How to order from Vishay Foil resistors?

[nour]

Nour, my understanding is that though foils may have a better tempco, precision wirewound will have better long term drift.  Tempco can be compensated for, but the only solution for long term drift is calibration -- pick your poison.  My plan is to (eventually) try both and compare

I actually don't know much about what you have mentioned to confirm it or not and also as you said I will try both.

But if wirewound is good enough, how much enough should be enough?, for example, is it good enough for LM399 or LTZ1000. if so I would go for it!

Originally I want two sets, one as a lab standard(1,10K) and the other set for building voltage reference standard!

[e61_phil]

Which resistors do you mean at the LM399? The divider to achieve 10V is critical. Alle other resistors (including LTZ1000) are less critical in terms of TC.

[nour]

Which resistors do you mean at the LM399? The divider to achieve 10V is critical. Alle other resistors (including LTZ1000) are less critical in terms of TC.

I mean the one to achieve the 10V, the 9K and the 20K



 And I will add another stage with voltage divider and a buffer to get 5 or 4.096V

[kj7e]

I'm in the same boat, KX LTZ1000 boards ordered, all other parts from Digikey ordered, stuck trying to get the VPG parts. I did speak with a VPG rep, who CC'ed me to a local sales rep at www.uhpresistors.com, but they have yet to respond, even to my follow up.

Edit, they just responded with a quote and 15 week estimate lead time.

[e61_phil]

I mean the one to achieve the 10V, the 9K and the 20K

--
 And I will add another stage with voltage divider and a buffer to get 5 or 4.096V

There is nothing special involved. The gain is simply 1+R1/R2. Therefore you can calculate the total derivative and see the impact of the resistors.

[e61_phil]

..or if you are a lazy boy (like me ). You could just "simulate" the error, like so:

Code: [Select]

tc = np.linspace(-5e-6, 5e-6, 1000)

R1 = 9.1
R2 = 20

exact = 1+R1/R2
ratio_err = []

for tc1 in tc:
for tc2 in tc:
ratio = 1+( R1*(1+tc1) )/( R2*(1+tc2) )
ratio_err.append( (ratio-exact)/exact*1e6 )
In this example the resistors have both +/-5ppm