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

Neil Johnson neil.johnson97 at ntlworld.com
Tue Mar 18 10:58:49 CET 2008


Karl,

> Well, speaking of wild-ass-guessing Neil!
> TC of MFR appears in some cases to come in +/-. Still you talk about
> 200ppm which one, the - or the + ,clearly you assumes the - sign but
> how do you know the MFR are? 

That's the point - you don't.  If you want some guarantee on temperature coefficient buy a resistor with specified temperature coefficient.  The temp.coeff of ordinary resistors is specified for the record so that the designer can accomodate it if the number is important.  Otherwise get your oven and ohmeter out and measure it yourself.

For comparison, check the tolerance specs of Y5V ceramic capacitors: -20% to +80%.

> Futher, about guessing, im not guessing im trying to figure out whats
> going on by puting up some questions here, and also by digging trough
> various sources this ppm thing in resistors are far from of one issue
> only there are plenty of quite delicate things as i have understood,
> also it differs between technology used, manufacturing process and
> application and between manufacturers and even mounting of the darn
> resistors.  

Exactly - many, many factors to consider.  That's what makes analogue electronics fun - everything is affected by temperature!  Capacitors, resistors, op-amp gain (and yes, even in closed-loop circuits you're interested in the open-loop gain), resistance of PCB tracks (don't forget they're copper too, maybe tinned as well).  Diodes are a good example, drifting the forward bias voltage by about -2.6mV/C.  

And unless you can guarantee that the entire circuit is ovened at a constant temperature you will get thermal gradients across the circuit, producing different thermal offsets (the reason we try to keep the temp.co resistor close  to the silicon junction we're trying to compensate for).

> Not completely just in fragments, since at that point in time few understood
> the complete picture. Still people don't understand the whole picture and i 
> *gladly* admit i don't understand either but other more qualified then me
> don't either.

Any non-trivial circuit will be difficult to understand the *complete* picture.  That's why we make approximations.  The temp.co resistor is just one such approximation.  It gets us "close enough" to perfection at minimal cost.

> So regarding the Rapid resistor you mentioned if you had actually read the
> Vishay datasheet you would have found a spec that says +/-ppm , there is
> not one single place in that sheet that speaks about -or + ppm only.

Why should I read the spec sheet?  I don't rely on a component not sold as a temperature-sensitive device to give me a specific temperature-related performance.  As long as it is below some arbitrary figure that a given design requires then that is sufficient.

> Actually, yes, in tiny fragment of their datasheet (how to order parts)
> Vishay actually says one can order the resistor in -25ppm tolerance!!
> How peculiar! Question arises,can one also order them in +25ppm?
> Which ppm has Rapid stocked them in???

No, don't be silly.  All Vishay are saying is that all of their resistors sold of that type will have a temperature coefficient in the range -25 to +25 (or whatever it is) ppm/C.  Nothing more.  In the same way as saying the tolerance on the value (at 25C) is within some +/-N% range.

Neil

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

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




More information about the Synth-diy mailing list