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

suggestion needed for BSIM4 in hspice

Status
Not open for further replies.
V

vikasnaidu

Newbie level 4
Joined
Jul 10, 2007
Messages
6
Helped
1
Reputation
2
Reaction score
0
Trophy points
1,281
Activity points
1,340
bsim4 in ads

I am trying to plot the id-vds curve for 32nm technology using the PTM files from ASU in the plot the id-vds curve is almost flat in saturation region even for l=160nm,32nm........ the spice engine is hspice 2006.03-SP1 even if i am using new version 2008 the same problem is there .Plz help me and let me know whether it is problem with tool or other because i feel syntax is correct for level 53 in MENTOR and HSPICE the values are same .But in level 54 BSIM4.0 problem is there

vikas naidu
 

bsim4 spice model extraction algorithm

Can you give us your problem more specifically ? I used BSIM4 before and BSIM4 has lots of more options than BSIM3.
 

id vds curve

Actually i am doing analog circuit design using BSIM4.0 level 54 mosfet model file in that for parameter extraction ie kp' lammda etc i plotted the id-vds curve but the plot is giving the lammda value as 0.05 V-1 even if the length is 32nm ie minimum length( i am simulating for 32nm node) .I need help whether i have to modify the model file or it is something else.I am attaching the id vds curves for both levls(54 & 53).
 

Hello,

I'm new to using BSIM. I just downloaded the source code for BSIM4 and I have no idea how to run this code and use HSPICE to get results. Could you please point me in the right direction? Is there a readme file that explains the set up?

Thanks.
 

aceyhan said:
Hello,

I just downloaded the source code for BSIM4 .
Click to expand...

What exactly have you downloaded? What you need are models, not the source code.

Keith.
 

Hello, Keith. Thank you for such a quick response. I have downloaded the source code at **broken link removed** for BSIM4.6.4. This is a C code. I am going to use the model cards from PTM at http://ptm.asu.edu/ for different technologies. I was expecting to find a .lib file written in SPICE syntax, which would model the nmos and pmos, but instead the code is written in C. That's why I'm confused. Do you mind checking this at the link that I have written above and let me know if I am actually on the right track to using BSIM4 models with HSPICE at your convenience?


Thank you
 

I don't know what PTM is about, but if you want to use BSIM4 models in Hspice then you need some .lib files as you expected. The Berkeley C code is for you to write your own simulator, I think, but you don't need that - you have Hspice.

The models I found at the PTM site looked like this:

* PTM Low Power 45nm Metal Gate / High-K / Strained-Si
* nominal Vdd = 1.1V
.model nmos nmos level = 54
+version = 4.0 binunit = 1 paramchk= ... etc

You should just be able to use those in Hspice.

They are under the menu "latest models":



They have a .pm extension, but are simply text files and should be fine with Hspice. You can always change the extension to .lib if you like.

Keith.
 

Thank you Keith, this has been very helpful. I was able to run the simulation for a simple netlist at the moment.

I'm writing now to clarify all the questions that I have in my mind. Apparently, I was confused about where to find the library files. It seems that the parameters in the .pm files are already defined in a built-in library file in HSPICE, am I right? Hence, it's possible to run simulations without an extra .lib file. Otherwise, how would SPICE know what to do with each parameter. I thought that I was supposed to use the C source code I was talking about in order to define how each of these parameters would be used to calculate MOS characteristics. I guess I was mistaken.

The netlist that I used at the moment is:

** Circuit Description **
.include modelcard.nmos

m1 2 1 0 b nmos L=0.09u W=10.0u NF=5 rgeomod=1 geomod=0
*+SA=0.5u SB=20u geomod=0 sd=0.1u
vgs 1 0 1.2
vds 2 0 1.2
Vb b 0 0.0

.dc vds 0.0 1.2 0.02 vgs 0.2 1.2 0.2

.print dc i(vds)

.end

In the instantiation for transistor m1, I used the model nmos. Is that what we really do when we want to use BSIM4 nmos model? Or is there another BSIM model that would be named, say, nmosbsim4?

Again, I thank you for your time and effort in responding to my questions.
 

aceyhan said:
It seems that the parameters in the .pm files are already defined in a built-in library file in HSPICE, am I right? Hence, it's possible to run simulations without an extra .lib file.
Click to expand...

The models are "complete" in that the contain everything required to run the simulation, but they aren't using anything built in to Hspice. Sometimes there is a separate params file which is also needed but that doesn't seem to be the case with the PTM models.

You need to be careful to make sure that your simulation is using the right models and not another library. Unfortunately they are often unimaginatively called "nmos"!

To check, rename the model in the PTM library so it is nmos4, for example i.e. the line

.model nmos nmos level = 54
would read

.model nmos4 nmos level = 54

and see if it does. If it still runs with the same results then you weren't really using the library. It will report the libraries it uses somewhere, but I don't use Hspice so I don't know where.

There should be a list file, .lst extension which should show the input "deck" and output data which should give a clue.

Keith.
 

Thank you very much. I changed the line and replaced nmos with nmos4 both in the PTM file and in the instantiation in my netlist file and the simulation results are the same as the ones that I got when I was using nmos instead. So, I'm not really using the BSIM4 library.

I will be looking into finding the library list and try to find out the model name for BSIM4.

I appreciate your help.
 

I think you have changed the model name too many times to prove anything. If you have changed it in two places it should still work. My idea was to change just the model file. The simulation should then give an error. If it doesn't, then it wasn't using the model file.

Keith
 

I mistakenly changed the handle twice, you're right. I corrected it and ran the whole thing again and the simulation aborts with an error, so I guess it is using the model file. Level 54 corresponds to BSIM4 model, so the results that I get should be what I'm really looking for.

Thanks.
 

can you explain me how hspice and bsim models are integrated. I am having licence for HSPICE2006-SP1 and want to use bsim4 model and want simulate subthreshold leakage, gate leakage power using 45nm PTM file from http://ptm.asu.edu/.
 

mallik.nimmagadda said:
can you explain me how hspice and bsim models are integrated.
Click to expand...

I am not sure what you mean by that. If you have a version of Hspice which supports BSIM4 and you have some Level 54 models (I think that is what Hspice uses for BSIM 4) then it will work. If your models are Level 47, 49 or 53 then they are not BSIM 4 but will run as the variants of BSIM3

Keith.
 

Hi Keith

I have read ur replies to different problems regarding BSIM. Can you give me some suggetions how to start with BSIM 4 as a new user. I am gonna use HSPICE simulator. waiting for ur reply
Sappana

---------- Post added at 21:58 ---------- Previous post was at 21:47 ----------

in the manual for BSIM 4 i c that when the depletion for gate resion is considered it is considered that the poly gate voltage has half the maximum poly electric field. Can anyone tell me the reason pls
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top