really need help!! Grounded Active inductor's Zin

[orz]

this is my circuit in ADS

this the theory's formula of Zin

this is the matlab program i used to calculate
the result is Z =

0.0001 - 0.1391i

but in the ADS


as you could see, the result is totally different. i am really confused. this is my homework please help! Thanks in advance!

 

[erikl]

1. I think the order of cox is wrong: in SI units this should be 8.4*10^-3 [F/m²]

2. The last contribution in the denominator of the Z_in(s) equation is cgd1 , whereas in your MATLAB equation you have cgs1
(but I'm not sure if this changes a lot).

 

[orz]

1. I think the order of cox is wrong: in SI units this should be 8.4*10^-3 [F/m²]

2. The last contribution in the denominator of the Z_in(s) equation is cgd1 , whereas in your MATLAB equation you have cgs1
(but I'm not sure if this changes a lot).

Thanks!
but after i corrected these two mistakes, the result remains the same.:-(

clear;clc;clf;

L=0.18*10^-6;
W=20*10^-6;
cgso=8.58E-10;
cgdo=8.58E-10;
cox=8.4222*10^-3;
gds1=0.00017156;
gds2=0.000172916;
gm1=0.006982440;
gm2=0.006982440;
cgs1=cgso+2/3*cox*L*W;
cgs2=cgso+2/3*cox*L*W;
cgd1=cgdo;
cgd2=cgdo;
w=2*pi*2*10^9;

Z=(gds1+j*w*(cgs2+cgd1+cgd2))/((gds1+gm1+j*w*cgd2)*(gm2+j*w*(cgs2+cgd1)))

 

[erikl]

2 more ideas:

1. You should write

cgs1=cgso+(2/3)*cox*L*W;
cgs2=cgso+(2/3)*cox*L*W;

... because otherwise MATLAB could interpret this as cgs2=cgso+2/(3*cox*L*W);

2. Are you sure MATLAB can differentiate between the upper case W for the transistor width and the lower case w which means ωf = 2πf ?

 

[orz]

2 more ideas:

1. You should write

cgs1=cgso+(2/3)*cox*L*W;
cgs2=cgso+(2/3)*cox*L*W;

... because otherwise MATLAB could interpret this as cgs2=cgso+2/(3*cox*L*W);

2. Are you sure MATLAB can differentiate between the upper case W for the transistor width and the lower case w which means ωf = 2πf ?

Thanks!

but after i changed these, the result remains the same.
Z =

0.0001 - 0.1391i

clear;clc;clf;

L=0.18*10^-6;
Width=20*10^-6;
cgso=8.58E-10;
cgdo=8.58E-10;
cox=8.4222*10^-3;
gds1=0.00017156;
gds2=0.000172916;
gm1=0.006982440;
gm2=0.006982440;
cgs1=cgso+(2/3)*cox*L*Width
cgs2=cgso+(2/3)*cox*L*Width
cgd1=cgdo;
cgd2=cgdo;
w=2*pi*2*10^9;

Z=(gds1+j*w*(cgs2+cgd1+cgd2))/((gds1+gm1+j*w*cgd2)*(gm2+j*w*(cgs2+cgd1)))

 

[erikl]

Z = 0.0001 - 0.1391i

With an Excel sheet calculation I get the same result: 0.00013662 - 0.13912i

Are you sure the ADS system uses the same equation?

 

[orz]

With an Excel sheet calculation I get the same result: 0.00013662 - 0.13912i

Are you sure the ADS system uses the same equation?

I used Zin in ADS , and I have tried to calculate the resistance only circuit, the result is ok.
I am now wondering if I have correct Cgs.
I got Cgso, Cox from mosis. Is it believable?

 

[erikl]

I used Zin in ADS , and I have tried to calculate the resistance only circuit, the result is ok.

Yes, but are you sure ADS uses the very same equation for Zin calculation?

I am now wondering if I have correct Cgs.
I got Cgso, Cox from mosis. Is it believable?

Yes, these values seem reasonable. The Cox value is right for a 180nm process.

However, the MATLAB result value is much more similar to an output impedance: Re(Z)=0.1mΩ , seems too low even for an output resistance. ABS(Z)=139mΩ also seems quite low.

 

[orz]

Yes, but are you sure ADS uses the very same equation for Zin calculation?

Well... Actually, i don't know how ADS calculates Zin. But i take it as the exact Zin of the circuit, and compare it with my formula(i also take it as the exact Zin).

Yes, these values seem reasonable. The Cox value is right for a 180nm process.

However, the MATLAB result value is much more similar to an output impedance: Re(Z)=0.1mΩ , seems too low even for an output resistance. ABS(Z)=139mΩ also seems quite low.

In this case, i really don't know what is wrong.
The Zin formula is from electronics letters. I guess it is right.

Thanks!