SV: Re: [sdiy] Compensating PTC and NTC resistors

[Neil Johnson]

From: karl dalen <dalenkarl at yahoo.se>
> Another small suprice are that so few are using resistor
> nets, SIL's as input buses to expos and signal inputs. In
> the old days these was strictly for digital
> pull dow/upp's and had pretty lousy specs but today 
> +/- 100ppm TC are bog standard. Alltought thick film tech
> and 1% but they are dead cheap.Interestingly
> to note are that above 50ohm individual TC tracking are 
> 50ppm. And one can get without much effort a R-net with
> 5ppm tracking and 15ppm over all and 0.5%.

Well, for small numbers of resistors they are not cost effective(*).  For example, consider a 4-commoned resistor array: four individual resistors would cost 6p, while the resistor array costs 9p (these are Rapid Electronics prices).  5-commoned is still more expensive than discrete resistors.  Its only when you get to 7 or more resistors that the resistor array is cheaper.

Also, with individual resistors you can choose which ones need to be high precision (0.1% or better), while others can be less precise.  For example, the CV input/feedback resistor pair would be high precision, while the course and fine tune inputs (and maybe exponential FM) can be lower precision, cheaper 1% resistors.

In terms of temperature tracking, considering the spread of PPM and physical spread the actual _difference_ in PPM between discrete resistors may be quite small.

Neil
--
(* Yes, I know, smaller PCB area reduces cost too, so what extra money you spend on the array you may save on the PCB area reduction.)

--
http://www.njohnson.co.uk

-----------------------------------------
Email sent from www.virginmedia.com/email
Virus-checked using McAfee(R) Software and scanned for spam