The Designer's Guide Community Forum

The Designer's Guide Community Forum
https://designers-guide.org/forum/YaBB.pl
Simulators >> Circuit Simulators >> problem with bandgap simulations
https://designers-guide.org/forum/YaBB.pl?num=1264073655

Message started by yvkrishna on Jan 21^st, 2010, 3:34am

Title: problem with bandgap simulations
Post by yvkrishna on Jan 21^st, 2010, 3:34am

I am working on bandgap voltage reference circuit(very much the same as shown in fig) and use a variant of spice for simulations, since I am not sure that simulator is capable of simulating selfbiased circuits, I started using .IC statements for avoiding the problem of startup(as amp here is self biased).

in the fig shown below I initialise the nodes biasp to 0V and n18 to vdd (as they move in opposite directions in an all zero soln when it fails to startup)

doing so i was not successful always, still it fails to give a correct op soln to the circuit. it settles in an intermediate op pt in which nodes are neither at 0,vdd(all zero soln) nor at the correct op pt i expect.

Is this way of using .IC for avoiding startup correct? does this have any issues?
OR
does this circuit really have another stable op pt ? if so how to avoid it?

thanks,
vamshi

Title: Re: problem with bandgap simulations
Post by ic_engr on Jan 21^st, 2010, 9:15am

Vamshi,

Did you perform transient or DC simulation. If DC and you are sweeping the VDD, then the .IC makes no difference since IC is initial condition when your VDD value is '0' and thus IC will have no affect, if VDD=0 at that point, circuit will not start-up properly.

Title: Re: problem with bandgap simulations
Post by yvkrishna on Jan 21^st, 2010, 8:30pm

yes i am aware of that, i was taking about transient simulations only.

thanks,
vamshi

Title: Re: problem with bandgap simulations
Post by ic_engr on Jan 22^nd, 2010, 5:33am

Okay.

The problem with BG circuits is that unless you have a proper start-up circuit, IC may not gurantee a proper star-up, unless the circuit noise itself is large enough to start the BG. But the BG you have is known to be hard to start-up, I refer you to Banba's paper.

First, you should try having a start-up circuit where the circuit will pull down the gates of PMOS switches, until the outptu is stabilized, and then use the output to shut-off the BG circuit.

ic_engr

Title: Re: problem with bandgap simulations
Post by yvkrishna on Jan 23^rd, 2010, 2:23am

thanks for the reply ic_engr,

probably i should try with a startup circuit as well.

thanks
vamshi