Manufacturing custom high-precision low-TC resistors

[TiN]

See picture attached... no change in temperature but change in resistance is a hint for a lack in temperature.

I'm lost with term "lack in temperature". Temperature need to be higher? By looking at time-domain graph it's easy to understand why there was a difference in resistance at +60C point.

It's pretty easy to miscalculate such paramter using the time domain data as high dynamic range of it hides true behavior.

Very true and correct, it's easy to miscalculate complex parameter, such as tempco (which is always combined figure of resistance tempco, setup quality, measurement noise, hysteresis and so on).
Representation format or axis definition do not make a difference. We can shuffle numbers and pretty plots, all we want, but end result is going to be same - complex equation with many unknowns (which effect we try to reduce when measuring interested parameter, with different level of success).

I don't expect anyone to calculate, I just present the data I got, that's it. I was interested to see if such DIY resistance device can be stable over time and small temperature span, not so much the absolute tempco itself. I got my answers, that it is not the case in this particular example, confirming my own experiments years ago, since which I decided it's not worth the time to try own DIY resistors. If you expected me to present traceable tempco calibration data with uncertainties and all, well, sorry, that's not happening. 

P.S. for all tempco numbers I ever post box method is used.

[branadic]

I was interested to see if such DIY resistance device can be stable over time and small temperature span, not so much the absolute tempco itself. I got my answers, that it is not the case in this particular example, confirming my own experiments years ago, since which I decided it's not worth the time to try own DIY resistors.

A conclusion based on one measurement with only one ramp up and down.

-branadic-

[Edwin G. Pettis]

Ramon:

That spool you acquired appears to be an older spool from Kanthal (possibly from the 1970s), I checked the spools I have in stock and even the older ones from the 1970s indicate clearly whether they are bare or enameled, the reason being that the spools used to be supplied in paper boxes with a clear window in the end for the spool label.  One did not want to have to open the boxes to see whether or not the wire was bare or enameled.  The boxes on my newer spools are plastic so you can see through but the labels still indicate bare or enameled as the boxes are almost always stacked so you still need the information on the label.  The spool's TCR is accurate, I've only had one incident in many years where the spool's TCR was wrong and verified as wrong.  I have one spool of wire with a similar TCR (not exactly the same) and it does indeed produce resistors with TCRs within the specified range if, and this is a very big if, it is made correctly.  All of my resistors are very close to the TCR printed on the spools unless something has gone wrong in production, which it can, there are a lot of variables to deal with and the resistance wire terminations are just one of a long list that affect the end product.

Cupron wire does indeed withstand high temperatures, well above the +140°C you indicated, in large gauge sizes, Cupron is used for heating coils, while Evanohm can operate at high temperatures, its TCR increases above approximately 150°C and if operated at much higher temperatures, the TCR can be changed.  That is why PWW resistors generally have +150°C as their maximum operating temperature (and that is at zero power mind you).  You are quite correct that Manganin (and derivatives) has a somewhat low operating maximum temperature of about 60°C, but you are also assuming that the 15 PPM/°C spec (maximum by the way) is what you get, it can actually be anywhere within the 15 PPM spec, even close to zero.  Yes it is very sensitive to processing, the act of soldering will change the local characteristics and keeping it within 15°C to 35°C makes it happiest for temperature.

Cupron is generally the alloy of choice for power resistors since it tolerates the +350°C maximum temperature without big changes in characteristics, you can't do that with Evanohm and keep its low TCR intact.

I can state with complete confidence that over the years a great many variations have been tried for terminating resistance wire to various types of lead assemblies, in my published paper, The Last Half Century: Wire Wound Resistors (in EDN, also posted here in another thread), we detail how the industry struggled with the new Evanohm alloy.  You are merely repeating history here and while I encourage it as a learning experience, you are not going to produce high grade precision resistors with any DIY methods unless you've got the money to setup the correct equipment and materials and that isn't going to be cheap.  I agree with TiN's assessment (and not just because I make PWW resistors), it has been tried many times with the same results, frustration and pretty much wasted time and money trying to reinvent the 'wheel'.

