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

Bulk of spiral inductor is not recognized !

Status
Not open for further replies.
M

Majid Rafei

Junior Member level 2
Joined
Aug 25, 2011
Messages
20
Helped
0
Reputation
0
Reaction score
0
Trophy points
1,281
Location
Tehran
Activity points
1,463
Dear all.

I am using 0.13 um PDK in cadence. In my circuit, I have 8 spiral inductors, which bulks are connected to the ground by means of a resistor. For more detail please refer to my post:

After doing LVS using Calibre, I have encountered an error, which says that bulk of spiral inductors are connected together. Someone addressed the problem, that Calibre does not recognize the bulk of spiral inductors.

Do anyone know how can I overcome this problem?
 
Last edited:

Perhaps the extraction of inductor bulk needs a specific
recognition layer. Perhaps the resistor you used, extracts
as a short circuit when it comes to inductor connectivity,
or perhaps the bulk is indeed the bulk handle / epi and
resistor trickery is correctly bypassed in favor of the real
global vertical connection.

I think inspecting the extracted netlist is the place to
begin.

Cadence is not responsible for the extract rules and
how they behave.That's the foundry PDK developers'
deal. Same for Caliber.
 

Dear dick_freebird,
Thanks for your attention.
I saw the extracted netlist of the layout. Inductors are known like follow:

.SUBCKT spiral_std$$209091628 PLUS MINUS 4
** N=5802 EP=3 IP=0 FDC=1
X0 PLUS MINUS 4 spiral_std w=1.5e-05 nr=4 rad=3e-05 lay=8 $X=-103500 $Y=-46055 $D=50
.ENDS

but the corresponding device in library is:

.SUBCKT spiral_std PLUS MINUS BULK
.ENDS

Therefore, BULK of spiral_std is not recognized. I don't know how and where to check for the error. Maybe, It is because of a typos in library of spiral_std or its layout definitions!
The PDK is from TSMC and many other guys work with it without error.

Thanks.
M. Rafei.
 

OK, so you have a node "4" assigned to one bulk terminal.
Now what else (if anything) is connected to "4"? What are
other inductors' bulk nodes mapped to? Is your bulk-degenerating
resistor connected to "4" at one end, or is the intended end
instead floating? Is there a disconnect in the extracted net(s)
if you try to follow from bulk-local, to resistor, to bulk-global?

The "library" (I'd assume this means the PDK models
hierarchy) subcircuit looks empty, thus malformed. Perhaps
there is some adjunct model hierarchy intended to replace
this null subcircuit with a useful one?

It does not appear that the assignment of BULK is a problem,
but perhaps the node assignment is not as envisioned.
 

Dear dick_freebird,
Thanks for the response.
Terminal 4 is the name of BULK terminal in subcircuit.
Terminal 4 of the subcircuit is named 27 in the netlist. BULK of all of the inductors are connected to the node 27.
Yes, intended end of resistors are float.
I think I have to use another (or new) PDK !
Thanks a lot.
 

So, sounds like a contact / connection extract issue at the
bulk tap-point. You might inspect your contact layout there.
Maybe there is a default assignment that is only overridden
by a well formed explicit contact (or, maybe the decision was
to make the bulk global because it is indeed global, and just
ignore any topside connection - but that seems wrong, to me.

Anyway, before you give up on the PDK make sure you don't
just have a low level construction problem with the cell as-
placed; maybe adding the right kind of contact, is on you
instead of a PCell built-in.
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top