Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Wide swing cascode mirror with variable input current

Status
Not open for further replies.

Junus2012

Advanced Member level 5
Joined
Jan 9, 2012
Messages
1,552
Helped
47
Reputation
98
Reaction score
53
Trophy points
1,328
Location
Italy
Activity points
15,235
Hello,

Below is the normal wide swing mirror,

mirror.PNG

I have swept the value of Iin and kept the current in the diode biasing constant as normal as we do for biasing the mirror in operational ampliifer or other circuit it used. Monitoring the output current I see the output is mirror of the input current (with mirror raio set to one) for a certain range, then it start to saturate and not following the input current. The doide biasing current is 10 uA and the mirror stop working around current more than 60 uA.

This is not an ideal behavoural of the mirror where it should always work to copy the input current to the output, thus it will create an issue when it is used as mirror load in operational amplifier during slew rate where the current is becoming more specially with adaptive biasing current of the differential pair tail current

I found one solution to extend the range is by decreasing the diode W/L further, but this still limited and has other drawback on the output range since it increase the overdrive voltage of the output mirror transistor

mirror_result.PNG
 

Is it possible that as you increase the current your MN4,5 transistors go into triode and become just like resistors? Than your output current mirror changes to the mirror MN3-MN0 with MN0 being degenerated by MN5. The ration of MN3 to MN0 is 1:8 and your output current saturates at about 60uA, maybe accounting for the degeneration.
 
Is it possible that as you increase the current your MN4,5 transistors go into triode and become just like resistors? Than your output current mirror changes to the mirror MN3-MN0 with MN0 being degenerated by MN5. The ration of MN3 to MN0 is 1:8 and your output current saturates at about 60uA, maybe accounting for the degeneration.

Dear Suta,

Recently I read many papers about this issue,
I found the solution for this, that is the only way is to bias the diode transistor with a current that track the input change, one other simple solution is given in the paper below, I have tried the latter one and it prooved itself as a good solution,

**broken link removed**

please let me know if you can not access the paper but basically here is the circuit

adaptive_mirroe.PNG

In the steady state DC operation the capacitor CBat is open and the biasing voltage VCN appears on the gate terminal of M3c M4c because the no current will pass through the huge pseudo resistor MR2. in the transient, this capacitor will hold the voltage and will not discharge quickly so it can keep M3C and M4C in the saturation region of any amount of input current

- - - Updated - - -

Dear Suta,

Ipresent below the mirror circuit with the above mentioned solution, for the purpose of comparesion I have also shown the classical one, you can see clearly how the mirror with the Cbat can mirror higher value of input current while the diode biase transistor is fixed by the current in either cases

batery_circuit.PNG

bater_result.PNG

I would noted here, for normal Opamp or OTA even the calssical one can work fine at the output stage since the current is not gonna to change in more than two orders, howver, for a circuit with adaptive biasing tail current the classical one will fail to mirror the current for a certain level then limits the slew rate below what is expected to be, hence the second circuit will be necessary to sove this problem
 

Yes, I agree that this approach can be use to extend the useful range of the current mirror. Of course, details remain about how big the battery cap has to be vs. how fast the mirror should respond to changes in the current.
My previous response was geared more towards finding the reason why there is a limitation of the range of the mirror as current was increased. Because once you know the reason, you can start thinking about solutions.
 
Yes, I agree that this approach can be use to extend the useful range of the current mirror. Of course, details remain about how big the battery cap has to be vs. how fast the mirror should respond to changes in the current.
My previous response was geared more towards finding the reason why there is a limitation of the range of the mirror as current was increased. Because once you know the reason, you can start thinking about solutions.

Thank you Suta,

Sure now one must consider the small signal behavioural, like the bandwidth of the circuit since the added cap will load the op-amp, therefore, I have selected the minimum cap available by technology that is 100 fF.

I am not sure if we can connect a transistor as a capacitor instead of the poly capacitor for this design or not ?
 

It would depend on the voltage across that MOS cap. It has to be large enough to invert the channel and build up some capacitance.
 
Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top