[branadic]

You are merely repeating history here and while I encourage it as a learning experience, you are not going to produce high grade precision resistors with any DIY methods unless you've got the money to setup the correct equipment and materials and that isn't going to be cheap.  I agree with TiN's assessment (and not just because I make PWW resistors), it has been tried many times with the same results, frustration and pretty much wasted time and money trying to reinvent the 'wheel'.

You force people to repeat history and reinvent the wheel, cause you don't want to share your "secret knowledge". I do understand that you live from selling wirewound resistors and thus don't want to share your solution. But by the looks of it you try to stop people from investigations on that topic, which seems like you are scared that someone can come up with a proper solution. And if this solution is shared in the web those guys won't buy wirewound resistors anymore. Just my personal impression also based on statements like this:

"...95% of what they're doing is nothing short of a waste of time and pointing it out to them will get you nowhere. They don't know the minutia of the components nor how to properly do tests and when to do the correct tests, they end up chasing their tails around in circles and can't understand why..."

I can remember a time, when people couldn't afford soldering irons nor soldering stations for home lab, when they couldn't afford a scopes analog or digital. I'm sure there will be a time, when you can produce precision components at home. Digital process chain such as Inkjet printing is getting better and better, first Inkjet printers enter homelabs. It's just a question of time and noone will stop this development.

-branadic-

[zhtoor]

You are merely repeating history here and while I encourage it as a learning experience, you are not going to produce high grade precision resistors with any DIY methods unless you've got the money to setup the correct equipment and materials and that isn't going to be cheap.  I agree with TiN's assessment (and not just because I make PWW resistors), it has been tried many times with the same results, frustration and pretty much wasted time and money trying to reinvent the 'wheel'.

You force people to repeat history and reinvent the wheel, cause you don't want to share your "secret knowledge". I do understand that you live from selling wirewound resistors and thus don't want to share your solution. But by the looks of it you try to stop people from investigations on that topic, which seems like you are scared that someone can come up with a proper solution. And if this solution is shared in the web those guys won't buy wirewound resistors anymore. Just my personal impression also based on statements like this:

"...95% of what they're doing is nothing short of a waste of time and pointing it out to them will get you nowhere. They don't know the minutia of the components nor how to properly do tests and when to do the correct tests, they end up chasing their tails around in circles and can't understand why..."

I can remember a time, when people couldn't afford soldering irons nor soldering stations for home lab, when they couldn't afford a scopes analog or digital. I'm sure there will be a time, when you can produce precision components at home. Digital process chain such as Inkjet printing is getting better and better, first Inkjet printers enter homelabs. It's just a question of time and noone will stop this development.

-branadic-

hello,

i don't think Mr. Edwin is forcing anybody to do anything. i think he is only pointing out the difficulty
of the process, and graciously so, if he is not willing to share his "secret knowledge", he is perfectly
within his rights to do so, since he has spent decades acquiring the necessary skills, and is not willing to
give it away for free (to un-grateful buggers like me!), which is his perfect right and sweet will.

whosoever wants to acquire the necessary skills should either work for it, or buy it, period.

as far as giving away your knowledge for free, it is anybody's sweet will, and i for one am
completely and utterly grateful for the knowledge i acquire, courtesy of all people on this
forum.

let us have a positive attitude and be thankful for the *free-bees* we get from
being members of this enlightened community.

as far as new developments are concerned, yes, time moves on, and low-tech countries become
developed, developed countries become poppers, these are historical cycles, and only the great
ones become timeless, and great ones are pretty rare.

regards.

-zia

[JoeN]

To M.M.:  I was replying to JoeN, as he did not mention how he connected the wires to leads (he didn't say what his leads were either), since it was very unlikely that he had an arc welder, my reply was geared to the more common CD welders which are more commonly available and just plain soldering.  If you have read my other postings you'll know exactly what I've said before (many times) about welding resistors.

I have not indicated any technique yet because I haven't tried to build my own resistors yet.  I am just gathering information to see how feasible it would be to do.  But no attempt has been made at this point.  Thank you all for your replies to this post so far.  I am reading the other indicated thread and some of Edwin's writings I have found on the internet to improve my overall understanding of this subject and why it is so difficult to make high-precision and low-tempco resistors.  Quite a bit more involved than it might be expected and it sounds like millions and millions of man hours has already been dedicated to this art.

