[volt-nuts] LTZ1000 VS LM399

[Frank Stellmach]

Jan,

obviously, you also have missed these long and exhausting discussions on 
EEVBLOG about those two references:

http://www.eevblog.com/forum/projects/ultra-precision-reference-ltz1000/
http://www.eevblog.com/forum/projects/lm399-based-10-v-reference/


The difference between both is much more distinct:

The LM399 is about 1-2 orders of magnitude less stable than the LTZ1000 
and appropriately selected external components.

The LM399 is good for stabilities of about 20ppm/yr. and <1ppm/K T.C., 
i.e. 6 1/2 digits class DMMs.

The LTZ1000 plays in totally different league of the 8 1/2 digit DMMs, 
calibrators and voltage standards, i.e. < 1ppm/yr. and < 0.02 ppm/K can 
be achieved on intelligently selected components.


The LM399 is fixed on 95°C oven temperaturer, including the buried Zener 
and all oven temperature and reference determining resistors.
The LTZ1000 on the other hand can be operated at 45°C, and the oven and 
reference determining resistors can be chosen for optimum performance, 
which is not true for the resistors-in-silicon of the LM399.
These differences explain the ultra stability of the LTZ1000.

Therefore, you may average as many LM399 as you like, you will hardly 
(never) achieve the ultra performance of a single LTZ1000.
If the averaging follows a 1/sqrt(N) law, then more than 400EA LM399 
would have a similar stability as the LTZ reference.

Therefore, solely the needed stability determines, which reference has 
to be used.

Frank