The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Design >> Analog Design >> Question about temperature influence on circuit performance
https://designers-guide.org/forum/YaBB.pl?num=1284816898

Message started by urian on Sep 18^th, 2010, 6:34am

Title: Question about temperature influence on circuit performance
Post by urian on Sep 18^th, 2010, 6:34am

As we run simulation of a circuit,we see from spectre or else that with temperature increasing,the performance of the circuit degraded.Like the speed becomes slower.
But what the exact reason for this phenomenon?
I can imagine two reasons,one is that the mobility is decreasing while temperature increasing, the other is that the R has positive TC so its value becoming large and the RC constant has increased account for speed down.
I wonder whether one of my option is right?
And what about the influence on the gain with temperature?I think gm is aslo decreaed with mobility reduction,right?

Best regards
urian

Title: Re: Question about temperature influence on circuit performance
Post by raja.cedt on Sep 18^th, 2010, 7:19am

hi,
for most of the cases speed(i mean switching speed, VCO oscillation frequency) will increase due to Vt reduction. But i didn't understand why R will come into picture, for most of the cmos circuits transistor on resistance will matter. Coming to last point amplifier gain it depends what kind of architecture you are looking for...for example gm/gm load doesn't have much impact.

Thanks.

Title: Re: Question about temperature influence on circuit performance
Post by RobG on Sep 18^th, 2010, 9:59am

You need to show your circuit if you want a meaningful answer.

But... in general analog circuits go faster with higher current and higher mobility (which both translate to higher gm).

Higher temperature decreases mobility and that would normally decreases speed. Similarly, if your circuit is biased with a positive TC resistor the speed would also tend to decrease as temp increases.

The general fix is to use a beta-multiplier bias circuit to keep the gm of the important devices constant over temperature by supplying a current that is roughly proportional to absolute temperature (aka PTAT).

rg

Title: Re: Question about temperature influence on circuit performance
Post by urian on Sep 20^th, 2010, 10:02pm

Thanks,RobG,and raja
All of the circuits I have simulated which used in a pipelined ADC show the same tendency. I think it all refers to the mobility since there are only transistor in my circuits.

urian

Title: Re: Question about temperature influence on circuit performance
Post by raja.cedt on Sep 21^st, 2010, 5:46am

hi,
please show me some results man..i didn't get any such type of result till now. how you are biasing you op amps in all stages..i guess you have to use at least one resister. isn't it?

Thanks.

Title: Re: Question about temperature influence on circuit performance
Post by urian on Sep 21^st, 2010, 8:32pm

hi,raja
I am doing comparator now,and havent move to MDAC yet.
and the comparator is like this.