Designer’s Guide Community

Welcome, Guest. Please Login or Register. Please follow the Forum guidelines.

Apr 26^th, 2024, 3:10pm

‹ Previous topic | Next topic ›

Pages: 1

Jmelt Jmax for BSIM in spectre (Read 194 times)

jbdavid

Community Fellow

Offline

Posts: 378
Silicon Valley

Jmelt Jmax for BSIM in spectre
Jan 17^th, 2007, 4:13pm

I was hoping to use these, to get current density warnings, but spectre manuals tell me these are Amps/um^2..
I'm assuming 1 dimension is transistor W, but the other cannot be Length.. , has to be depth..
of WHAT?

What I want to specify is similar to TSMC manual, Jeff = Id/w

BSIM spec does NOT mention these parameters at all.. and Spectre manual doesn't explain how to get Jds from Ids!

Guess I have to write assertions for this instead..
If someone knows what thickness would be used, please post it here..

jbdavid
Mixed Signal Design Verification

IP Logged

marekm New Member Offline Posts: 8	Re: Jmelt Jmax for BSIM in spectre Reply #1 - Jan 17^th, 2007, 4:35pm I don't know about Spectre, but usually those explosion limits apply to p-n junctions so I would suspect that they refer to the s/d areas for the junction diodes.
Back to top	Marek Mierzwinski IP Logged

jbdavid Community Fellow Offline Posts: 378 Silicon Valley	Re: Jmelt Jmax for BSIM in spectre Reply #2 - Jan 17^th, 2007, 5:56pm Id wouldn't flow thru the junction diodes though.. and these aren't described in the Junction Diode parameter section. But I hadn't really considered that possibility.. jbd
Back to top	jbdavid Mixed Signal Design Verification IP Logged

Ken Kundert Global Moderator Offline Posts: 2384 Silicon Valley	Re: Jmelt Jmax for BSIM in spectre Reply #3 - Jan 17^th, 2007, 7:00pm I think marekm is right. I think those are parameters for the substrate junctions. -Ken
Back to top	IP Logged

Geoffrey_Coram Senior Fellow Offline Posts: 1998 Massachusetts, USA	Re: Jmelt Jmax for BSIM in spectre Reply #4 - Jan 18^th, 2007, 4:14am imelt, jmelt are in the diode section imax, jmax are in a section with alarm, vbox which aren't diode parameters. Jonathan - did you try it? Can you get a jmax warning with drain current? It might be that the units listed are wrong (or A/m^2 is used for the diodes, and A/m is used for the drain current). -Geoffrey
Back to top	If at first you do succeed, STOP, raise your standards, and stop wasting your time. IP Logged

Ken Kundert

Global Moderator

Offline

Posts: 2384
Silicon Valley

Re: Jmelt Jmax for BSIM in spectre
Reply #5 - Jan 18^th, 2007, 8:25am

Both the max and melt parameters apply to the junction diodes. The max parameters specify where the IV curves deviate from the expected exponential so as to improve convergence, and the melt parameters specify when a warning should be given. This is all explained in "spectre -h bsim3v3" under a special section titled "Imax and Imelt".

-Ken

IP Logged

jbdavid Community Fellow Offline Posts: 378 Silicon Valley	Re: Jmelt Jmax for BSIM in spectre Reply #6 - Jan 18^th, 2007, 3:46pm And Geoff verified that thru simulation.. - so for me thats a dead end, since what I need is Ids/Wmetal < Jmax Metal.. Appreciate the quick help!! jbd
Back to top	jbdavid Mixed Signal Design Verification IP Logged

Ken Kundert Global Moderator Offline Posts: 2384 Silicon Valley	Re: Jmelt Jmax for BSIM in spectre Reply #7 - Jan 19^th, 2007, 12:29am You might look into the new assertion features in Spectre. -Ken
Back to top	IP Logged

jbdavid Community Fellow Offline Posts: 378 Silicon Valley	Re: Jmelt Jmax for BSIM in spectre Reply #8 - Jan 22^nd, 2007, 8:50pm Reading about Jmax made me hopeful that I could avoid that.. but Geoff dashed those hopes.. guess its about time to play with those .. Jonathan
Back to top	jbdavid Mixed Signal Design Verification IP Logged

Andrew Beckett Senior Fellow Offline Life, don't talk to me about Life... Posts: 1742 Bracknell, UK	Re: Jmelt Jmax for BSIM in spectre Reply #9 - Jan 23^rd, 2007, 2:04am Hi Jonathan, assert is very easy to use. `spectre -h assert` gives you a very good starting point (as you'd expect). So there's no need to be fearful of using asserts! Regards, Andrew.
Back to top	IP Logged

Pages: 1

‹ Previous topic | Next topic ›