxit -- I wonder if it can trim 10 milliOhms off the resistance of the 100 Ohm/Step decade...

Still, for $140, not too bad. IET wants $2300 for a new one.

[ian.ameline]

Given the accuracy of the 34401A on low ohms range, a fair number of your readings are on the 'ragged' edge of being out of tolerance.

For the low ohms range, the 34401 is 0.01% of reading + 0.004% of range (which is 100 Ohms.) For the 1K range it is 0.01% of reading + 0.001% of range.

So for 500 Ohms, it would be 50 milliOhms + 10 milliOhm -- 60 milliOhms of uncertainty. My reading for the 500 Ohm setting on the 1433-M is 500.062 Ohms. Just outside of the uncertainty of the meter -- So without a more accurate & precise meter, I don't know that I can say that the 1433 is out of spec. (Which should be +- 52 milliOhms for a setting of 500 Ohms.)

[Conrad Hoffman]

...
The IET documentation says to lubricate using Synco Super-Lube with PTFE -- would not the PTFE particles suspended in the synthetic grease increase contact resistance? This unit at 0 measures at under 2.5 mOhm using kelvin probes on my HP34401A. The General Radio specs for these switches was  < 0.5 milli-Ohms of contact resistance. With the exception of the three values above, it is so precise, I'm reluctant to fuck with it.

Any advice?

AFAIK, the PTFE shouldn't be a problem. The Super Lube products are a PAO oil, supposedly an excellent lube for electrical contacts. The thickener is fumed silica rather than the usual soaps. It would be interesting to re-grease it and run the resistance test again just to be sure. I think the old GR literature suggested Vaseline for lube, and maybe some other grease, depending on the vintage. One has to remember that the GR resistance boxes are very good, but they were never intended as metrology grade calibration standards.

[ian.ameline]

...
The IET documentation says to lubricate using Synco Super-Lube with PTFE -- would not the PTFE particles suspended in the synthetic grease increase contact resistance? This unit at 0 measures at under 2.5 mOhm using kelvin probes on my HP34401A. The General Radio specs for these switches was  < 0.5 milli-Ohms of contact resistance. With the exception of the three values above, it is so precise, I'm reluctant to fuck with it.

Any advice?

AFAIK, the PTFE shouldn't be a problem. The Super Lube products are a PAO oil, supposedly an excellent lube for electrical contacts. The thickener is fumed silica rather than the usual soaps. It would be interesting to re-grease it and run the resistance test again just to be sure. I think the old GR literature suggested Vaseline for lube, and maybe some other grease, depending on the vintage. One has to remember that the GR resistance boxes are very good, but they were never intended as metrology grade calibration standards.

I also have some deoxit lithium based grease (L260Np), I'll give that a try on the 100Ohm decade along with the deoxit cleaner and see what effect that has.

[Vgkid]

A quick impromptu teardown. L&N 4221-B

[Conrad Hoffman]

With all the concern about spot welding and wire fastening in other threads, and the impossibility of it being done well enough my mere mortals and DIYers, it's hard to believe the crude construction of the 4221 yields as good performance as it does.

[Vgkid]

With all the concern about spot welding and wire fastening in other threads, and the impossibility of it being done well enough my mere mortals and DIYers, it's hard to believe the crude construction of the 4221 yields as good performance as it does.

Very true , especially since it was still getting calibrated until 2006. Then again, I wonder what the ultra precision low ohms resistors look like(ohm-labs , Measurements International). I'm think not much different.

[Edwin G. Pettis]

You are comparing apples and oranges here, these resistors such as the L&N 42xx series are made with Manganin and are soldered, PWW resistors such as I make are made with Evanohm alloys which are not solder-able and must be welded.  It is not the 'crude' construction as much as the processes used to make them stable.

These resistors require entirely different methods to produce the end result, using the same methods on Evanohm resistors wouldn't even bother them.  Some of the Soviet style resistors similar to the L&Ns were actually made with a relatively thick sheet of Manganin with heavy copper leads soldered to the sheet.  Others such as L&N used heavy gauge wire instead with smaller gauge wire being used for higher values.  On the whole, the techniques were not that different between the Manganin type standards.  None the less, it does take considerable care and handling to produce such stable Manganin resistors, it isn't an off-the-shelf type of thing.  Manganin does remain very sensitive to handling and its environment unlike the Evanohm standards such as the SR-104.

[HighVoltage]

With all the concern about spot welding and wire fastening in other threads, and the impossibility of it being done well enough my mere mortals and DIYers, it's hard to believe the crude construction of the 4221 yields as good performance as it does.

Yes, I expected somehow a "much better", whatever that means.
By the look of it, we all could have built something like this L&N 4221-B ourselfs

[Vgkid]

Here is some reading material:
https://www.google.com/url?sa=t&source=web&rct=j&url=http://www.dtic.mil/dtic/tr/fulltext/u2/a284623.pdf&ved=2ahUKEwiT5MGh8JjeAhVBnOAKHVgUA_4QFjADegQICRAB&usg=AOvVaw1gjJqlrXjH1vYJwjzwvhrI

[BradC]

By the look of it, we all could have built something like this L&N 4221-B ourselfs

I will just say that years of hard won experience (across many different disciplines) will indicate that looks can often be deceiving. I'm sure there is an awful lot more to it than grabbing a spool of manganin and soldering on some binding posts when it comes to any post assembly treatment to manage long term accuracy and drift.

Of course I could be wrong.

[Vgkid]

By the look of it, we all could have built something like this L&N 4221-B ourselfs

I will just say that years of hard won experience (across many different disciplines) will indicate that looks can often be deceiving. I'm sure there is an awful lot more to it than grabbing a spool of manganin and soldering on some binding posts when it comes to any post assembly treatment to manage long term accuracy and drift.

Of course I could be wrong.

The link I posted goes into some of the processes involved.

[Mr. Scram]

I remember from some of Dave's videos that high grade decade boxes tend to look similar, but that they appear to be trimmed after assembly. Crude looking isn't necessarily crude.

[calija]

Dear all,

I am new and I got resistor fluke 742A-1.
One of the pin starts to moving on - can you help me to show how I can open this type of resistor.

Best regards!!!!

[MegaVolt]

help me to show how I can open this type of resistor.

https://www.eevblog.com/forum/metrology/teardown-standard-resistors/msg891672/#msg891672

[calija]

yes I find this pictures but I can t find screws  !!!

