Integrating reworked spice3 into Cadence

[rahul]

Hi,

I have added new models of TFTs inside
spice3/ngspice and want to integrate spice3
with cadence using the OSS interface.

I have briefly gone though the OSS manual and
it seems that I need to customize the simcap.il
file.

Has anyone done such customization and can I get
an example simcap.il file from cadence (other than
the one from src/ directory) ?

Any help will be greatly appreciated

Thanks
-Rahul

ASU
Flexible Display Center (FDC)
Tempe, AZ

 

[Stuart Brorson]

rahul <rahuls@hrl.com> wrote:
: Hi,

: I have added new models of TFTs inside
: spice3/ngspice and want to integrate spice3
: with cadence using the OSS interface.

That's cool! Any chance you can submit some patches with this feature
enhancement to the ngspice project?

Stuart

 

[rahul]

Hi Stuart,
I certainly will, I have to do some more testing
The dc and ac analysis works fine and I am currently testing
the tran (NIintegrate portion is a bit fuzzy) and temp analysis
section.

I am a EE so my c-programming skills are not as sharp
as you or Paulo's.

The model is a RPI level-15 tft (n-channel) our process
doesn't have a p-device as yet and probably will never have
(the mobility of amorphous Si is orders lower than
crystalline Si and we can barely meet the specs at 15V Vdd)
Although I have added a patf variable with type=-1

Another problem is that I use "letter a" for atft
which I think might have conflict with xspice, I had to comment
out some portion to get around that.

I will send you the files once my advisor/group oks it.
There is nothing proprietary.

Thanks
-Rahul

ASU
Flexible Display Center (FDC)
Tempe, AZ




Stuart Brorson wrote:

rahul <rahuls@hrl.com> wrote:
: Hi,

: I have added new models of TFTs inside
: spice3/ngspice and want to integrate spice3
: with cadence using the OSS interface.

That's cool! Any chance you can submit some patches with this feature
enhancement to the ngspice project?

Stuart

 

[Kevin Aylward]

rahul wrote:

Hi Stuart,
I certainly will, I have to do some more testing
The dc and ac analysis works fine and I am currently testing
the tran (NIintegrate portion is a bit fuzzy) and temp analysis
section.

I am a EE so my c-programming skills are not as sharp
as you or Paulo's.

The model is a RPI level-15 tft (n-channel) our process
doesn't have a p-device as yet and probably will never have
(the mobility of amorphous Si is orders lower than
crystalline Si and we can barely meet the specs at 15V Vdd)
Although I have added a patf variable with type=-1

Another problem is that I use "letter a" for atft
which I think might have conflict with xspice, I had to comment
out some portion to get around that.

It is not advisable to use "a" unless it is an xspice code model. Add
another level to the mosfets.


Kevin Aylward
salesEXTRACT@anasoft.co.uk
http://www.anasoft.co.uk
SuperSpice, a very affordable Mixed-Mode
Windows Simulator with Schematic Capture,
Waveform Display, FFT's and Filter Design.