On Fri, 26 Jan 2001, Harold M Hallikainen wrote:

> On Thu, 25 Jan 2001 19:34:19 -0700 Richard Ottosen <rottosen@IDCOMM.COM>
> writes:
> >
> > Trivia time. I seem to remember that the best reference zener diodes
> > (either 6.2 volt or 8.2 volt, I think) are actually a zener diode in
> > series with a signal diode. This is done so that the temperature
> > coefficients of the 2 diodes cancel out. Does anyone else remember
> > the
> > same thing?
> >
>
>         I STILL like the LM399H. http://www.national.com/pf/LM/LM399.html
> Looking at my old drawings, I first designed it into a circuit in 1978.
> It's tempco is .0002%/degree C (2ppm/degree C). I then spent a couple
> bucks each for resistors with a 5ppm/degree C tempco to divide the 6.95V
> reference down to what my A/D needed. I'd always thought National
> should've made the chip with voltage divider resistors on the chip where
> they'd be in a constant temperature environment...


I concur. However, to obtain that very stable tempco, the LM399 has a heater to
keep the silicon junction at a relatively constant temperature. Keep this in
mind for those low power applications.

Scott

--
http://www.piclist.com hint: To leave the PICList
mailto:piclist-unsubscribe-request@mitvma.mit.edu