how to do stability analysis for bandgap circuit?

[ashish_chauhan]

bandgap stability

Hi,

Can you put some light on how to do stability analysis of bandgap circuit?

 

[transbrother]

bandgap stability analysis

You will have to analyze both the positive and the negative feedback loop. That is cut the loop in both and analyze to see if you have more negative feedback than positive.

 

[DenisMark]

bandgap circuit

Generally U shouldn't check phase margin in bandgap (and other biasing circuits), because result of AC simulation may be deceiving. But always check stability during transient analisys with slow and fast supply voltage ramp.

I cite :
" Verification of the biasing circuit has to be done by transient simulation,
with slow ramping of the supply voltage first, followed by switching of the
load current and supply voltage. AC small-signal simulations may be
deceiving. For example, it is interesting to notice that AC simulations of the
phase margin in the voltage regulators provide results that would be
unacceptable in the OpAmps (5o phase margin, for example). Yet, the
simulations of step-response transient or experimental verification of this
response on the silicon chip show quite sufficient excess stability. Also, the
start-up problems in biasing circuits are likely to cause conditional
instability, and AC simulations do not detect this circuit problem."

 

[ashish_chauhan]

bangap circuit

Well I asked this question because I am not getting desired response on silicon.

My transient simulations are perfect but small signal did sometimes showed instability.

And on silicon as well I am getting small ringing... can supply noise and probe capacitance cause this?

 

[yschuang]

bandgap circuit stability

Hi, DenisMark

It is always true to run the transient analysis after doing AC simulaiton. However, I do not know why the result of AC simulation may be deceiving. I do not quite understand texts you quote. Please give me some examples if you can.
Thanks

 

[DenisMark]

bandgap stability analysis

Hi, yschuang
U can do AC simulation before TRAN analysis for BG reference. But there both positive and negative feedbacks in bandgap. Actually i don't know how stability criteria changes, but it isn't a same as for operational amplifier. Internally TRAN analisys more accurate than AC analysis. And it doesn't matter if AC analysis shows good results that in TRAN all will be OK.
In summary. If TRAN analysis doesn't show problems with BGR stability than it's enough. AC analysis in this case can only help with choosing frequency compensation technique and compensation capacitance value. Only TRAN analysis can gives u results close to truth not AC!

I onlt cite good book writen by good senior designer.

 

[srinivasan]

slow ramp band gap

In general stability analysis can be performed by Middle Brooks Method to find the return ratio and loop gain using a spice simulator. This method involves injecting a small voltage/ current change at a suitable node in the loop and observing the return ratio.

While performing stability analysis by breaking the loop, make sure the loading conditions are same before and after the loop is broken.

 

[peni]

breaking the loop during stability simulation

Usually analysis stability , you must break the loop , and add a small signal voltage input to check the loop gain and phase margin. But generally we don't need analysis stability in bandgap circuit.

 

[cloudfly88]

midldle brook stability criteria

I agree with DenisMark's views about the stability analysis, but AC simulation is desirable to choose the preferable compensation network and components's value

 

[WindRay]

do we break feedback loop for stability analysis

In general, TRAN analysis is good because you actually subject your circuit on the real conditions (DCOP and the load/feedback conditions), so when the signal is injected, you are causing a distribution to the circuit. However, you cannot tell how stable is the circuit in term of values or quantity. You can only tell what is the behaviour of the system ie overdamp,underdamp, critical damp... Not so easy to determine the phase margin, etc.... Normally, I want the system must settle down within a few number of rings.. at all conditions when subject to a small step noise.

As for AC analysis, it is always good to bias the circuit to the correct DCOP which is the same as what you are going to use the circuit. As you know the transistors parameters, gm, cgs, etc ... depends on the DCOP. The load must be included to for more accurate analysis if possible..

As for the bandgap, ensure the opamp have enough margin and it is normally stable because the negative feedback is more than the positive feedback if you connected correctly. Take note on the high frequency parasitic componemts within your opamp bandwidth. So to ensure you need to run what "transbrother" sugguested in AC analysis and what "Denismark" sugguested in TRANS analysis.

 

[stocol]

loop stability analysis on bandgap

there are negative feedback and positive feedback at the same time.
u can simulate the loopgain and phase margin keeping the feedback loop closed.
by inserting a Vac into the location between the output of opa and the gate of pmos mirror u can simulate Gv,and adding a Iac between the output of opa and gnd can simulate Gi.Finally u can get the loopgain using the equation G+1=(Gv+1)||(Gi+1).
i have compared this method with the troditional one,and the former is more precise.
u can have a try.