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.

[ADS] Spiral inductor parameterization in ADS

Status
Not open for further replies.

pragash

Advanced Member level 2
Joined
Jul 4, 2005
Messages
512
Helped
64
Reputation
128
Reaction score
59
Trophy points
1,308
Location
Oakland
Activity points
4,939
I designed and simulated my schematic and layout of a spiral inductor. the layout is created using "Generate and Update Layout" button in ADS. However, the correlation is not good between the schematic and the layout of the spiral inductor. I want to tune/optimize the spiral inductor layout. can anyone guide me on how can i do that? Any document also welcomed.

i believe its called parameterization.

the correlation between schematic and layout is given below. blue is schematic and red is the layout.

L2.png
 

Why do you have negative inductance? Please show a wideband plot down to DC.

my filter circuit required negative inductance in order to meet the goal of insertion loss and rejection.

attached the inductance from DC to 2.6GHz.

Inductor.png

Leff is circuit simulation whereas Leff2 is layout simulation.

Q and Reff.png

Please refer to the Q and Reff of the layout simulation of the negative inductor.

- - - Updated - - -

my filter circuit required negative inductance in order to meet the goal of insertion loss and rejection.

this is when i simulated with CPW components in ADS
 

Dr Volker,
i followed the steps you have given in the document and i got below error.

View attachment error message.txt

- - - Updated - - -

Dr Volker,
i followed the steps you have given in the document and i got below error.

View attachment error message.txt

this is solved.

- - - Updated - - -

Dr Volker,
i followed the steps you have given in the document. when i tried to optimize the coil to meet the circuit simulation, nothing happened. meaning, the coil parameters did not change.

i was intended to change N (number of turns), W (width) and S (Gap). i did Customize Pcell and re-simulated the layout but the optimizer still not able to change the N, W and S. Please help.

- - - Updated - - -

now im not getting any error message but the layout is not changing i believe..

- - - Updated - - -

the optimizer is saying no improvement is possible..
 

now im not getting any error message but the layout is not changing i believe..

Have a look again at my appnote, near middle of the page, title "Bug or feature?"
https://muehlhaus.com/support/ads-application-notes/em-parameters-momentum

I guess this is your problem: You need to change a setting from “Use this layout” to “Parameterized“.


my filter circuit required negative inductance in order to meet the goal of insertion loss and rejection.

Have you tried replacing the negative inductance by a capacitor?
 
Have a look again at my appnote, near middle of the page, title "Bug or feature?"
https://muehlhaus.com/support/ads-application-notes/em-parameters-momentum

I guess this is your problem: You need to change a setting from “Use this layout” to “Parameterized“.

parameterized.png

please refer to the attached pic. i did changed to parameterized but its not still not changing. i repeated the exercise many times but still the same..

- - - Updated - - -

Have you tried replacing the negative inductance by a capacitor?

I did but it did not work. I think it because my filter is seeing negative to positive reactance as the optimum point.
 

please refer to the attached pic. i did changed to parameterized but its not still not changing. i repeated the exercise many times but still the same.

That looks fine.
Have you defined all the cell parameters, and pass them to the spiral ?
You can test your parameterized cell by placing an instance into a new layout, and see if it changes when you modify parameters.

I can have a look at your element, please copy to a separate library/workspace and upload here.
 
Have you defined all the cell parameters, and pass them to the spiral ?
You can test your parameterized cell by placing an instance into a new layout, and see if it changes when you modify parameters.

no, I didn't do that. i only did every step in your link that you sent me.

- - - Updated - - -

View attachment LUMPED_COMPONENT_EM_SIMULATION_wrk - Copy.rar

attached the project. Please let me know how to define cell parameters and pass them to the spiral. I think thats what im missing.
 

Your workflow/method is not clean.

There must be no schematic in the cell, that is what causes you trouble: you have defined variables in the schematic VAR block. That is wrong. The values that you pass to the cell must be defined as cell parameters (design parameters). Only then, you can control the values when placing an instance with parameters.

pass_param.PNG

Delete the schematic, there must be no schematic in this cell. Define missing cell parameters and make the inductor use them.

cellparam.PNG

Use a separate cell for your schematic testbench.
 
Last edited:
Delete the schematic, there must be no schematic in this cell.

done. please see the pic below. my new layout name is "1".

1.png

- - - Updated - - -

Define missing cell parameters and make the inductor use them.
done. please see the picture below.

2.png

- - - Updated - - -

Use a separate cell for your schematic testbench.

also done. please refer to the pic below.

3.png

but unfortunately, my optimizer still not able to change the parameters!

- - - Updated - - -

if you able to optimize the spiral layout in the file that i sent you, please let me know. im still not able to do it

- - - Updated - - -

dr volker,
i didnt define Ri in the cell parameter above. i have redone the simulation after defining the Ri. However, my optimizer says "no improvement possible" in the 1st Iteration/Trial as usual! so no difference is detected by adding Ri.


hpeesofsim (*) 490.shp Nov 12 2018, MINT version 4
(64-bit windows built: Tue Nov 13, 2018 02:20:54 +0000)

OPTIM Optim1[1] <LUMPED_COMPONENT_EM_SIMULATION_lib:new-sche:schematic> optIter=(0->25000)

SP Optim1[1].SP1[0] <LUMPED_COMPONENT_EM_SIMULATION_lib:new-sche:schematic> optIter=0 freq=(2.3 GHz->2.6 GHz)

Iteration/Trial #0:
CurrentEF: 311.629187474297
Optimization variables:
X2.N_turns = 4.5487
X2.Ri_inner_radius = 616.705974612e-06
X2.S2_gap = 251.149866e-06
X2.W2x_width = 155.585668e-06

- - - Updated - - -

X2.Ri_inner_radius = 616.705974612e-06
X2.S2_gap = 251.149866e-06
X2.W2x_width = 155.585668e-06

i dont understand why "mil" is defined as e-06 here! Is this the rootcause?
 

done. please see the picture below.

View attachment 151556

But this is wrong - you have fixed values for the MRIND, instead of using the cell parameters.

To test, create a new layout in a new cell. Place an instance of you parameterized cell and change parameters. If it scales - great. If it does not scale, something is wrong.

- - - Updated - - -

I have created a quick example workspace for you

View attachment scale_mrind_wrk.zip

ind1.PNG

ind2.PNG

ind3.PNG

ind4.PNG
 
Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top