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.

Register Log in

Purpose of transistors attached with source degeneration MOS

Status
Not open for further replies.
I

ICdesignerbeginner

Member level 5
Joined
Nov 22, 2017
Messages
86
Helped
1
Reputation
4
Reaction score
2
Trophy points
8
Activity points
729
Hi

I am attaching one picture. I can understand that transistor M14 and M15 are source degeneration MOS but what is the purpose of M3 , M16 where M17 and M18 are biasing the transistor M14 and M15 gate. What is M3 and M16 purpose. Can anyone explain? I have attached the diagram

OTA SOURCE DEGENERATED.jpg
 

M14 and M15 operate in deep triode region, so they can replace resistors. But they need some Vgs to be opened, so M16 is used to generate a Vgs for M14 and M15.
M3 is the current source, M17 is swallowing the current. M16 is placed between the source and the sink, so a biasing current can flow through it.
This current on the diode-connected M16 will generate the necessary Vgs.
 

ICdesignerbeginner said:
What is M3 and M16 purpose.
Click to expand...

M3 sources the current ISS - like M4 & M11 - mirrored from master M9, and so provides a high impedance to the center point of the M14, M15 & M16 sources. This current flows through M16 & M17, the latter one mirrors and sinks the same current ISS from the mirror master M18.

Diode connected M16 provides a reasonable Vgs ≈ Vth for M14 & M15, to set them to the required impedance for degeneration of the active differential pair MM1/M1 & MM2/M2.
 

Thanks for your reply.

I have seen source degenerated MOS with out M3 transistor only M16 is used for generating a vgs for M14 and M15. Is it necessary to have a diode connected M16 to provides vgs to M14 and M15 cant it be a simple current mirror
 

If the drain terminals of M14 and M15 are changing due to the change in gate voltage of differential pair then the resistance of source degenerated MOS will vary. Will this M16 help in adjusting the resistance.

- - - Updated - - -

What is the advantage of providing a high impedance to the sources of M14 , M15 and M 16

- - - Updated - - -

What is the advantage of providing a high impedance to the sources of M14 , M15 and M 16
 

ICdesignerbeginner said:
Is it necessary to have a diode connected M16 to provides vgs to M14 and M15 cant it be a simple current mirror
Click to expand...

frankrose has already answered this question above:
Without M16 there's no current to be sunk into M17. Together with M3 & M17, M16 provides the correct Vgs to M14 & M15.

ICdesignerbeginner said:
What is the advantage of providing a high impedance to the sources of M14 , M15 and M 16
Click to expand...

To not short-circuit the signal at the connection point of the 2 degeneration "resistors" M14 & M15 to VDD (virtual GND).
 

ICdesignerbeginner said:
If the drain terminals of M14 and M15 are changing due to the change in gate voltage of differential pair then the resistance of source degenerated MOS will vary. Will this M16 help in adjusting the resistance.
Click to expand...
Yes. You can't ensure a fix Vds, but at least you can ensure a stable Vgs by M16. So resistance variation of M14 and M15 will be lower, but obviously the resistance will change with input differential voltage. If you don't want this use normal resistors. Most straightforward solution, doesn't require huge area, extra biasing circuit and very linear.
ICdesignerbeginner said:
What is the advantage of providing a high impedance to the sources of M14 , M15 and M 16
Click to expand...
Not the high impedance is the point here, I think. If you don't use M3 all current of the M17 has to flow through M16 and M14,M15. If you pull DC current through M14,M15 it will increase the resistance of them which is quite uncontrollable. With higher DC current M14 and M15 will leave the deep triode region, even they can leave triode region and step into saturation region where they act as a current source, not as a resistor.
 

What are you exactly asking for? "source degenerated MOS" is not the same as "source degeneration MOS".

Please clarify the question referring to the circuit explanation given in post #2 and #3.
 

I am asking for source degeneration MOS.

Source degenerated MOS is different I have heard this for the first time that they are different. Can you also explain the difference?
 

If the term "source degenerated" makes sense at all, it applies to M1 and M2 which have degenerated gain.
 

ICdesignerbeginner said:
I am asking for source degeneration MOS.

Source degenerated MOS is different I have heard this for the first time that they are different. Can you also explain the difference?
Click to expand...

I'd think the different terms mean the actively resp. passively involved transistors: In the schematic of your post #1, M14 & M15 are the source degeneration MOS transistors (active participants concerning the degeneration), whereas MM1, M1, M2 & MM2 are the source degenerated MOSFETs (passive participants; their gain being degenerated (decreased)).
 
  • Like
Reactions: FvM

    FvM

    FvM

    Points: 2
    Helpful Answer Positive Rating
Thankyou for your reply. My question was can M14 and M15 be biased by simple current mirrors?
 

My question was can M14 and M15 be biased by simple current mirrors?
Click to expand...
How? As explained by frankrose and erikl in post #2 and #3, you need a circuit to generate the gate-source voltage for M14 and M15. Effectively at least an additional transistor in combination with a current source respectively two current sources. If you have a specific circuit topology in mind, please clarify.
 

Simple current mirror like the ones used to bias the differential pairs. One figure is attached

curren mirror.jpg
 

And where should this mirrored current come from? The M14/M15 gate voltage would be undefined or zero. Try and simulate!
 

If you don't see why this can't work, perform a simulation and look at the node voltages.
 

erikl said:
Diode connected M16 provides a reasonable Vgs ≈ Vth for M14 & M15, to set them to the required impedance for degeneration of the active differential pair MM1/M1 & MM2/M2.
Click to expand...

I am also working of source degeneration MOS adn I think this thread is quite old but I still feel like posting. you said that vgs =vth. Do the size be kept larger of M16 to make vds small. I think the drain and sources of M14 AND M15 should be at the same potential.?
 
Last edited:

The basic idea is to run M3 and M17 current sources with equal current and thus keep M14 and M15 free of DC bias current. If so, there's only AC signal across M14/M15 drain-source. Scaling M16 varies M14/M15 Vgs and changes the source degeneration resistance respectively.
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top