[volt-nuts] LTZ1000 resistors, lower collector currents

[Frank Stellmach]

John,

all drift parameters of R1, R4, R5 were attenuated by the circuitry by a 
factor of 100, R2 and R3 by a factor of 300. Therefore, bulk metal foil 
is not needed, also wirewound precision resistors will do.

Their T.C. is between < 1 - 3ppm/K, compared to (+/-1 +/-2) ppm/K of the 
S, K types, which is exactly the same.
Even the Z202 type Vishay resistors are no better than (0.2 +/-2) ppm/K.

Therefore, the contribution to T.C. for all types is about 0.02.. 
0.03ppm/K or less. I think, that's to be neglected, as 0.1ppm/K total is 
very fine.

Drift over time of ww. vs. molded metal foil is exactly the same, i.e. 
20ppm/yr shelf life, contributing 0.2ppm/yr. max., compared to the 
1..2ppm/ yr. typical for the LTZ @ 45°C.

I have built my two LTZ references on wirewound resistors, afters 3 
years of nearly continous powering, I have not seen a drift bewteen 
both, against a Fluke 5442A, and a third LTZ in my 3458A in the last 1,5 
years of more than 1..2 ppm. Temperature drift effects I also could not 
distinguish from other stationary/reversible drifts.

If you are living in UK (what is: 'me.uk'?) , visit 
rhopointcomponents.co.uk.

They have those 0.1% wirewound resistors 'econistor 8E16/8G16' on stock, 
I have used for the LTZ

They sell also those very stable metal foil resistors "FLCY" from 
japanese manufacturer alpha electronics (swallowed recently by Vishay), 
which I have used for my Hammond type reference divider, which is also 
ultra stable after years.

Its a little bit more difficult to find FLCY values for R2/R3, but 
25K/2K should do the job...


Frank