[Diy_efi] NTK wideband sensor

[Peter Gargano]

Brian L Massey wrote:
> 
> ... The only way to get an actual calibration curve, for
> example, is to get a statistically significant (say at least a hundred
> or two) sample of sensors ...


This is completely bogus.

A single sensor will give you as good a curve as any other. Two hundred
sensors will give you even more confidence that your single sensor had
just as good a curve as any of the other.

If you're talking about how to experimentally determine how the factory
supplied calibration resistor would be used in a compensation circuit,
then yes, you may need say 20 or more to be confidant you had cracked
the mystery. 200? Sounds like a "my sensor inventory is longer than yours"
pissing competition story.


Peter.

-- 
Tech Edge Pty. Ltd.     http://techedge.com.au

_______________________________________________
Diy_efi mailing list
Diy_efi at diy-efi.org
http://www.diy-efi.org/mailman/listinfo/diy_efi