[Echo88]

I agree with zhtoor: Lets be grateful that experienced people like Edwin, TiN and Mister Diodes spend their time here and give hints how to do it. You can not expect them to write all their knowledge down and give it away for free.
If you expect such a behavior i can point to lymex who did an absolutely remarkable job in sharing his knowledge on http://bbs.38hot.net/thread-485-1-1.html about a wide range of thinks including rare beasts like DCCs, Warshawsky Bridges and other things. A pleasure to read, just a little bit tiresome due to the sometimes crude translation with google translate.

[ramon]

That spool you acquired appears to be an older spool from Kanthal (possibly from the 1970s)

I really thought it was quite new ! I compared several pictures of many different spools and the one I bought didn't seem like 70s style. What do you think is the date of this other spools? (this is what is currently available at auction sites)

Even the 2012 catalog from Kanthal shows the picture of my spool (please, check the smallest spool on page #3). https://www.kanthal.com/globalassets/kanthal-global/downloads/materials-in-wire-and-strip-form/resistance-heating-wire-and-strip/s-ka026-b-eng-2012-01.pdf

Oh my good, those sweeden guys really love retro style. (The more I look at the catalog picture, the more I think they made the picture with some wood camera and kodachrome film).

I don't pretend to reinvent the wheel, this is just like any other hobby. Some people collect stamps, or coins, or even insects. I just enjoy winding wire on weird substrates and check resistance value with a meter (with needle or leds). It takes little time, almost no money, don't need that much space, and is relatively safe for childrens ... and It's funny !!

Please, could you help me to buy some bobin of cupron (or anything else)? I havent found any on the action site, and no manufacturer is willing to sell me.

[branadic]

Noone wants Edwin to offer his knowledge here, all I'm asking for is to not stop people from investigating on their own.
And never underestimate the power of curiosity and analysis capabilities. It took a minimum amount of time using EDX to get a material composition of the intermediate material used in his resistors.

-branadic-

[zhtoor]

Noone wants Edwin to offer his knowledge here, all I'm asking for is to not stop people from investigating on their own.
And never underestimate the power of curiosity and analysis capabilities. It took a minimum amount of time using EDX to get a material composition of the intermediate material used in his resistors.

-branadic-

hello,

just because you know the composition of the intermediate wire using x-ray spectroscopy,
would you be able to supply PWW resistors at a cost lower than Edwin?, i think not.

i am pretty sure a *lot* of work goes into making ppm level stability PWW's.

regards.

-zia

[branadic]

just because you know the composition of the intermediate wire using x-ray spectroscopy,
would you be able to supply PWW resistors at a cost lower than Edwin?, i think not.

Who would have thought that people are able to have voltage references at home? Who would have thought that people do have the possibilities to weld at home? Who would have thought that people do have inert gas and nitrogen at home? Who would have thought that people do all the metrology stuff you guys are presenting here?
You think it is air that you are breathing? Okay, you better take the blue pill

-branadic-

[Magnificent Bastard]

It took a minimum amount of time using EDX to get a material composition of the intermediate material used in his resistors.

-branadic-

Are you going to share this info?

[Magnificent Bastard]

just because you know the composition of the intermediate wire using x-ray spectroscopy,
would you be able to supply PWW resistors at a cost lower than Edwin?, i think not.

Who would have thought that people are able to have voltage references at home? Who would have thought that people do have the possibilities to weld at home? Who would have thought that people do have inerat gas and nitrogen at home? Who would have thought that people do all the metrology stuff you guys are presenting here?
You think it is air that you are breathing? Okay, you better take the blue pill

-branadic-

Agreed.  But also, since Edwin is making resistors for profit, he has to keep the material and labor costs low so he can sell his resistors at a reasonable price.  There can be no doubt that there were compromises made in his resistor designs to hit a price target.  Conversely, a hobbyist can take all day long (if we want to) to make just one resistor, and since (we) are only making one (or a few), more expensive (even exotic) materials can be used, where that would be totally impractical for a resistor production line.  This means that (we) can make resistors even better than Edwin's (and even better than VPG foil resistors), if enough time and effort is put into it-- and this process will go much faster if we share information.  PWW resistors are an old technology-- so there is a lot of information available (for free) from the various patent filings and also from the many national metrology laboratories.  If you have access, there are also many peer-reviewed scientific papers available on this subject.

[branadic]

Are you going to share this info?

