The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Design >> Analog Design >> Question about Vos drift vs temperture
https://designers-guide.org/forum/YaBB.pl?num=1206866402

Message started by James Bond on Mar 30^th, 2008, 1:40am

Title: Question about Vos drift vs temperture
Post by James Bond on Mar 30^th, 2008, 1:40am

Hi, all

Does anyone know how to simulate dc offset voltage drift vs temperture for an fully differential operational amplifier? or for an fully differential system, such as base band filter. Thank you very much.

Best Regards
james

Title: Re: Question about Vos drift vs temperture
Post by Visjnoe on Mar 30^th, 2008, 2:18am

Dear James,

1.. You could run Monte-Carlo for different temperature settings. As in all Monte-Carlo simulations, you
will depend on the accuracy of the models provided by the foundry...

2. You could model the offset as an input-referred voltage source. You could then make this input-referred voltage
source temperature-dependent (equation).

Kind Regards

Peter

Title: Re: Question about Vos drift vs temperture
Post by James Bond on Mar 30^th, 2008, 2:59am

Dear Visjnoe,

Thank you for you reply.

As for the 1 solution, if I want plot Vos versus temperature, I have to sweep temperature in the dc simulation, and run montecarlo simulation based on this dc setting, is that correct?

I don't quite understand solution2, I want to simulate my circuit and find out how the temperature will affect Vos(I have seen somewhere a monotonic reducing curve) , if I can model them, I already know the answer. Could you explain it a bit further?

Best Regards
james

Title: Re: Question about Vos drift vs temperture
Post by thechopper on Mar 30^th, 2008, 5:30am

Hi James,

Actually it does not make much sense to simulate offset over temp except for the monte carlo approach. The reason is that the weights of the offset contributors in your amp will randomly change, hence will randomly change your temperature coefficient too.
In other words, the offset temperature drift for your circuit is in itself a random variable.

Hope this helps.
Tosei

Title: Re: Question about Vos drift vs temperture
Post by James Bond on Mar 30^th, 2008, 9:03am

Hi, thechopper,

I think you are right, I just managed to simulate the offset voltage versus temperature under monte carlo runs, it sometimes increases with temperature increasing, sometimes reduces. But I think the benefits of doing this is to find maximum offset voltage drift of my circuit and then at the circuit output, I will be able to know the maximum offset range, then I will know if this will affect the following circuit performance. What do you think ?

Best Regards
james

Title: Re: Question about Vos drift vs temperture
Post by thechopper on Mar 30^th, 2008, 2:56pm

Hi James,

I think that is the way to go and I would do the same thing in order to estimate the offset range you need to accommodate. I do not know of any other method that might give you offset drifts over temp close to actual ones (assuming appropriate models for mismatch).

Tosei