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

finding MOS kp(kn) value?

Status
Not open for further replies.
A

analogartist

Junior Member level 2
Junior Member level 2
Joined
Mar 18, 2005
Messages
21
Helped
0
Reputation
0
Reaction score
1
Trophy points
1,281
Activity points
1,543
cox mos

Hi all,

This might sound silly to most designers here.....

How can I find kp (or kn) for a MOS device... i.e Up*Cox(Un*Cox)

(In school most projects I did mentioned this value ..so starting a design was easier .. but in industry... how is it done?)...

thanks for your help
 
how to calculate kn in mos model

1. if you have spice model you can look up "Uo" and "Tox".
2. otherwise, run the simulation in the deep triode region; get the trandcondutance (gm) then calculated the Kn/Kp.
ps. choose the channel length and channel width around the design range.
 
nmos kn

due to short channel effect and other parameters, this value is not very accurate so it's utility is limited in professional design. but I know what you mean - it's nice to have a starting value for some paper calculation!

so just build a current mirror with a fairly large area and in simulation, check the dc operating point. From this, you can get Vdsat and GM from sim value to back-calculate k'p and k'n.

For example, in a 0.5um process I had been using recently, I built mirrors of W=5um, L=8um and extracted values of 40uS for k'p and 128 for k'n. Using this for other circuits (op amp) gave a good starting point on sizing the class-AB control for my desired vdsat.

good luck!
 
kp nmos

Is there a reason for choosing larger area..?mismatch??
 

ibm 250n mos device parameters

I think that is not for the mismatch of the transistor, since you are running in ideal simulation. The mask will be the exactly area you define. I think the larger size is to get rid of the LD and WD effect, which makes your drawing L and W will be close to the effective channel length and width.
 

mos kp

Hi
In fact, in todays technologies particuarly to design low-voltage circuits, due to become short channel; this values are not very accurate and it should measure for every current, bias (VGS) and Vds values in a special application. but you can typical values of these parameter (U0, Cox or better Tox, Esilicon, ect;) in offered libraries by manufacturers.
I suggest to understand analysis of circuit and systems perfectly, then go on circuit design and simulation.
Good luck!

Regards,
SAZ
 

finding kn for nmos

**broken link removed**
 

how to get kn nmos

In most of the high-order model,it uses a more complex way to calculate Kp or Kn than the square-law equations.
In some simualtion enviroment, you can get BETA from the simualtion results directly. So you can use this value to calculate the Kn or Kp though it is just a reference value to your hand calculation.But,I think,it is a good starting point and after some iterations,you will get a trade-off between your simulation results and hand-calculating models
 

kn and kp

NMOS PMOS

AMIS

1500n
K' (Uo*Cox/2) 36.6 -11.7 uA/V^2
500n
K' (Uo*Cox/2) 56.2 -18.9 uA/V^2
350n
K' (Uo*Cox/2) 93.1 -22.2 uA/V^2

TSMC

350n
K' (Uo*Cox/2) 92.9 -32.1 uA/V^2
250n
K' (Uo*Cox/2) 125.9 -25.9 uA/V^2
180n
K' (Uo*Cox/2) 171.9 -36.5 uA/V^2

IBM

500n
K' (Uo*Cox/2) 77.0 -20.9 uA/V^2
350n
K' (Uo*Cox/2) 91.4 -22.7 uA/V^2
250n
K' (Uo*Cox/2) 110.5 -27.0 uA/V^2
180n
K' (Uo*Cox/2) 153.7 -32.8 uA/V^2
130n
K' (Uo*Cox/2) 276.0 -47.6 uA/V^2
 

spice mos kp

i think u can find it by a simple simulation, because beta=k'*w/l, in a normal dc operating points, u can find parameter betaeff in results display widow, then calculate it will obtain k'.
 

pmos kp

If you are talking about long channel processes; 0.18u and above, you can get what you want just by looking at the model file... so easy and straight forward...
If you are using an advanced process; 0.13u - 45n technology, it is going to be hard to follow the model file thread. My advice here is to run a simple simulation as the other post mentioned and find "the effective kn(kp)"... This might be faster for you than figuring it from the model file....which you many not be able to understand anyway....
 
(un/up) for mos

I think the parameter using simulation to get .but not accurate
 

retrieve nmos values k

Thanks everyone. for all your feedback....

I did check the model file for the values.. but these were hidden in layers and layers of hierarchy set by the company...

as analog_guru said.. I am working on 135n and less... and the values seem to vary crazily... so I just set up a MOS in saturation( with a fixed current source) and ran a DC on it....this gave me a very very approximate value.... I am hoping by trial and error I would find the right value...

-AA
 

Re: kn and kp

MikeR said:
NMOS PMOS

AMIS

1500n
K' (Uo*Cox/2) 36.6 -11.7 uA/V^2
500n
K' (Uo*Cox/2) 56.2 -18.9 uA/V^2
350n
K' (Uo*Cox/2) 93.1 -22.2 uA/V^2

TSMC

350n
K' (Uo*Cox/2) 92.9 -32.1 uA/V^2
250n
K' (Uo*Cox/2) 125.9 -25.9 uA/V^2
180n
K' (Uo*Cox/2) 171.9 -36.5 uA/V^2

IBM

500n
K' (Uo*Cox/2) 77.0 -20.9 uA/V^2
350n
K' (Uo*Cox/2) 91.4 -22.7 uA/V^2
250n
K' (Uo*Cox/2) 110.5 -27.0 uA/V^2
180n
K' (Uo*Cox/2) 153.7 -32.8 uA/V^2
130n
K' (Uo*Cox/2) 276.0 -47.6 uA/V^2
Click to expand...

Is k' & Kp same?
or kp=k'/2?
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top