[MegaVolt]

yes I find this pictures but I can t find screws  !!!

It looks like they are under the sticker on the front panel.

[MegaVolt]

Or, back under the black elements there are just 4 of them.

[calija]

Yes, you are right  !!!!

  what a stupid solution!!!!

Thanks!!!!

[calija]

Under sticker  !!!!

[HighVoltage]

Under sticker  !!!!

Can you post some pictures?

[calija]

here it is!!!!

[Dr. Frank]

Hello and welcome, calija!

You're really new here.. we all would like to see pictures of the interior, of course 
Would you mind taking some of the interior construction and post them here, please?

Frank

[calija]

that's all folks!!!!

[calija]

I see internal picture in this blog
https://www.eevblog.com/forum/metrology/teardown-standard-resistors/
first page!!!!

[HighVoltage]

Thank you for the pictures calija!

I am surprised to find the screws under the metallic label.
The is not an easy production step, since the label is squeezed to the housing by the binding posts.
Very interesting.

[TiN]

It's actually very good anti-tamper feature

[notfaded1]

Here's some of mine... first is esi dekabox db62

B

[notfaded1]

Second is venerable General Resistance RTD-100

B

[notfaded1]

Next is Ohmite Decade-Ranger

B

[notfaded1]

I have more but need to take pics... amazing the variations in the designs inside and craftsmanship.    I have a Dekavider RV622A I've never opened before still with a 10V reference attached to it from an old CAL lab I wanna open.  For as old as it looks it was pretty on.  I'm not sure if they (the lab) made the top part or if this was some product someone sold?

[Vgkid]

Thanks for the teardowns, i was always curious what the esi low ohm decades looked like.
It looks like there are several GR rtd simulator owners here.
So you bought the rv622a mystery combo

[Edwin G. Pettis]

No that grey chassis is a home grown attachment, not made by ESI.

[Vgkid]

I could easily tell it was not esi  made.

[notfaded1]

Thanks for the teardowns, i was always curious what the esi low ohm decades looked like.
It looks like there are several GR rtd simulator owners here.
So you bought the rv622a mystery combo

I'll have to post some 3458 measurements of the mystery combo.  I spent a while last night working on raspberry pi3 with latest kernel getting Keysight 82357B GPIB working.  Thanks for your guide TiN it helped me look in the right direction.  Getting the kernel source version correct so I could compile kernel modules was the trickiest part.

Bill

[notfaded1]

I new the top part wasn't made by ESI but it almost looks factory produced.  Both the 10V reference and the 100:1 divider have matching cases on the back.  From preliminary tests it's pretty solid.  I'm sure the Dekavider won't be giving my F720A a run for its money but it seems to work pretty well.

B

[serg-el]

An interesting site.  A lot of information.
http://etoysbox.jp/1_TestEquipment/17_Std_R/Std_R.htm

[serg-el]

ADVANTEST 10k, find no spec, 1 hermetically sealed WW resistor inside.

[ArthurDent]

Here's a photo of the resistor string inside a Guildline 4281-G volt box.

[ArthurDent]

Here's a photo of the resistor string inside an ESI SV-194 volt box.

[wolfy007]

Australian made J.L.William Scientific Instruments (https://en.wikipedia.org/wiki/Les_William) decade boxes, serial numbers 3548 & 5136, showing a slight change in enclosure but otherwise identical internally.

Also have a Lindeck Potentiometer by the same maker, but havent opened that one up yet.

[ArthurDent]

You don't need switches to select different resistor values. Unlike most of the resistance boxes shown above, some of the really old decade resistor boxes used tapered shorting plugs to connect for the various combinations. Here is a Welch Scientific 3 decade box that goes from 0 ohms to 111 ohms in 0.1 ohm steps. The tapered shorting plugs have very low contact resistance.

Some of the larger units could probably be used as cribbage boards as well.   

P.S. - the meter in the background is reading my LTZ1000A 10 volt reference.

[TimFox]

Do you have any information on who made these resistor units sold by Welch?  Welch was an old Chicago company, which merged with Sargent in 1968, and now operates as Sargent-Welch from the Chicago suburbs, dealing in educational products.

[Gyro]

As the thread is slowly migrating cover decade boxes and dividers too, I'll link my Muirhead D-801-D voltage divider.

As it is a Reyleigh divider, it has 20 matched resistors per decade. This allows it to maintain a constant input resistance (true potentiometer). The switch contacts are self-lubricating Silver-Graphite on Silver, with non-inductive windings on Mycalex cards. The residual resistance is <1ppm.




https://www.eevblog.com/forum/testgear/kelvin-varley-reistor-porn/msg773770/#msg773770

[fenland787]

Hi,

Hope it's okay to jump on this thread but maybe you can help. I just got a Muirhead D-801-D delivered today - my first brush with a 'proper' voltage divider box - won on a well known auction site.

At first sight it seemed to function correctly other than a few switch positions being slightly sensitive to 'jiggling' the switch but then I noticed the '0.01' range went squiffy with the output voltage higher than it should be, in the 8 and 9 positions.

On opening the box, to my relief there was no sign whatsoever of anything having got hot, and apart from some of the switch contacts looking like some TLC with cotton buds and IPA would not go amiss all looked clean and good.

It was quickly evident that the end-end resistance of one side of the decade in question was exactly and precisely one resistor's worth too high at 1K0 rather than the 900R I was expecting, sooo:

1/ Hard to be sure by looking but based on the number of solder joints I think I see per stage, am I correct in my assumption that each of the resistors is actually wound as two in parallel to cancel winding inductance?

2/ If (1) is correct then my hope was that it could just be a 'dry joint'  (a cold joint on the other side of the Atlantic!)  but the 2.5 joints I can actually see/get to look fine so I have not tried a reflow (yet!).

3/ If (2) is actually a possibility, how easy is it to take things apart enough to get to any solder joints on the reverse side of the card and, indeed, are there any?

As a check I soldered a 200R type RC55Y precision resistor I happened to have in stock across the appropriate switch wires and the thing now works as it should.

As ever, thanks in advance for any pearls of wisdom....

Adrian

[Gyro]

Hi Adrian, welcome.

Congratulations on your acquisition, if it's the one I think it is, the listing says "All switches and resistors visually checked. Cal checked" and mentions the 6-7 switch issue on the 0.01 decade to be fixed. I'm not sure if you have any comeback there.

