Active inductor simulation on cadence

Hi friends,

I am simulating an active inductor with resistive feedback on cadence (Fig.1). I am performing the two port s-parameter analysis (Fig.2), but the plot for the inductance is coming out to be negative (Fig.3). Shouldn't the inductance plot be positive? For the equivalent circuit, I am considering a resistance and inductance in series between the two ports. The formula I am using in the calculator is

(10**9)*imag(1/YP(1 1))/(twoPi*xval(YP(1 1))).

I have tried all I could think of to get the proper plot, but have been unsuccessful. Can someone please tell me where I am going wrong?

Thanks
Kas

hiiiii

i'm getting a similar plot when i found the inductance curve of the active inductor using sp analysis.were you able to correct the plot...?

can u help me?

Well, first off, i was not providing a Vdd (I don't know how I had overlooked that). Next, try changing the input port and the output port. Also, there will be a very small band within which the plot is +ve.

Try this. If you still don't get a result, message me again, I'll try and find out what is wrong.

Kas

https://obrazki.elektroda.pl/42_1285080103.png
https://obrazki.elektroda.pl/89_1285080103.png

this is my circuit and the waveform i'm gettin...
wen i put the ports ifound out that my dc operating points are changing...so i put a capa to block the dc...my circuit has to work at 434Mhz..so i put a capa value of 1uF...

can u chek it and tell me what is wrong
thanx 4 helping me

hey riajames,

i haven't had the time to go through your circuit. however, i did give it a glance, and you may try out the following suggestions, see if they work.

connect the base to the source for both the transistors, and provide a gate voltage. instead of connecting the drain of both the transistors together, make it a common source cascode configuration, with the drain of mn2 connected to source of n2. remove the ground from the source of n2. (look at the following two papers). also, you are using the one port simulation, so you might remove the 2nd port and just ground the drain of n2 through a resistor.

the papers:
1. S. Hara, T. Tokumitsu, T. Tanaka, M. Aikawa, “Broad-band monolithic microwave active inductor and its application to miniaturized wide-band amplifiers”
2. S. Hara, T. Tokumitsu, T. Tanaka, M. Aikawa, “Lossless broad-band monolithic microwave active inductors”

Hope this helps.
Kas

As far as i have tried, the funtions work well. I suggest you to check your DC biasing networks!
Regards.

hiii....
I used ac simulation to find my inductance and Q factor..I will recheck the s-p analysis later.
There is another problem....I implemented a cascode active inductor....it is meeting my specs....Now i have to implement tunability...What do you suggest is d best way...i'm thinking of changing the bias current..So my gm changes nd thus L....is der a better method?
Now i'm tryin to replace my ideal current sources with mosfet ones....I'm attaching my schematic with this....my C-D biasing current source is working perfectly....But my C-S cascode set-up current source is not working properly....
ie...wen i connect the current source to the cascode set up,my voltage across the current mirror mosfet is around 0V...so no current....
wat mite be d reason...
https://obrazki.elektroda.pl/58_1287719469.png
In the schematic i'v attacehd..i've placed the ideal current source itself...on the right hand side is my pmos current mirror which shud replace the ideal one....

please helppp!!!!!my submission date is nearing....
thnx
ria

hey, here's sumthing u can try. u must have assigned values to all the parameters while doing the analysis, right? like w, l, etc. instead of assigning values to them, just put them as variables... width of first transistor as w1, of the second as w2 etc, voltages as V1, V2, and so on. now when u open the window for analysis (i forget the name of the window) u have to first assign the values for all these parameters. next, instead of plotting the curve against frequency, choose one of these variables as the varying factor, and plot the curve against it, for some particular frequency (see the frequency plot and choose an optimum frequency). u will be able to tune your circuit varying most of these variables. optimize, and u will get the best results.
as far as the ideal current sources are concerned, i don't know if i can be of much help with that. just make sure ur dc biasing circuit is operating fine, and the mosfets u are using are aligned properly.

i hope this helps; do tell me if it works.
Kas

hi,

I want to ask another doubt...I implemented a tunable active inductor....Now i'm doing the corner simulations...Did you do corner for ur circuit ?how was it...how much variation u got form the typical value.....I'm getting very large variations....My typical value is 50nH and some of my corner sweeps ar showing values more than 100nH.....

riajames said:

hiii....
I used ac simulation to find my inductance and Q factor..I will recheck the s-p analysis later.
There is another problem....I implemented a cascode active inductor....it is meeting my specs....Now i have to implement tunability...What do you suggest is d best way...i'm thinking of changing the bias current..So my gm changes nd thus L....is der a better method?
Now i'm tryin to replace my ideal current sources with mosfet ones....I'm attaching my schematic with this....my C-D biasing current source is working perfectly....But my C-S cascode set-up current source is not working properly....
ie...wen i connect the current source to the cascode set up,my voltage across the current mirror mosfet is around 0V...so no current....
wat mite be d reason...
https://obrazki.elektroda.pl/58_1287719469.png
In the schematic i'v attacehd..i've placed the ideal current source itself...on the right hand side is my pmos current mirror which shud replace the ideal one....

please helppp!!!!!my submission date is nearing....
thnx
ria

Click to expand...

Can you tell me how to simulate the inductance, Q factor of the active inductor in cadence?
thanks!