Question about bandgap stability

[lics]

bandgap stability

Hi, all
I don't know how to simulate whether the bandgap circuit is stable. I use such a way as the figure. I use a small step signal V. Then I run transient to see whether the output is stable. Does that method right? Why? Can someone tell me how to simulate whether the BGR is stable?
I will appreciate for your reply!
Thank you in advance!

 

[gte582w]

There's at least two feedback loops, one is positive
and another one is negative. So they fight with each
other.

You might also try to step at the + input, - input terminal
of the OpAmp, and supply lines (something like CM step).

 

[lics]

Hi gte582w,
You said my measure is right and I also need to add step signal at two inputs and supply. But someone said I should AC simulation which I am not good at. I am misunderstanding which method is right.
Can somebody explain?

 

[pfd001]

you should simulate the AC. Break the loop the input of the amplifer, put the ac input signal to the input of the amplifier, then you can get the phase margin at the e of the Q1

 

[layes2]

loop gain sim

 

[lics]

Hi pfd001,
I have done as you said. But the gain is negative, the phase decrease from 0 degree. I don't know how much the phase margin is. I want to ask another question. You said I should put the ac input signal to the negative input of the amplifier. Can I put the ac signal at the positive input?

 

[gte582w]

you will use stb to do AC check of the loop gain and phase margin,
but remember AC is small signal, therefore, I would trust ultimately
the transient, that is why step the inputs and supply lines are must.

 

[heartwide]

I think that you can test it by adding a step power supply and then do some transient simulation .

 

[ankit12345]

Search for the Thesis papers in edaboard.I have seen a paper long back which addressed these problems.

 

[laglead]

try two loop gain simulation

 

[pfd001]

Hi pfd001,
I have done as you said. But the gain is negative, the phase decrease from 0 degree. I don't know how much the phase margin is. I want to ask another question. You said I should put the ac input signal to the negative input of the amplifier. Can I put the ac signal at the positive input?

It must be sth wrong. If it is true, your bandgap can't work. You put the differential signal at the positive and negative input. you can simulate it just like you did for amplifier

 

[lics]

Thanks for reply!
But I still don't know how to do it.
I will apreciate if someone can give me papers or links about it.

 

[winsonpku]

As i know, i will do two simulations about the stable problem of the bgr.
first i will give a setp power and do the tran simulation of the circuit, then you will find the circuit is stable or not and wether it can work properly.
Second i will add AC signa on the power suplly and do the ac simulation, you will get the psrr and bandwidth of the bgr, these give you a reference.

Hi, all
I don't know how to simulate whether the bandgap circuit is stable. I use such a way as the figure. I use a small step signal V. Then I run transient to see whether the output is stable. Does that method right? Why? Can someone tell me how to simulate whether the BGR is stable?
I will appreciate for your reply!
Thank you in advance!

 

[shake2004]

you can refer to hurst's paper about loop gain or return ratio simulation. you can refer to the ook--<<the designer's guide to spice and spectre>> also