It sounds as if it's generally in good condition anyway, and I can't see how any problem on that specific setting could be caused by any sort of overload. You might find the occasional unit that has been used as a rheostat, but for voltage divider use, they are pretty bulletproof. I would expect it to be well within spec anyway.

The contacts are solid silver stud with silver graphite wipers. The listing mentions DeOxit has been used. If the contacts are showing 'crud' then, yes, a good clean with IPA would be certainly be a good idea. The switches are supposed to be self-lubricating, but if it's been sitting for a long time, there's still the possibility of oxides or sulphides forming. Exercising the switches should be enough to clear that but I've found that the smallest trace of Electrolube EML, or even Vaseline keeps the surfaces well protected.

It was quickly evident that the end-end resistance of one side of the decade in question was exactly and precisely one resistor's worth too high at 1K0 rather than the 900R I was expecting, sooo:

1/ Hard to be sure by looking but based on the number of solder joints I think I see per stage, am I correct in my assumption that each of the resistors is actually wound as two in parallel to cancel winding inductance?

From memory, and looking at my photo above and the linked post, yes, I think most of the decades use parallel counter-windings to minimise inductance. I'm not sure about the highest one, I suspect that that one has enough resistance to swamp out the stray inductance.

2/ If (1) is correct then my hope was that it could just be a 'dry joint'  (a cold joint on the other side of the Atlantic!)  but the 2.5 joints I can actually see/get to look fine so I have not tried a reflow (yet!).

I suppose it's possible that one of the joints is dry, but construction quality is very high. Make sure you aren't seeing the effect of the next decade down, which is being switched in series with the resistor chain that is fooling you - it's best to measure the individual resistors and make sure you understand where the wiper position is when measuring them. I've attached the data page, which includes basic schematic below.

3/ If (2) is actually a possibility, how easy is it to take things apart enough to get to any solder joints on the reverse side of the card and, indeed, are there any?

Yes, that's entirely possible, the resistor cards would have been adjusted and matched before they were soldered into the switch banks. Maybe you can peer through between the solder joints from the other side of the switch with the aid of a strong light. Unfortunately, as you can see in the above photo, some of the 'pins' of the resistor card extend into the little holes in the switch body (bottom in the photo) for mechanical support, so you would need to unsolder all of its switch stud connections to get a card out. Entirely possible with care, but you need to be sure you need to first.

As a check I soldered a 200R type RC55Y precision resistor I happened to have in stock across the appropriate switch wires and the thing now works as it should.

Hmm, it does sound as if there's a problem then, but if the switch is twitchy then I'd address that first.

[fenland787]

Hi Chris,
Thanks for the speedy (and full) reply!

The switch contacts responded well to a clean and all is now stable.

Yes, I had checked the individual sections of the suspect decade and one measured 200 ohms where all the others were 100 and adding a 200 ohm resistor in parallel with the appropriate section brought everything right.

I'll now have a closer look. I think the fault can only be that one end of one winding is just not soldered or there has to be a break in the wire - given the simplicity of the device and if I rule out the supernatural, no other failure modes occur to me!

Adrian

[Gyro]

You're welcome.

That looks pretty definitive then. Yes, it seems unlikely that it would  be a break in the wire - unless there is any visible damage (which would be pretty evident).

One thing you could try... very carefully scrape the insulation on the wire, just short of each solder joint and measure the resistance to that same stud. At the faulty one you should get a reading of 400R. If 3 joints are visible, that hopefully gives you a 75% chance without disassembly (or excessive disturbance)!

[fenland787]

Ah-ha, problem found - and fixed?

I had decided that the best way to see what was going on was to remove the switch complete - only four easy solder joints to undo - having done that I could get it under my microscope easily. 

With the light shining 'just so' the last few mm of one of the windings looked a slightly different colour under the varnish with a very slight greenish tinge. Investigation with the tip of a scalpel revealed that the last 5mm or so was no longer wire but a brittle green 'string'!

Long story short, I simply scraped back 'till I had about 3mm of silver wire, scraped the varnish off the copper 'bus-bar' next to it, re-tinned both, taped the wire so it was in contact with the bus-bar and re-flowed the joint. The result was now about 66 ohms, removed the 200 ohm resistor I'd added and bingo - 100.002 (ish!) ohms.

I'm guessing some flux was left behind originally which over a long time ate up the nichrome(?) wire.

Now just clean everything up, re-assemble and see what we've got......

Adrian

[Gyro]

Good catch! 

That's a little disturbing, I thought the varnish gave good protection. As you say, it could have been residual flux (combined with a damp environment maybe?).

Afaik, the wire is Manganin.

Chris


P.S. I'm going to have to do a paranoid inspection of mine now!

[fenland787]

Hi Chris,

Yes, looking at it the varnish there do seem to be odd gaps here and there - and quite a few entombed brush hairs! I suppose I should re-annoint the area I've worked on, I use 1B31 conformal coat here for hand coating PCBs so I could use that but perhaps I should use something more like the original, is there a 'classic' type of varnish I should be using?

Thanks for the info on the wire type, I'm sure you're right and that would explain why it soldered so well without my having to reach for my bottle of Rubyfluid flux which is full of substances nobody in their right mind would want within 100 yards of an instrument like this!

All went back together fine and, yes, I think I can pronounce it fixed and tomorrow I can get on with using it for the job I intended, re-linearising the Data Precision 8200 following replacing a duff pot!

Thanks for the assistance, I am most appreciative!

Adrian

[Gyro]

You're very welcome Adrian,

Yes, it's fortunate that you didn't need anything aggressive to re-make the joint. Regarding the varnish, I would think that using modern conformal coat would be sensible. I suspect that the original varnish is too recent to be shellac - a gentle test of solubility in IPA or meths would be a clue there. It might even be something like the yacht varnish of the day (I don't know when polyurethane came in). I would just go with a nice blob of 1B31 on the repair and any gaps. It sounds as if they had a new painter on the job -  or an end-of-life paintbrush in use, the day that card was built!

I'm glad it ended well, I hope the linearity is up to expectations, as far as I can measure, my unit (S/N 370700) comes in better than .02% ratio accuracy.

Chris

[wolfy007]

Just had this Fluke 750A Reference Divider open on my bench for a look, so took a few quick pictures.

[doktor pyta]

I found a mechanically damaged standard so I tore it down and saw something interesting.

