Email: Password: Remember Me | Create Account (Free)

Back to Subject List

Old thread has been locked -- no new posts accepted in this thread
???
04/03/05 13:10
Read: times


 
#90900 - Tolerance
Responding to: ???'s previous message
The Tolerance value usually stated is the resistance measuredat some standard temperature. The thing you are usually MOST interested in is the temperature coefficient - the variation of R with temperature, which WILL affect final precision from the likes of ADCs. With careful design, even the effects of this can often be ameliorated by the use of designs where the temperature drift upwards of one stage is compensated by a drift downwards in the ouyput of some other stage.

So, the bald "tolerance" of a cheap resistor in the market is not going to be examined properly by sticking it on a resiatmce meter,unless you are going to carefully oven the devices.

We generally use Phillips MRS25, which have a TC of 50 PPM. Over the limited temperature range of operation of our equipment, that gives an acceptable precision near 12 bits, with the compensation schemes we employ.

Much more critical is often the precision of the voltage reference used, and the number of designs where a 12 bit converter is fed from a zener/resistor combination is quite laughable.

Steve

List of 8 messages in thread
TopicAuthorDate
WEOT: Sorting descrete components.            01/01/70 00:00      
   Tolerance            01/01/70 00:00      
   continuing ..........            01/01/70 00:00      
      Selections and tolerance            01/01/70 00:00      
         Re: selection on tolerances.            01/01/70 00:00      
            Risk taking - logically            01/01/70 00:00      
      Re: Continuing            01/01/70 00:00      
   The other aspect of precision            01/01/70 00:00      

Back to Subject List