Not in detail, all I say is: It's a NiCr alloy.

-branadic-

[RandallMcRee]

Back to the basics--

Why could not the wire be terminated by some mechanical means? Some sort of screw terminal?

(No one would want this in production, but should be ok for hobbyist use, surely).

[amspire]

Back to the basics--

Why could not the wire be terminated by some mechanical means? Some sort of screw terminal?

(No one would want this in production, but should be ok for hobbyist use, surely).

Definitely try it. It is just it doesn't take much to ruin the stability. If you use 10 meters of wire to make a 1ppm/C stable resistor, then it only takes a problem in 10s of micrometers of the wire to affect stability. You would want a screw terminal with an accurate edge of contact and minimal stress on the wire.

[Magnificent Bastard]

Back to the basics--

Why could not the wire be terminated by some mechanical means? Some sort of screw terminal?

(No one would want this in production, but should be ok for hobbyist use, surely).

Well, the position of the wire can move, changing the value of the resistor and/or the joint can corrode, changing the value of the resistor-- and also, there might be more thermal EMF than a properly welded joint.  Just my $0.02 worth...

[Gyro]

Back to the basics--

Why could not the wire be terminated by some mechanical means? Some sort of screw terminal?

(No one would want this in production, but should be ok for hobbyist use, surely).

I was thinking the same thing - with the right shaped clamps / washers / whatever, it ought to be possible to make a gas tight contact with the wire at a very defined point (soft knife edges?). They could also be arranged to be isothermal.

[amspire]

I was thinking the same thing - with the right shaped clamps / washers / whatever, it ought to be possible to make a gas tight contact with the wire at a very defined point (soft knife edges?). They could also be arranged to be isothermal.

Even with the soft knife edges, at the sub mm level, the contact point will have a width and the leading edge of that contact point may only have 1% of the pressure further down the wire. Initially, that area with the 1% pressure will be making contact, but with time and corrosion, it may stop making contact. That is all it needs to change the resistance by perhaps 10-100ppm.

I imagine that the welds Edwin uses have a very precise and mechanically stable contact boundary.

[Magnificent Bastard]

If you can't afford the thousands of US$'s for a Sunstone Pulsed-TIG welder, and you don't want to DIY, then here is a less expensive alternative:

http://www.aliexpress.com/item/220V-Stainless-Steel-Spot-Laser-Welding-Machine-Automatic-Numerical-Control-Touch-Pulse-Argon-Arc-Welder-for/32818492226.html

You would still need the industrial Argon tank for 99.99% Argon gas, and a regulator for the tank, plus some tubing to hook up the gas from the regulator to the back of the welder.  I use an aluminum 60-L tank I bought from eBay-- you might find a good used one near where you live (or rent one from the gas supplier).

Bonus:  You can do a hell of a lot more with this tool than just make resistors.

This should be a "must have" item in your shop.

If you want to DIY, just a suggestion that you can buy the stylus and a matching connector from one of the sellers on aliexpress; then just do all of the rest of the electronics yourself.  You might end up spending more than just buying one (above) though.

Don't worry about "ruining" the TCR of the wire; the pulse is so fast that it does not heat the wire except very near where you are actually welding.

[JoeN]

If you can't afford the thousands of US$'s for a Sunstone Pulsed-TIG welder, and you don't want to DIY, then here is a less expensive alternative:

http://www.aliexpress.com/item/220V-Stainless-Steel-Spot-Laser-Welding-Machine-Automatic-Numerical-Control-Touch-Pulse-Argon-Arc-Welder-for/32818492226.html

How long have you owned and how much have you used this?  I always have a problem buying stuff on Aliexpress more than a hundred dollars or so because Chinese stuff can work for a while and crap out.  Seems this happens quite often.

[Magnificent Bastard]

If you can't afford the thousands of US$'s for a Sunstone Pulsed-TIG welder, and you don't want to DIY, then here is a less expensive alternative:

http://www.aliexpress.com/item/220V-Stainless-Steel-Spot-Laser-Welding-Machine-Automatic-Numerical-Control-Touch-Pulse-Argon-Arc-Welder-for/32818492226.html

How long have you owned and how much have you used this?  I always have a problem buying stuff on Aliexpress more than a hundred dollars or so because Chinese stuff can work for a while and crap out.  Seems this happens quite often.