It was made in Poland in the seventies by the 'Inco' factory.
It is 0.1 ohm, 1W in air/ 3W in oil. tolerance 0.01% Manganin.



It is built as parallel connection of two coils wound in opposite directions.
I measured the inductance and obtained 200nH.

[TimFox]

Can you tell what insulation was used?  I know that cotton and silk were common, years ago, but I don’t know what was preferable in oil.

[doktor pyta]

First of all the manganin tape is coated with laquer or enameled.
It is wound on an aluminum bobbin covered with some kind of insulating paint.
The twine is used only as a spacer and mechanical holder for the manganin tape.

[HighVoltage]

It is wound on an aluminum bobbin covered with some kind of insulating paint.

Interesting resistor, I have never heard of that company.

Are you sure it is an aluminum bobbin?
May be it is an insulator made of Aluminum Oxide?

What is its reading, is it still stable?

[doktor pyta]

Yes, double checked that. The bobbin is for sure made of Al, but the winding has some free space to move.
Value measured: 0,0999975 ohm  (setup uncertainty approx. +/-30ppm).
The insulator looks rather like enamel which is sometimes used on cooking pots

[HighVoltage]

That seems to be a very nice Manganin resistor, they usually have 10ppm/K temperature drift.
May be you can find a nice new housing for this resistor.

[doktor pyta]

New puppy in the lab.


R4030-M1 (Р4030-м1) made in USSR wirewound 1G ohm standard. Umax 2500V
Tol 0.01%, TC in my case: alpha: 8ppm/C; beta: 0 (quoting individual specs sheet).


It is made as series connection of two microwire hermetic resistors (bigger than shown on a picture) and a selected resistor or potentiometer.

[tszaboo]

Just had this Fluke 750A Reference Divider open on my bench for a look, so took a few quick pictures.

What's that block in the middle? PN184986
Is that a lot of TO-18 style package resistors in a block?

[Gyro]

Just had this Fluke 750A Reference Divider open on my bench for a look, so took a few quick pictures.

What's that block in the middle? PN184986
Is that a lot of TO-18 style package resistors in a block?

That's the container for the main divider resistors. It's oil filled (note the filler bolt at the end). The 'TO18 style' things are hermetic feedthoughs.

[antintedo]

A look inside hermetically sealed Tinsley 5685A. Photo from DOI:10.1109/CPEM.2008.4574696
100 ohm variant specified at 2ppm/year and 2ppm/K.

Bifilar winding on a fiberglass bobbin with fiberglass spacers. PTFE and bakelite top cover. Paper authors point out poor design of the terminals. Joints seem to be arc welded. Selection of two wire diameters and the way wires are bent before welding makes the process very forgiving, which could indicate they use a simple method like manually pulsed TIG.

[antintedo]

HRU standard from Alpha AE (now Vishay).
Specified at 0.1ppm/year (!) 0.05ppm/K. Offered for ~$10k.

Metal foil process, but without any substrate. Ceramic, hermetic, oil filled resistive element body. The element is free floating and only attached by one edge to the body.
Interesting construction details can be found in DOI 10.1109/TIM.2011.2108613, 10.1109/TIM.2018.2879997.

[MiDi]

Still waiting for the 10k 

For convenience with links - access to IEEE needed, but abstracts are for free:

Novel 100-Ω Metal Foil Resistor - 2011

Development and Evaluation of High-Stability Metal-Foil Resistor With a Resistance of 1 kΩ - 2015

Characterization of 1 kΩ Metal-Foil Standard Resistors and Continuing Drift-Rate Evaluation of 1 Ω and 10 Ω Standard Resistors - 2018

A set of standard resistors is a key component of resistance calibration. We have been developing a series of compact superstable standard resistors that meet calibration laboratory demands. In this paper, we describe manufacturing and evaluation of 1 kΩ standard resistors. All the resistors elements are made with the “stress free bulk metal foil” technology. We found that all standard resistors show extremely small average drift rates, e.g., smaller than 10 nΩ/(Ω year) and small temperature coefficients typically around 35 nΩ/(Ω °C) at 23 °C. Moreover, continuing the evaluation of drift behavior of 1 Ω and 10 Ω standard resistors is discussed.

8x1kΩ, 1715 days eval: 3.3nΩ/Ωyr avg (-0.8 ... 5.4 max), TC a23 35nΩ/ΩK avg (-77 ... 55 max), TC b 0.83nΩ/ΩK² avg (0.0 ... 1.5 max)
*1nΩ/Ω = 1ppb = 0.001ppm

[branadic]

With respsect to the PTB website

we meanwhile can expect a different resistor inside the wekomm resistor standards compared to the VHA518 that was seen in the early units:

https://www.eevblog.com/forum/metrology/teardown-standard-resistors/msg891676/#msg891676

-branadic-

[binary01]

We are a surely a strange bunch to be so interested in pictures of resistors!  But to me they look like fine pieces of artwork and I can appreciate the craftsmanship...
I'll add a couple of mine to the collection: 
1) Guildline 9200 Resistor / DC Shunt
Looks to be circa 1994, based on NRC (Canada) calibration sticker.

2) J. L. William (Australia) 1k Standard Resistor
I believe J. L. William was well regarded for carefully hand-made standard resistors with low TC in a since-gone era of Australian manufacturing.  This one was originally used by NMI Australia.

I have some others but they have been covered already in this thread.

[tszaboo]

ELC DR-08 decade box, 8 decade, 10MOhm to 0.1 Ohm
1% nominal accuracy


The gray beards soldered 3 SMD parts, otherwise done by the machine.
Banana for scale.

[trobbins]

Ok not in the same league as the GR1433 in post #170, and a wee bit older, and the subject of some abuse, but can anyone identify the type of wire used in the Gambrell Bros 4x decade resistance box (photo below) ?  The wire can make a joint in solder, and has a form of insulation covering.  All coils are wound on the same size wooden bobbin, and are wound bifilar.  I'm guessing 1930's but not much info available on Gambrell except for them exhibiting their instrument range in the 1920's and 30's, and they are from England.  Nearly every bobbin measures within about 0.1% (only a crude measurement so far with a Keithley 197) so this was likely a base grade lab 1R to 9,999R box for its time.

[Edwin G. Pettis]

The wire is Manganin or a similar alloy, wood bobbins would be somewhat out of the ordinary as ceramic was the standard bobbin material even back in the 1920s.  Constantan was another alloy commonly used but it had a higher TCR than Manganin (1892) and generally fell out of use as a precision resistor alloy but is still used today.  Both alloys can be soldered.

