Designer’s Guide Community

nus_lin Community Member Offline Posts: 48	questions about PTAT Jan 13^th, 2011, 11:40pm i simulated a basic PTAT core based on SMIC18MIX technology. the simulated ptat output (based on ideal resistor) is actually proportional to T-(2E-4)*(T-300)^2, and is nothing to do with the area ratio and the current density (except for very low/high current ). somebody tell me why there a nonlinearity exists?
Back to top	IP Logged

RobG Community Fellow Offline Posts: 570 Bozeman, MT	Re: questions about PTAT Reply #1 - Jan 14^th, 2011, 6:04am A proper PTAT source should not have any curvature so you have something non-ideal kicking in. It is impossible to say what it is unless you show us the circuit that you are using. rg
Back to top	IP Logged

nus_lin Community Member Offline Posts: 48	Re: questions about PTAT Reply #2 - Jan 15^th, 2011, 2:31am dear rg, it is a textbook structure, with two pmos on the top, two bjt in the bottom, a resistor connects to on of the bjt with bigger size, and an ideal amplifer. if you do derivation to the current output, you will see how good is your ptat.
Back to top	IP Logged

avlsi Community Member Offline Posts: 91 singapore	Re: questions about PTAT Reply #3 - Jan 15^th, 2011, 9:03am Hi, In reality, the pmos are non-ideal, BJT itself is non-ideal. Please post the simulated response and test circuit to be more clear in terms of non-linearity.
Back to top	Winners never Quit,Quitters never Win IP Logged

HdrChopper

Community Fellow

Offline

Posts: 493

Re: questions about PTAT
Reply #4 - Jan 17^th, 2011, 4:49pm

nus_lin wrote on Jan 15^th, 2011, 2:31am:

dear rg, it is a textbook structure, with two pmos on the top, two bjt in the bottom, a resistor connects to on of the bjt with bigger size, and an ideal amplifer.

if you do derivation to the current output, you will see how good is your ptat.

Most probably the circuit in the text book neglected, among other things, the effect of the bjt base current to simplify the calculations.
This, on top of the pmos not being ideal as avlsi suggested, might be the sources of your circuit�s non-ideal behavior over temp.

Best
Tosei

Back to top

Keep it simple

IP Logged