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.

[SOLVED] How to insert SRAFs using the resist model and optical model in Calibre?

Status
Not open for further replies.

CongJ

Newbie level 5
Joined
Dec 4, 2022
Messages
8
Helped
0
Reputation
0
Reaction score
0
Trophy points
1
Activity points
62
Hello,
I'd like to do SRAF insertion on the vias layout. I've got the optical model, resist model, and GDS, but I don't know what to do next?
I read Calibre's user guide, but only the templates that use litho model.
Thank you in advance.
 

If you can describe geometrically where you want the SRAFs, you could try OPCSBAR, which doesn't use models. The instructions for it are in the Calibre OPCsbar User's and Reference Manual. (calbr_sbar_useref.pdf in your Calibre installation)

Alternatively, it isn't that hard to turn models into a litho model; just package them in a directory with a Lithomodel file. The format is pretty basic. If you have a WORKbench license, there's a built-in lithomodel wizard to turn existing models into one. (Searching on "wizard" in the WORKbench manual should land you in the right chapter.) If you don't have WORKbench, you can do it manually.

1) Create a directory that is accessible from anywhere you are running Calibre.
2) In the directory, create a file named Lithomodel (no extension).
3) For just a resist model and optical model (assuming single exposure/not double patterning):

version 1
resist <filename>
mask 0 {
background <transmission>
mask_layer 0
optical <filename>
}


4) Replace <filename> with the name of the resist/optical model. Ideally, you copy them to the directory with the Lithomodel file, but relative and absolute paths also work; you just can't count on being able to copy the directory around where you need it.
5) For <transmission>, use whatever is on the "background" line of your setup file. (I'm assuming you have one of those.)

There are a bunch of other parameters you can specify (see "Litho Model Format" in the OPCverify manual), but it sounds like you have a simple case.
 

    CongJ

    Points: 2
    Helpful Answer Positive Rating
If you can describe geometrically where you want the SRAFs, you could try OPCSBAR, which doesn't use models. The instructions for it are in the Calibre OPCsbar User's and Reference Manual. (calbr_sbar_useref.pdf in your Calibre installation)

Alternatively, it isn't that hard to turn models into a litho model; just package them in a directory with a Lithomodel file. The format is pretty basic. If you have a WORKbench license, there's a built-in lithomodel wizard to turn existing models into one. (Searching on "wizard" in the WORKbench manual should land you in the right chapter.) If you don't have WORKbench, you can do it manually.

1) Create a directory that is accessible from anywhere you are running Calibre.
2) In the directory, create a file named Lithomodel (no extension).
3) For just a resist model and optical model (assuming single exposure/not double patterning):

version 1
resist <filename>
mask 0 {
background <transmission>
mask_layer 0
optical <filename>
}


4) Replace <filename> with the name of the resist/optical model. Ideally, you copy them to the directory with the Lithomodel file, but relative and absolute paths also work; you just can't count on being able to copy the directory around where you need it.
5) For <transmission>, use whatever is on the "background" line of your setup file. (I'm assuming you have one of those.)

There are a bunch of other parameters you can specify (see "Litho Model Format" in the OPCverify manual), but it sounds like you have a simple case.
Hello slizak,

Thank you for your answer.

I have tried litho_models obtained using the "wizard", but did not successfully insert the SRAF. The output file is still the same as before only vias.

I don't know if this is due to the litho model. I found the following warning in the log:

<
//-- WARN: nmSraf: Found zero magnitude transmission 0. This may result in all zero gradient. Check either layer commands or Lithomodel (if used) to ensure nonzero transmissions are specified for target and/or sbLayer.

//-- WARN: nmSraf: Detected very small overall maximum value of the gradient: 0. Please check the gradient and make sure that the target features are printing. Otherwise please consider using an open step (drc size or opc iterations) to construct and provide a correction layer with enlarged target shapes orconsider using a different gradType.
>

Thanks for your help.
 

That can happen if the background and layer use the same value -- they should be opposites. (In the simplest case, a dark/clear pair.) What are the lines for the "background" keyword and the "layer" keyword in the setup file? (Not the SVRF part, but the ones with the setlayer commands. Typically SVRF rules are in all caps, whereas the setup file uses lowercase, but this isn't an absolute.) Oh, and the mask_layer (plus its subblock) from Lithomodel would be helpful.

Because you didn't start with a litho model, the layer command would look something like "layer my_layer visible 0 0" or "layer my_layer visible clear" rather than "LAYER my_layer 17" (that one is the SVRF command). If it is just "layer my_layer", it was created to be used with a litho model from the get-go, and you may want to talk to whoever gave you the files.
 

That can happen if the background and layer use the same value -- they should be opposites. (In the simplest case, a dark/clear pair.) What are the lines for the "background" keyword and the "layer" keyword in the setup file? (Not the SVRF part, but the ones with the setlayer commands. Typically SVRF rules are in all caps, whereas the setup file uses lowercase, but this isn't an absolute.) Oh, and the mask_layer (plus its subblock) from Lithomodel would be helpful.

Because you didn't start with a litho model, the layer command would look something like "layer my_layer visible 0 0" or "layer my_layer visible clear" rather than "LAYER my_layer 17" (that one is the SVRF command). If it is just "layer my_layer", it was created to be used with a litho model from the get-go, and you may want to talk to whoever gave you the files.
Hello Slizak,

Thanks for your answer.

Your advice is what I need.

I modify the SVRF file according to your advice, but since I am a freshman in this, achieving it is still a little challenging.

I have started a conversation to show you my svrf file, could you give me some other advice?

Thanks.
 

Status
Not open for further replies.

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top