[serg-el]

YEW Type 2792 10 kOhm 1977

[Henrik_V]

Since the ESI DB62 already showed up here, and I just got my hands on three of them.. hold on , one is different, looks custom..
called SP 3972

pics to follow

[Henrik_V]

The 10 Megs are 005, but no point , the 1Megs are market with +/- .003% , so I assume the 10 Megs are +/- 0.005% 

[ZhuraYuk]

New puppy in the lab.

R4030-M1 (Р4030-м1) made in USSR wirewound 1G ohm standard. Umax 2500V
Tol 0.01%, TC in my case: alpha: 8ppm/C; beta: 0 (quoting individual specs sheet).

It is made as series connection of two microwire hermetic resistors (bigger than shown on a picture) and a selected resistor or potentiometer.

Here is how they actually look inside of P4030, without M1 letter.

[HighVoltage]

Thanks for sharing the inside.

So, there are 2x 500 MOhm resistors in series with a small calibration resistor.
Can you measure the resistance of the small resistor?

[ZhuraYuk]

Thanks for sharing the inside.

So, there are 2x 500 MOhm resistors in series with a small calibration resistor.
Can you measure the resistance of the small resistor?

Small one is 603.187 KOhm, big ones are 500 MOhm with precision 0,1%. Resistors model is МВСГ (MVSG) but the label is made of paper, such resistor are custom ones and are selected by tempco and can be created with non standard value. USSR industry was creating МВСГ resistors with printed labels and those are standard values and tempco.

[Nanitamuscen]

Resistance coil R310 0.01 Ohm USSR.
The inside is filled with oil.
https://drive.google.com/drive/folders/1tfFxWcOP6kJ6ihT4HXamelHuIbY_r7rb?usp=sharing

[donlisms]

Here's one of the ubiquitous Leeds & Northrup 4000-series resistors;  this one is 10 ohms.  It arrived leaking oil, and the cap was not difficult to unscrew, so I carefully opened it up.

A simplified build order must have been something more or less like:

• Varnish the 30mm brass tube;
• Tie the folded end of the Manganin near the top with (probably) silk thread;
• Wind one layer all the way down, and a second layer part way back up;
• Tie off the end of the winding with silk thread;
• Tie two varnished flat brass strips up along the both sides of the coil;
• Attach one winding end near the bottom of each strip, and a short heavy wire from the terminal near the top;
• Connect those two with (probably) another piece of Manganin;
• More silk thread as necessary;
• Solder those joints;
• Measure, and file the side wires to tune the resistance;
• Measure again, and so on;
• Seal it all up with red paint or shellac or something;
• Fill the can with oil, and tighten it all up.

You can see the 12 little marks where they grabbed the can with some kind of chuck internally at the open end and cut the threads.

The red paint flaked off the silk thread and settled to the bottom of the can.  I will be cleaning all that out and replacing the oil.  I also need to find a sealing method for the lid; it was recommended (at least in the early days) to fasten the lid with shellac; my other resistors really don't want to unscrew.

[donlisms]

More pictures, because... limits!

[donlisms]

And this one... 1 ohm, somewhat historical, from 1930.  This is the earlier earlier model, with a hard rubber top.  It's not exactly as Rosa described, because the hard rubber top does not screw on, it just fits tightly, and the thread extending out the bottom of the central tube screws into a nut soldered onto the inside of the floor of the can, pulling the whole assembly down tight.  Well, more or less tight -- the can is about half full of oil.

No red varnish or paint or whatever, though it's much darker inside than the newer resistors.

This one is two bifilar winds, both wound together in parallel, starting from the same point.  One layer.  The two pairs of winding ends are brought all the way up to the top of the side rails, possibly with a separately-joined piece of Manganin or whatever.  You can see the abrasive tuning on this one, too; it's a bit more aggressive than the 10 ohm resistor.

I like it because it's a little bit historical: baked in 1930, sealed or checked in 1931, nice temperature coefficient (at least as they measured it at the time), and owned by the National Research Council.  I don't think there were all that many of these resistors around at that time.

And this year-long wait after baking is what makes me nervous when you guys put your resistors through some kind of heating adventure, and then wait 15 minutes before measuring!   

[donlisms]

By the way, the 1930's 1 ohm and newer 10 ohm both measure very closely to my best other standards, within a fraction of a milliohm. 

More of the old one ohm:

[Conrad Hoffman]

Can't be sure but the red stuff is usually Glyptal. The stuff goes back a long ways, patented in 1914 and used in paint 1930s. Good properties and I'd never replace it with anything else during a rebuild or spruce-up.

[MegaVolt]

A set of wire-wound!!!!! resistors of 1 GΩ.

[MegaVolt]

Continuation

[bobAk]

Mgsv(мгсв) did not differ in reliability with poor storage.  So it's better to check, perhaps they have changed their characteristics a lot.  Pay attention to the yellow insulators - amber, most likely.

[MegaVolt]

Mgsv(мгсв) did not differ in reliability with poor storage.  So it's better to check, perhaps they have changed their characteristics a lot.  Pay attention to the yellow insulators - amber, most likely.

Yes. This is amber. Lots of amber )

This is a measure of resistance Р4043 (cyrillic)

Resistors are hermetically sealed from thin wire covered with glass. It takes a lot of effort to ruin them. All of them seem correct at first glance. Accurate measurements of 1 GΩ are not so easy


Here are more photos.

[MegaVolt]

...

[MegaVolt]

...

[Conrad Hoffman]

I figure a 1 Gohm resistor would have enough wire to circumnavigate the earth!

[MegaVolt]

I figure a 1 Gohm resistor would have enough wire to circumnavigate the earth!

https://ru.wikipedia.org/wiki/%D0%9C%D0%B8%D0%BA%D1%80%D0%BE%D0%BF%D1%80%D0%BE%D0%B2%D0%BE%D0%B4

The thinnest wire has 15 kOhm/m. This is somewhere 67km of wire

[chekhov]

One more decade resistor box - Р4078, 100 Meg/decade

It pretends to be quite hermetic, screws that hold 'legs' are sealed with paint from the internal side.
On the back there is a cap that holds plastic compartment where little pack of silica-gel is stored.

[babysitter]

P4078 - things with faces!

