The influence of self bias to the gain of opamp

[rogerqin]

When simulating opamp, I found that when applying a constant voltage to NBIAS, the gain of the opamp is nearly 50dB, but when self biasing the gain of the opamp boosted to nearly 90db.
Who can explain this problem detailed for me, thanks!

 

[hr_rezaee]

Hi
I can't understand your circuit.
explain the left side circuit more.
sketch your baising circuit.
regards

 

[analog_prodigy]

Give the left side circuit clearly, so that one can understand easily

 

[DenisMark]

It seems to me that ur self-bias circuit is incorrect. It must form positive feedback with unit gain. I see that this condition isn't satisfied. Try to connect the gate of P7 to gates of P8,P9 may be it resolves ur problem. Follow reference link for example.

A doubt may rise with ur left side.



If my help is usefull don't forget press "Helped me" button

 

[rogerqin]

Hi,all. Thanks for your help. This opamp is used in bandgap. I simulated a simpler circuit as the figure below. The same phenomenon occured. I tried to analyze it using small signal model, but failed. Please help me to check it, thanks a lot!

 

[hr_rezaee]

Hi
what is NBIAS?
How did you test it?
What is the meaning of "failed"?
regards

 

[rogerqin]

Hi, hr_rezaee.
NBIAS is the bias voltage of tail current.
When I applied a constant voltage to NBIAS, the gain of opamp is nearly 50dB, but when self biasing as the figure above, the gain boosted to 90dB.
The meaning of 'failed' is that I can not explain it with small signal model analysis.
Please help me to check it, thanks.

 

[fanshuo]

could it be the feedback loop from Vout to the source of the N0,1?

 

[liuyonggen_1]

why does the output be connected to the gate of another transistor?
there will form a positive feedback loop to increase the vout!

 

[hr_rezaee]

Hi
Does "self biasing" mean that the Nbias is float?
if answer is yes this circuit has positive feedback.
if answer is no you must not have change in your gain.
regards

 

[watersky]

Usually stability analysis need cut the loop, this question perhaps use iprobe and analyze stability.

 

[rajanarender_suram]

it looks like a +ve feed back ckt

 

[rogerqin]

Yes, this circuit has a positive feedback. My question is why this loop is stable and the gain is boosted from 50dB to nearly 90dB. How can I analyze this circuit? Thanks.

 

[DenisMark]

Ok. I've understood ur problem.
U try to design bandgap like

or


Ur main mistake is that u simulate only error amplifier without actual feedback influence. U really need to simulate the whole circuit of bandgap to obtain true dc operating point of ur citcuit, than u need to disconnect positive and negative feedbacks and apply some AC source to simulate AC characteristic. To do it in spectre u can use sp1switch component from analogLib.
However that may be the AC small-signal simulations may be deceiving when applied to bandgaps.
I cite the quotation:
" 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."