I never bought one from AliExpress; I have a Sunstone welder.  You could buy one of these, and gut the electronics and then build your own; and make it as reliable as you can.

[MisterDiodes]

Don't worry about "ruining" the TCR of the wire; the pulse is so fast that it does not heat the wire except very near where you are actually welding.

Actually that's about opposite of what happens.  The weld joint is absolutely critical to keep low TC, and you're not even at the tip of the iceberg.  How you weld is even more critical.

You take a spool of wire labeled 3ppm TC, and that's what it was when it was placed on the spool.  The exact moment you unwind it, or tug or bend or crimp or weld or look at it the wrong way:  The TC is going up.  That TC spec on the spool is about as good as it gets from the day it was alloyed and drawn at the wire mill.

Again:  Crimp, fold, bend or kink the wire and TC (and noise) goes up.  So much for crimped or screwed connections.  That's why.  Take the time to research the long history of PWW resistors to find out all the mistakes made by trying to using mechanical clamp/crimp methods.  In other words: Read and Learn. 

Look:  I'm not being negative here, but the fact is:  I'm absolutely positive you probably never make a precision resistor for less money than thousands of good purchased parts that will work better, and are built on better equipment by people who actually know the process (and many decades more experience).  This is not really a hobbyist project - if you're in it to save money on your precision resistor needs.  Can you DIY a mediocre, UN-stable, noisy resistor?  Probably.  You'll still be out many times the cost of a good part though.

Bran: Edwin has been very patient here, and he's giving you a lot more info than any other manufacturer - for free.    He's offered to help you replace your broken parts for free.  He's giving you tons of information - and why on earth should he divulge the exact process he uses??  That's called Intellectual Property, and that has a lot of value - and you never expect that knowledge at no cost.

Buy some resistors if you want support of the manufacturers.  Expect to pay a fair price if you want those quality parts again next year, because you -want- the suppliers to stay in business if YOU are going to stay in business.  That's how it works.

Always true: If you don't like the price of the parts, then you're not buying enough parts.  Nobody is running a charity for parts or exact manufacturing tech info - please don't expect that to happen.

[amspire]

Look:  I'm not being negative here, but the fact is:  I'm absolutely positive you probably never make a precision resistor for less money than thousands of good purchased parts that will work better, and are built on better equipment by people who actually know the process (and many decades more experience).  This is not really a hobbyist project - if you're in it to save money on your precision resistor needs.  Can you DIY a mediocre, UN-stable, noisy resistor?  Probably.  You'll still be out many times the cost of a good part though.

Great post. And I am still actually going to have a go at making some resistors myself. I just want to see what happens.

Probably start out with a divider that just needs temp coefficient tracking, but it is just interesting to see how good or bad the results are. I am particularly interested in short term stability for transfers - and particularly trying to get a very low resistance versus voltage coefficient.

If I was making a 10V reference standard for home, I would be buying resistors for sure. I could never trust a home made resistor in a primary reference.

[Edwin G. Pettis]

I am encouraging experimentation, you could possibly understand better how resistors work even if you can't throw together a <5 PPM°C TCR.  Knowledge is always better when you learn from doing even if you fail.

It wasn't that long ago that making resistors under 10PPM/°C was difficult, then here and there somebody managed to get closer to 5 PPM/°C.  Mind you those 'improvements' came along with significant losses in yields.  Even today resistors with very low TCRs, such as those very expensive Vishay hermetic with <2PPM/°C come with yield losses, they don't have 100% yields.  While I consistently have greater than 90% yields at ±3 PPM/°C, there are still small losses for various reasons, sometimes the TCR is off a tad and I know why in almost every case.  This all comes from experience and interestingly enough, most failures are from mechanical sources (in my case).  In some cases, I know of yields that are only 60% to 70% for a given set of specs at some other competitors.  Some of those resistor manufacturers have been at it for decades and have yet to even get past 80% yields for tight parameters, that is just the beast that PWW resistors are.  And it isn't just wire wounds, film/foil have their own set of alligators in the swamp to contend with, different construction, different problems.

Depending on who is making the resistors, their yields are going to be most likely different to some degree than someone else's, TCR and stability are just two parameters that are hard to hit.

So have at it and try to have some fun while learning.