[Nanitamuscen]

A resistive element from the resistance standards produced by the Krasnodar plant of measuring instruments.
It is made of nichrome foil, sintered with a substrate.

[alm]

A resistive element from the resistance standards produced by the Krasnodar plant of measuring instruments.
It is made of nichrome foil, sintered with a substrate.

Thanks for the picture! What kind of resistance standard would contain this element? The only Krasnodar standard resistors I'm finding on Google are the Soviet-era wirewound resistors like P321, but that's probably because I'm not searching in Cyrillic.

[Nanitamuscen]

P3030, P3026 and many others  (р3030, р3026 Cyrillic)

[bsw_m]

The only Krasnodar standard resistors I'm finding on Google are the Soviet-era wirewound resistors like P321, but that's probably because I'm not searching in Cyrillic.

As an example: https://www.ebay.com/itm/334903961471?hash=item4df9d4e77f:g:3YkAAOSwENZd4SpX

[bsw_m]

Microprovod R4033 100MOhm standard:
The standard was purchased in NOS condition.
But it has a problem with the large value of the standard deviation. For this reason, it was disassembled in the hope that the problem was with the additional resistance. But alas, the main resistive element turned out to be problematic here.

Measurement of this standard:
Total number of comparison = 25
Rx value peak to peak = 15.96 ppm
Rx value standard deviation = 4.96 ppm

And the measurement of another good R4033 standard:
Total number of comparison = 25
Rx value peak to peak = 2.76 ppm
Rx value standard deviation = 0.88 ppm

Measurement of problematic main resistive element:
Ambient T = 25.7
Total number of comparison = 15
Rx value peak to peak = 12.65 ppm
Rx value standard deviation = 3.91 ppm
Rx value = 99.86598 MOhm

The main resistive element has  alpha=+2ppm, beta=-0.4ppm and -133kOhm from 100MOhm this data from the paper on it.
The installed potentiometer can adjust the rating within 200 ppm

[0.01C]

ESI SR1050 1M

[0.01C]

[ Specified attachment is not available ]Kelvin-Varley Voltage Divider Model 2891

[0.01C]

Vishay Bridge Module Model 1109

[chekhov]

P3030, P3026 and many others  (р3030, р3026 Cyrillic)

Here is an example of slightly beaten 1R Р3030 resistance standard that was opened for verification

[TimFox]

On your last photo (the resistive element), are any features of the visible pattern used during construction for trimming the value?

[chekhov]

It's hard to look deep inside, and I cannot disassemble it further, so 'scratch' in the center of this image is the only visible trim that I see clearly.
Also, by how light is reflected from surface, there is one more trim at the central line in the middle of the 'die', and two or there little ones in other places.

[TimFox]

That "scratch" could be a trim made at the factory.
I have seen similar trim marks on Vishay foil resistors.
The features to which I referred are the geometric white lines that seem to bound areas of the element.

[Nanitamuscen]

The features to which I referred are the geometric white lines that seem to bound areas of the element.

You look at my post earlier in this thread, it shows exactly the same resistive element, but at a higher resistance.

[branadic]

Not a teardown, more like a tearup sort of thing, DIY hermetically sealed resistor filled with mineral oil. We can choose between Fluke-like brass or red gold (copper).

-branadic-

[r6502]

hello branadic,

interesting , where did you get the feet trough's?

BR Guido

[branadic]

Bought off ebay long time ago. Attached are the old links:

https://www.ebay.de/itm/10pcs-Glass-Hermetic-1-Pin-Feedthrough-Feedthru-Military-Product/323116442647
https://www.ebay.de/itm/10pcs-Glass-Hermetic-1-Pin-Feedthrough-Feedthru-Military-Product/323116442642

-branadic-

[Gyro]

They originate from the former Soviet Union (NOS military). I purchased some similar ones around 10 years ago. As far as I can see, supplies seem to have dried up on ebay, but that may be due to the current situation. This type is useful because the wire can pass straight through them, removing any additional thermocoupe effects.

[chekhov]

One more decade resistor, Р40106 this time. 10^7 and 10^6 in a single box.

[Vgkid]

One more decade resistor, Р40106 this time. 10^7 and 10^6 in a single box.

Thanks for the teardown . I find it interesting that the smaller resistors are not always connected to the larger resistors. I guess it is a sot trim resistor.

[chekhov]

Sorry, haven't scanned schematics in good quality yet, but indeed I missed more description.

All little resistors are indeed trim resistors, but the ones without connection are actually back-up 
All resistors have the same value 2M or 20M, but in series or in parallel connection allows to form all 1-10 values, so if some of them go off limit (which happens pretty frequently for these MRX resistors with time) or they just burned by too high voltage, they could be replaced by these pre-selected pairs in the field (there was not even any protective label on bottom cap - open and nobody notices).

[KirstyA]

Hello Everybody,

I have just acquired this resistor as part of a set I am building up.
Unfortunately this one appears to be open circuit which is very disspointing as it has the last measured value on it as well.
Can I get this resistor apart to inspect/repair it, or is it soldered together?

[alm]

I don't know this specific type, but with the similar looking Rosa type resistors, the top unscrews. Early ones were press fit I believe. I've also seen resistors with a small screw near the top of the can. Usually these standard resistors, unless hermetic, can be disassembled for oil replacement.

[donlisms]

The top is pretty clearly a different piece, and it will have had to be held on somehow.  Not only that, but (almost certainly) it has to be a method accessible from the outside, so if it's soldered, you should be able to see some clues - soldering is never quite that perfect.

Unless... the bottom comes off. More pictures would be helpful!  But just continuing to look and think, looking for tiny clues, might yield your answers.

I'm kind curious about that screw on top.  None of my Leeds and Northrup or Honeywell resistors have such a screw.  Ground lug?  Structural?  Pressure thing?  Hm.

It is kinda pretty, though!  Is that gold paint?

[guenthert]

I thought that screw might be connected to an adjustable resistor, trimming the total, but I was mistaken.  I think I've seen such with resistors made by -hp- and Tinsley, but at least for the Tinsley 5695 it is a connecting point to the shield (guard), see http://www.tinsley.co.uk/wp/downloads/Tinsley_5685-and-5695.pdf

[KirstyA]

I don't know this specific type, but with the similar looking Rosa type resistors, the top unscrews. Early ones were press fit I believe. I've also seen resistors with a small screw near the top of the can. Usually these standard resistors, unless hermetic, can be disassembled for oil replacement.

