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

current mirror problem

Status
Not open for further replies.
A

ahmed709

Junior Member level 1
Joined
Dec 22, 2006
Messages
18
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Activity points
1,425
I am making a design for a charge pump, and i want to eliminate the mismatch current as much as i can, but the problem is when i made a current mirror with a same (W/L) then the the current i got from the mirror was not equal to the source current, i.e the source current is (5u A) and the mirror current is (5.4u A), so what is the problem and what can i do

the second question is: what is the advantage of putting an op-amp in the architecture of the charge pump, and where i can get the schematic of the op-amp

thanks in advance
 

try adding high swing cascode mirrors, the opamp enhance the dynamic response as the switching node will see similar MOSTs
 

    A

    ahmed709

    Points: 2
    Helpful Answer Positive Rating
safwatonline said:
try adding high swing cascode mirrors
Click to expand...
i am sorry but can you tell me how to do this

thanks in advance
 

You can also increase the transistors' L while keeping W/L constant .This decreases the channel length modulation effect while keeping the same headroom. The drawback is the increased capacitance but I don't think that this is critical for CP used at low reference frequencies .
 

    A

    ahmed709

    Points: 2
    Helpful Answer Positive Rating
ieropsaltic said:
The drawback is the increased capacitance but I don't think that this is critical for CP used at low reference frequencies .
Click to expand...

my reference frequancy is (5M Hz) and my technology is (130n M)
so what is the maximum length i can use in transistors.

thanks in advance.
 

the maximum length that u can use for MOSFET can be confirmed from the library/process file u are including.For example,in TSMC018 the range for length is from .18µm to .22µm.However, in some VLSI Design pages in web it's adviced to use .20µm as a good compromise.
 

the VDS between the source current and the mirror is same?
 

I agree with williamy,you can use an opamp to make the vds is same.Maybe the opamp is constant -gm opam
 

    A

    ahmed709

    Points: 2
    Helpful Answer Positive Rating
zhangsai04 said:
I agree with williamy,you can use an opamp to make the vds is same.Maybe the opamp is constant -gm opam
Click to expand...
instead of using opamp, i would recommend using cascode structure because when using opamp stability issue may arise if not compensate properly. However if the design is in low power, cascode structure might be a challenging task.
 

    A

    ahmed709

    Points: 2
    Helpful Answer Positive Rating
you can increase L by put resistance in series to PMOS
 

    A

    ahmed709

    Points: 2
    Helpful Answer Positive Rating
I think u can add opamp to get a large res or cascode-structure to match the current.
 

ahmed709 said:
I am making a design for a charge pump, and i want to eliminate the mismatch current as much as i can, but the problem is when i made a current mirror with a same (W/L) then the the current i got from the mirror was not equal to the source current, i.e the source current is (5u A) and the mirror current is (5.4u A), so what is the problem and what can i do

the second question is: what is the advantage of putting an op-amp in the architecture of the charge pump, and where i can get the schematic of the op-amp

thanks in advance
Click to expand...

some things can be applied.
1. longer channel length
2. cascode
3. source degneration
 

    A

    ahmed709

    Points: 2
    Helpful Answer Positive Rating
source degeneration is just put some impedance/resistor between the source of transistor and gound.
by doing so, you have to also take headroom and noise issues into consideration.
 

High swing cascode current mirror is a good choise.
 

wide swing cascode mirror is good it helps in keeping same vds for both the transistors and setting higher vgs helps especially vth{threshold} mismatch in the mirror .Probably montecarlo will still be worse if transistor dimensions are not good .
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top