question about TSMC spiral inductor

when i use TSMC spiral inductor(ind3 std) i can set a same value with difference nr and rad now my question is which one is better smaller nr vs larger rad or larger nr vs smaller rad??

and the point that i show with question mark in the picture must be grounded or no itsnt necessary???

Maybe you should study passive spiral inductors in theory before trying to use them.

http://www.home.agilent.com/agilent...ckey=10102:eesof&lc=eng&cc=US&nfr=-34360.0.00

Read Niknejad's books: http://rfic.eecs.berkeley.edu/~niknejad/pubs.html

thanks for your references
but i study some notes about spiral inductor theory but i cant find my questions answer till now if you know these questions answer please tell me

Hi rezaee,

You must check values of Rs & Q for both and you has the answer_I think...
If you will calculate these values in some versions_you will find out that you has an optimal aspect of Diameter-outside to Diameter-internalside. You can also scale your coil to that-if you has enough place...
Good luck!
K.

rezaee said:

thanks for your references
but i study some notes about spiral inductor theory but i cant find my questions answer till now if you know these questions answer please tell me

Click to expand...

You can not find reference on differential inductors, ground planes and shields, and how to simulate/measure parallel resistance and quality factor??

Read this paper: https://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=1004055

Sorry, I dont know & read thes article...
But my be it helps you a lot more:
**broken link removed**
Electronically tunable floating inductance simulator - Elsevier
On-Chip Differential Mult-ilayer Inductor: www.freepatentsonline.com/7091814.pdf$
Comment on “Electronically tunable floating inductance simulator ...
Comment on “Electronically tunable floating inductance simulator ...
linkinghub.elsevier.com/retrieve/pii/S1434841109002258 (S. Minaei, O. Cicekoglu, H. Kuntman and S. Turkoz, ?)
K.
PS:
I wrote Rs, their=Serial resistance & L/Rs (as Q )has an optimum(maximal value) too.
Of course both are not at same parameters:-(...

Added after 3 minutes:

And this topics?:

THanks karesz
of course in the link you reffered i cant find my questions answer but i think your idea is true for calculation of Q i do this and this idea is a good idea for have a better choice but i cant find the answer for my second question i dont know when i simulate my circuit i should ground the point that i show in picture or no??

Hi,
Sorry-I overseed it...
If distances between dot & coil are in both (all) versions same_I think must be not relevant, than you are on the finding for relative values, differences..
K.

Added after 55 minutes:

You can find some description over the proposed calculations in a relative old book: ISBN 3-89649--430-9,
P. A. Passeraub: An integrated inductive proximity sensor_a thesis so 10 years before.
He finded out for exampl. that a thicker coil separation form Si (called SU8 process) brings ca. 4x higher Q_with same parameter as N, D_ and no correlation in diverse thicknesses of Si.
In 1. solution:
Standard 10um gold bumping layer over 4um Si dioxide. (usual TAB process)
In 2. solution:
SU8 thick photoresist is in process,
Copper lines are 60um thick,gap to Si(formed by SU8Layer) is 15um.
The self resonat frequency changed from ca. 22 to 35 MHz too... References: [2.27 & 2.28]

For me is important from ther book, chapter 2.4: "Flat coil in spiral":
Rint/Rout referenced to maximal L value is at ca. 0.12, but L/Rs max is at ca. 0.3.
I belief, if you se the diagramms, it is a good compromise at Ri/Ro by ca. 0.2...
K.