I'm beginning to think press fit as well

[KirstyA]

The top is pretty clearly a different piece, and it will have had to be held on somehow.  Not only that, but (almost certainly) it has to be a method accessible from the outside, so if it's soldered, you should be able to see some clues - soldering is never quite that perfect.

I'm kind curious about that screw on top.  None of my Leeds and Northrup or Honeywell resistors have such a screw.  Ground lug?  Structural?  Pressure thing?  Hm.

It is kinda pretty, though!  Is that gold paint?

It's a dual can assembly. Looks like the two cans are soldered from the bottom, but I don't see any solder on the top, just what looks like a possible red gasket. Presumably meaning a press fit of the top and coil assembly into the two cans?

The screw on the top you can remove and change the oil as this resistor is oil filled.

This one has brass arms and copper vertical inserts. Some of the others I have in the set seem to have copper arms.

[KirstyA]

The bottom of the Cambridge is soldered, what about the top?

The top is soldered. But the connection bar seals are far from hermetic. The Cambridge type is oil filled, not air, so there is also an oil screw at the top plate.

Looking back right through this thread it seems this Cambridge resistor is soldered top and bottom, so not easy to get into without wreaking it.
I'm suspecting somebody overloaded it and burn it, although I find that hard to believe since it is oil filled and therefore would have had to have been a big overload because of the oil cooling?
It's a shame because it's part of a set of 5 I bought from 1 Ohm to 10K, all with UKAS cal stickers on and all with their actual measured value on as well and although the values wouldn't be completely accurate now, (UKAS cal lab measured in 2007), they would have been more than accurate enough for me.
Looks like I am going to have to return it for a refund.

[trobbins]

By the way, what resolution did the latest measured value provide, given say a nameplate of 10,000 ohm would indicate within 0.5 ohm.

[Victorman222]

Here is a teardown of a soviet decade resistance box R327 (Р327 cyrillic) made at the Krasnodar ZIP plant. It boasts an accuracy class of 0.01%. This one was assembled from two slightly broken units.

[Victorman222]

Close up photos of the resistors

[Gyro]

Nice switches. Those butt joints on the lowest decade look 'interesting'. Presumably they are spot welded but I don't think I'd like to put any pressure on them.

[Conrad Hoffman]

"Interesting" is a good description. I think they're hard soldered (silver brazed?) but could be something else. Doesn't seem like a great method but probably works just fine.

[bsw_m]

Interesting find. Standard resistor R3030/2 produced by the Krasnodar ZIP plant. There is no information about them, and they seem to be experimental samples.
The accuracy class of these standard resistors is 5ppm.
I compared them with each other, the difference was less than 3 ppm.
A little later I will measure the true values of these standard resistors.

[manganin]

Standard resistor R3030/2 produced by the Krasnodar ZIP plant. There is no information about them, and they seem to be experimental samples.

Small size secondary standards based on Soviet bulk metal foil technology.

[bsw_m]

Measurement results for these standard resistors (See attached photos).
All standard resistors are out of tolerance.
They also have a very high TCR (about 10ppm/K)
Conclusions - these standard resistors are interesting artifacts from the past, but nothing more.

[HighVoltage]

Measurement results for these standard resistors (See attached photos).
All standard resistors are out of tolerance.
They also have a very high TCR (about 10ppm/K)
Conclusions - these standard resistors are interesting artifacts from the past, but nothing more.

What instrument is it that you using to measure these resistors in your pictures?

[bsw_m]

What instrument is it that you using to measure these resistors in your pictures?

Measurement International 6242B Resistance bridge.

[manganin]

All standard resistors are out of tolerance.

The marked "Class" should be considered as an initial adjustment accurary rather than the long term stability. The tight tolerance comes from the very easy trimming of the bulk metal foil element - compared to the Krasnodar wire wound standards, which were generally more stable.

I think the resistors have performed surprisingly well, age considered.

Also the direct measured value doesn't tell the whole truth, you need to include the change of the Ohm as well as the original calibration uncertainty in your calculations.

[Victorman222]

Might be a bit off topic but here is a teardown of a soviet hermetic metal foil resistor S5-61 (С5-61 Cyrillic). There resistors were made with tolerances up to 0.005% and TCR grades 30/20/10/5 ppm/℃ in temperature range from 20℃ to 125℃.
This resistor had two chips but the second "trim" chip was cracked and not bonded. It's a 340R 0.05% 30ppm/C resistor.
The last photo is of a S5-62 resistor, that one is the caseless version, but the ones i got were destroyed by moisture and decomposing foam. The value is 6.42K 0.05%.

[TUMEMBER]

What instrument is it that you using to measure these resistors in your pictures?

Measurement International 6242B Resistance bridge.

Respect if you operate such a machine.
A question like this: do you "check the entrusted material"?
I recently came across something like this on Aliexpress
https://pl.aliexpress.com/item/1005005989851239.html?spm=a2g0o.cart.0.0.11c5452cIMCvgC&mp=1&gatewayAdapt=glo2pol

but I have no way to verify whether what they declare is reality or marketing.
It looks cheaper than the famous Vishay and I am tempted by the 200 ohm, 2 kohm, 20 k ohm, 200 kohm series.

[bsw_m]

but I have no way to verify whether what they declare is reality or marketing.

I don't have these resistors and can't say anything about their performance.

[TUMEMBER]

but I have no way to verify whether what they declare is reality or marketing.

I don't have these resistors and can't say anything about their performance.

Thank you for your answer 

[Kosmic]

What instrument is it that you using to measure these resistors in your pictures?

Measurement International 6242B Resistance bridge.

Respect if you operate such a machine.
A question like this: do you "check the entrusted material"?
I recently came across something like this on Aliexpress
https://pl.aliexpress.com/item/1005005989851239.html?spm=a2g0o.cart.0.0.11c5452cIMCvgC&mp=1&gatewayAdapt=glo2pol

but I have no way to verify whether what they declare is reality or marketing.
It looks cheaper than the famous Vishay and I am tempted by the 200 ohm, 2 kohm, 20 k ohm, 200 kohm series.

Check out this article: "Temperature drift test results of RJ711 and other plastic metal foil resistors" https://www.kechuang.org/t/89682

[TUMEMBER]

What instrument is it that you using to measure these resistors in your pictures?

Measurement International 6242B Resistance bridge.

