Long term stability of matched resistors?

Feb 13, 2017 33 Replies

If I select metal foil resistors with 1% tolerance so that they match to 0.1% can I expect that the matching keeps stable for a longer time or will the values drift sooner or later to their 1% tolerance? In other words: Do I need resistors with 0.1% initial tolerance for long term 0.1% matching?


Dipl.-Inform(FH) Peter Heitzer, peter.heitzer@rz.uni-regensburg.de

** Hang on a mo,

"metal foil" resistors are typically 0.01% tolerance or better.

Maybe you mean "metal film".

.... Phil

Even if you buy 0.1% resistors, that does not mean that they will stay within 0.1% forever, nor that they will always stay matched even to the initially guaranteed 0.2%.

Some resistors have stability specifications, but these are usually so loose as to be fairly useless, except for extremely expensive resistors.

Vishay seems to have bought almost all of the precision resistor companies. If you want a second source it is getting difficult.

If the resistors are selected from the same manufacturing lot, are used well below their specified voltage/power/temperature ratings under similar operating conditions they should match fairly well. It's hard to guarantee, though, absent manufacturer matching specs.

If the application is really critical and is a one-off, so presumably less cost-sensitive, you may be best off buying a matched pair in a package.

Read the data sheet about drift, load life and so on. It's not just a simple tolerance specification. The tolerance refers (usually) to

soldering, aging etc.

Generally, chances are excellent that resistors of similar or equal resistance value from similar lots will match well over time, but there are no guarantees, especially if the environment or service (eg. pulses, large DC bias) is not benign.

Since looser tolerance resistors may be made with inferior materials or construction, they may tend to deviate more over time or temperature etc. Or not.

--sp

Best regards, Spehro Pefhany

Yes, of course. I fear my technical english vocabulary is rather unsufficient.

Dipl.-Inform(FH) Peter Heitzer, peter.heitzer@rz.uni-regensburg.de

We've been impressed with the long-term stability of the Susumo surface-mount thinfilm resistors. Some of our products that depend on them are showing drifts of low tens of PPMs per year, with lots of other things contributing.

What specific resistor types did you have in mind?

John Larkin Highland Technology, Inc lunatic fringe electronics

Just ordinary metal film types like vishay CCF55

formatting link
or thick film chip resistors with tempco of 100 ppm. I do not develop products, I use them only in hobby projects.

Dipl.-Inform(FH) Peter Heitzer, peter.heitzer@rz.uni-regensburg.de

Looks like the temperature coefficient will be a bigger error than ageing.

The Susumu's typically have tempcos in the single digits of PPMs. Here in the USA, we can get them inexpensively from Mouser or Digikey.

John Larkin Highland Technology, Inc picosecond timing precision measurement jlarkin att highlandtechnology dott com http://www.highlandtechnology.com

Hi Peter,

schau mal auf Seite 2 im eben zitierten Datenblatt unter Performance. Da stehts:

montiert (alleine schon wegen der thermischen Drift), da sollte dann

darauf verlassen soll, dann sollte man wenigstens ein paar Testreihen aus verschiedenen Chargen testen.

Wenn es aber ohnehin Geld kostet, die Teile zu selektieren, dann ist es

So long

Marte

Sorry about the german writing here, but it is much easier for Peter an me ;-)

Google translates "so long" into English as "so long." It actually sounds Asian to me.

John Larkin Highland Technology, Inc picosecond timing precision measurement jlarkin att highlandtechnology dott com http://www.highlandtechnology.com

Hi John,

may be it is such a denglish word that "should" sound like english but really is a nativ german slang like "Handy" which means mobile but sounds like an english word and therefore more cool;-)

So long at the end of a paragraph means not more or les than "it's enough for now".

So long Marte

Is "so long" originally German?

We'll happily steal from anyone.

(I'm 3/4 German myself, but I never learned the language. I pretend to be Irish.)

John Larkin Highland Technology, Inc picosecond timing precision measurement jlarkin att highlandtechnology dott com http://www.highlandtechnology.com

Den tirsdag den 14. februar 2017 kl. 23.53.09 UTC+1 skrev John Larkin:

afaui "so long" is an Americanization of the greeting shalom or salaam

To me "so long" is a form of "good bye", "till I see you again". a parting thought, "so long, for now"

I've got direct male decent back four generations to a George Herold who came over from the Deutschland, I'm probably less than 1/8 German, but I've always felt more German than I am.

George H.

I agree,

formatting link

no sense of humour, preoccupied with order or take over the world German? ;)

Right, I'm totally none of that, UB (my grad school) had an exchange program with a German school. (I'm thinking Wurtzburg, but that is probably wrong.) I got to know many Germans who where both fun and hard working. Some more this way, some more that... basically like everyone else.

George H.

Wurtzburg that is in Bavaria, I'm not sure they really consider themselves German, Bavaria is like the German Texas or something :)

Ahh well then, I think my ancestors came from some where south of Wurtzburg, they were Catholics, coming to America. :^)

George H.

It's not our fault that we speak a form of English. The Brits made us do it.

Do they have barbeque in Bavaria?

John Larkin Highland Technology, Inc lunatic fringe electronics

Join the Discussion

Have something to add? Share your thoughts — no account required.

Didn't find your answer?

Ask the community — no account required