What is the maximum gain of my opamp?

[jutek]

op amp gain

hello

what is the maximum possible gain of this circuit?

i can't get nothing more 200V/V with vdd=1.3V and ibias 2uA

regards

 

[omsi]

opamp's gain

The gain of ur OP-AMP is depends on feedback that u are providing.
If u provide feedback in suchway that ur resistor in FB is good then u can get the gain what ever u required.

 

[jutek]

Re: opamp's gain

The gain of ur OP-AMP is depends on feedback that u are providing.
If u provide feedback in suchway that ur resistor in FB is good then u can get the gain what ever u required.

???

i was talking about open loop gain, which is only 200V/V

 

[mists]

opamp's gain

I think you should diode bias M1,M2 in the schematic, and the circuit gain should be larger than 60db

 

[pixel]

Re: opamp's gain

This configuration is not amplifier because there is not signal path from input to output (no signal flow from drain to gate!!!). Your simulation results are probably incorect, and gain should be 0.
If you wanted to make folded cascode then m1, m2 should not be there...
If you wanted to make m1,m2 diode connected (simple OTA amplifier) there is no sense to have cascode m7-10 in parallel with simple mirror m3-4 at the output stage.

 

[jutek]

Re: opamp's gain

This configuration is not amplifier because there is not signal path from input to output (no signal flow from drain to gate!!!). Your simulation results are probably incorect, and gain should be 0.

the circuit comes from Design of Low-Power Analog Drivers Based on Slew-Rate Enhancement Circuits for CMOS Low-Dropout

article's link

i had to cut off M7,M8 pair to ensure work with vdd=1.3V did i do right??

so as you can see, it is an ampflifier. in the article it has about 70dB but with M7,M8 pair.

regards

 

[elbadry]

Re: opamp's gain

you simply have to diode-connect the 2 transistors M1 and M2 ( Connect their drains to their gates ).

The schematic you sent will make their gates floating.

This circuit is a current-mirror Opamp. the current of 1st stage is mirrored to the 2nd, hence m1 and m2 have to be diode-connected transistors. You can find more details on it in Johns and Martin.

Also, to improve the gain of this circuit, you can use the Leveraged current-mirror opamp.

Regards

 

[jutek]

Re: opamp's gain

you simply have to diode-connect the 2 transistors M1 and M2 ( Connect their drains to their gates ).

of course M1 and M2 are diode connected, i forgot about it when i was redrawing.

but the simulation is still the same, cause i use hspice


now, it's the right circuit

 

[pixel]

Re: opamp's gain

OK now is amplifier. As a first see if all mosfets work in active region. If your gain is small it seems that some of them are in triode region (maybe bias1 voltage is too small). It would be good if you show dc voltages. For example try to use normal current mirror, and see what happens. Notice that you should add M7-M8 transistors (sre.gif circuit), or cascode mirror has no sense.

 

[ricklin]

Re: opamp's gain

2 stage without cascode architecture, gain=200 is the normal case for .13um techs in 1.2v supply, unless you don't care bandwidth and output swing.

 

[beckchm]

Re: opamp's gain

you must make sure the transistors are in sat

 

[corel]

Re: opamp's gain

Notice that you should add M7-M8 transistors (sre.gif circuit), or cascode mirror has no sense.

Maybe its supply voltage is only 1.3V.I think adding M7-M8 transistors (sre.gif circuit) won't let all transistors operate in saturation,so he got rid of it.

 

[pixel]

Re: opamp's gain

Maybe its supply voltage is only 1.3V.I think adding M7-M8 transistors (sre.gif circuit) won't let all transistors operate in saturation,so he got rid of it.

Yes it could be problem for 1.3V. I wanted to note that there is no much sense connect simple current mirror in parallel with cascoded current mirror, because output impedance is defined by smaller of theese two. In this case you should not expect high gain.
M7,8 from opamap.gif can maximally improove gain for factor 2x (in reality smaller), which is much smaller than expected from cascode output stage.