Respect if you operate such a machine.
A question like this: do you "check the entrusted material"?
I recently came across something like this on Aliexpress
https://pl.aliexpress.com/item/1005005989851239.html?spm=a2g0o.cart.0.0.11c5452cIMCvgC&mp=1&gatewayAdapt=glo2pol

but I have no way to verify whether what they declare is reality or marketing.
It looks cheaper than the famous Vishay and I am tempted by the 200 ohm, 2 kohm, 20 k ohm, 200 kohm series.

Check out this article: "Temperature drift test results of RJ711 and other plastic metal foil resistors" https://www.kechuang.org/t/89682

And thank you, I probably would never have found it on my own. After all the effort of Google translate, his - is it kenji?

Fortunately, there are pictures and it's somehow easier to "understand" 

[Kosmic]

What instrument is it that you using to measure these resistors in your pictures?

Measurement International 6242B Resistance bridge.

Respect if you operate such a machine.
A question like this: do you "check the entrusted material"?
I recently came across something like this on Aliexpress
https://pl.aliexpress.com/item/1005005989851239.html?spm=a2g0o.cart.0.0.11c5452cIMCvgC&mp=1&gatewayAdapt=glo2pol

but I have no way to verify whether what they declare is reality or marketing.
It looks cheaper than the famous Vishay and I am tempted by the 200 ohm, 2 kohm, 20 k ohm, 200 kohm series.

Check out this article: "Temperature drift test results of RJ711 and other plastic metal foil resistors" https://www.kechuang.org/t/89682

And thank you, I probably would never have found it on my own. After all the effort of Google translate, his - is it kenji?

Fortunately, there are pictures and it's somehow easier to "understand" 

The auto translater in Edge says it's "Simplified Chinese". Try this link instead: https://www.eevblog.com/forum/metrology/rj711-foil-resistors-from-china/?action=dlattach;attach=1946766

[C-47]

Adding a recent acquisition to this post, a J.L William 1 Ohm after it made a journey across Australia, where unfortunately it lost quite a bit of oil, question is, which is better, Paraffin or White Mineral Oil (food grade), as both of these are locally available and to what level should it be refilled.
After opening up to inspect the internals as the readings obtained were not stable, it was noted that winding #2 (from top) has been replaced at some time, noticeable due no varnish and newer looking twine, should varnish be applied considering its already oil soaked, if so, any leads on cleaning.
After the coil unit has been out of the body housing for a couple of days the readings have somewhat stabilised using a 4 wire setup into a calibrated HP34401A.
Any advise appreciated. Thanks, Alan.




 

[trobbins]

I would have thought that a transformer grade mineral oil would have been used, due to its high resistance and stable performance and general use in other standard resistor metrology devices (even though a 1 ohm resistance is unlikely to be affected by distributed oil resistance like say a 100k device).

Imho the repair looks like it was done by the manufacturer, given the same wiring foldback technique and some obvious trimming.  I wouldn't think varnishing are the use of non-standard transformer oils is a good path to follow. 

The instability may be more related to the busbar bolting, and soldering of the wire elements to busbaring, but the photos don't really show that continuity region.  What variation were you seeing with the 34401A ?

[C-47]

Thanks for the input, the variation was initially in +/- 100th of an Ohm, but this has since reduced back considerably just using the alligator clips of the leads, probably should be using proper terminal spades. The photo shows the connections, while #2 coil lead looks like its not as good as the others. Will look into transformer oil.

[trobbins]

Ta - that's a better photo for awareness of the electrical joints.

Maybe the variation was thermal EMF related - was it drift or fluctuating ?  You would have 1mA test current, so 1mV FS reading.  Can you confirm acceptable shorted measurement value using test leads to one of the device terminals?  You may be able to log over a day to get a feel for the tempco, given it is now in free air.

It's not easy to get a small commercial quantity of transformer oil in oz - I looked last year.  Perhaps ring around any remaining transformer manufacturer/service/recyclers in your area - they may have a drum out the back.

[C-47]

Ok thanks, will see what I can find re power supply and getting readings, they were fluctuating rather than drifting, I'll see if I can get better connection setup at the end of the 4 wires. Have to be tomorrow as work intervenes.

[guenthert]

Thanks for the input, the variation was initially in +/- 100th of an Ohm, but this has since reduced back considerably just using the alligator clips of the leads, probably should be using proper terminal spades. The photo shows the connections, while #2 coil lead looks like its not as good as the others. Will look into transformer oil.

    The -hp-34401A is a fine bench multimeter, but it's not ideally suited to measure low value resistors.  It's lowest resistance range is 100Ohm, so 100th of an Ohm is only 100ppm of its range which is outside of its comfort zone.  You might get more consistent results with a Kelvin bridge configuration or if you happen to have a stable current source (some 10 to 100mA), use that and measure the voltage drop across the resistor instead.

    And yes, copper spades or bare solid wire should get better connections than alligator clips (which might also introduce fairly high thermal EMF).

[HighVoltage]

If it was really transformer oil inside this resistor, be careful. Old transformer oil can contain polychlorinated biphenyls (PCBs) and other toxic chemicals that can be harmful to human health. If possible, stay away from transformer oil, which is also not needed here at all.

[Gyro]

Here's the definitive standard resistor filling thread, well as definitive as it's got on here anyway...  https://www.eevblog.com/forum/metrology/oil-for-a-leaking-tinsley-standard-resistor/

[Kosmic]

In Fluke Calibration Philosophy in Practice, page 14, they mention "light mineral oil" when describing NBS style resistor:



You're going to laugh, but a cheap source of good quality mineral oil is actually horse laxative 

[C-47]

Thank you to all for input and suggestions, indicates that I have a bit to learn when dealing with low value Ohms, but as this is a hobby to keep me out of the pub at night, wife reckons the pub maybe cheaper.
Anyway a quick reading shows its now in the ballpark of 1 Ohm, will pickup some White Mineral Oil on Monday, any left over I can use to clean the horse out  . Thanks to all, including beanflying, useful input.

[beanflying]

Sweating for the cause. 36C 'outside' and topping 40C in the shack and climbing 

My Williams 0.1 Standard plugged into my 34461A on 4W is there or thereabouts. To hot to move the Quadtech milliohm meter to the bench or to pull mine down for a comparison.

Motivation ceases to the rest of the day.....

[C-47]

Crikey, didn't think it got so hot there, more amber liquid refreshment needed !!