Susumu resistors

Is that their RG series, or what?

Right.

Huh... so is there a value that is totally flat? The geometric mean is about 10k ohm. (Unless it's to early for me to do math.)

George H.

Interesting that they're linear. I infer from the datasheet that Z-foil resistors have more of a parabolic curve.

Of course they're something like two orders of magnitude better so I guess they've done a first order correction to a smallish linear tempco like the Susumus.

--Spehro Pefhany

Not having tested many of them, I'd guess that there is variation from part to part, and reel to reel. And there is no "totally flat"; keep zooming the y-axis until it's ugly again.

There are expensive Vishay parts well below 1 PPM. These Susumu's average around 20 cents.

Pretty much any thing that's tweaked to zero out the 1st order term leaves behind a residual parabola.

Best I could trim an oscillator tempco:

Most physical things correct pretty well with a 2nd order polynomial. We correct lots of things with polynomials, in uPs or FPGAs sometimes, and we generally go 3rd order. Higher orders can get wild.

I'm lately doing "linear" timing ramps with a simple RC, polynomial corrected. A resistor is sure a nice wideband, stable, cheap current source.

If trimming is done anyhow, consider the positive-tempco resistors that are proportional to absolute temperature. The linear-in-temperature character CAN be mass-produced easily and cheaply, it just didn't win the market.

TI used to make a line of PTAT resistors that would have been perfect for this kind of use. From a materials-science point of view, that's the stable solution with the simplest correction curve, following Drude model for resistivity.

John, how'd you get that tempco data?

Hi Win

HP3458A 4-wire measurement and my environmental chamber (beer fridge) :)

IIRC I used a keithly 2001 with an RTD to get the temperature, and some python script to collect the data.

In 2012, when I must have had time for the fun stuff.

I searched for Z-210 resistors... no luck. can you give me a bit more of the part number?

Thx, George h.

Vishay Z201, $16 each.

Thanks Win, (a silly typo on my part.) GH

Thanks, some pretty ugly curves there.

--Spehro Pefhany