SkyWater PDK

[Abel Joseph John]

Will Electric add support for the to be released SkyWater PDK?

[Joselito Morallo]

Dear Electric Group:

We can propose to electric users to work together and collaborate to create a SoC for the 180 nm.

Google plans s free shuttle tape put for interested chip makers to use the process for both analog and digital design, what is good from this process is they have already thr digital logic cells. 

A collaboration for analog design engineers to use electric to make the analog cells with one goal.

Both 32/64 bit Core can be used as efabless Tim Edwards mention in some workshop.

I can both participate in doing analog design and layout.

Best Regards, 

Joselito 

On Mon, Sep 14, 2020, 11:07 AM Abel Joseph John <eddy....@gmail.com> wrote:

Will Electric add support for the to be released SkyWater PDK?

--
You received this message because you are subscribed to the Google Groups "Electric VLSI Editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email to electricvlsi...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/electricvlsi/8d6feba0-e425-47ff-a0b9-4acf8ab0a3b1n%40googlegroups.com.

[Estevao Teixeira]

I have interest on development of an analog/mixed-mode design using 180nm SkyWater on Electric. Do you know how can we get SPICE models for this process?

Regards,

Estevao

[Joselito Morallo]

Dear ElectricVLSI Group:

We can start reading the docs of the skywater sky130 pdk.

There are Files that you can run simulation to design analog and mixed integrated circuits. 

We can check together the documentation if they already exist.

Best regards,

Joselito 

To view this discussion on the web visit https://groups.google.com/d/msgid/electricvlsi/23204cdc-4723-473f-b1b4-f29c28a01378n%40googlegroups.com.

[Joselito Morallo]

Dear ElectricVLSI Group:

This is also a follow up in support to Sir Estevao Teixeira, 

Does anyone here can help provide the HSPICE Simulation Tool?

Schematic drawing can be done using ElectricVLSI and check compatibility else, we need open source community to help enable it in ElectricVLSI to do Analog design - - >this will be another proposal. 

Best Regards, 

Joselito 

On Mon, Sep 14, 2020, 7:09 PM Estevao Teixeira <est...@engenharia.ufjf.br> wrote:

To view this discussion on the web visit https://groups.google.com/d/msgid/electricvlsi/23204cdc-4723-473f-b1b4-f29c28a01378n%40googlegroups.com.

[Abel Joseph John]

All things related to  Sky130 is supposed to be happening in a slack channel here, https://invite.skywater.tools/ (there is none for Electric)

You can find the necessary files for this here, https://github.com/RTimothyEdwards/open_pdks/tree/master/sky130 it generates the spice, lib, LEF, etc for magic and KLayout.
Is there any documentation on how to add a new PDK to Electric?

[Joselito Morallo]

Dear Electric VLSI Group:

I think we write an email/letter to Skywater Foundry on our Initiative to collaborate with them using their Sky 130nm PDK as 

Abel Joseph John wrote to us.

Best Regards,

Joselito

To view this discussion on the web visit https://groups.google.com/d/msgid/electricvlsi/2854adc8-13f7-4c6b-ad2b-8fd9b8d65a15n%40googlegroups.com.

[Patrick Mendes]

I am also interested in using the technology and willing to contribute to having a PDK version to Electric. I just don't know which steps should be taken. One thing is to use a tool. Another one is to develop things to be used with it. With any guidance, I could work on this. My interests are also in analog/RF/mixed-signal systems.

I am pretty satisfied with NGSpice as a simulator, so far. It needs more coding but is free. I did some comparisons with the same spice models with NGSpice and HSpice, when I had access to it. There were no differences in the results, for different models in different circuits/analyses. 

Best regards, Patrick.

To view this discussion on the web visit https://groups.google.com/d/msgid/electricvlsi/CALy8%2BG6pZ80Xmeb4zr6wZ%2BDpukLxws5dU8Z7DUsb-FDr7fZcrQ%40mail.gmail.com.

--

Patrick Mendes dos Santos, Professor, M.Sc.

Departamento de Engenharia Elétrica / Electrical Engineering Department

Federal Center of Technological Education of Minas Gerais (CEFET-MG).

Campus II, Av. Amazonas, 7675, Nova Gameleira. CEP: 30510-000.

Belo Horizonte, Minas Gerais. Brazil.

Phone: +55  31 3319 6834. 

e-mail: patric...@gmail.com

Website: http://www.eng-eletrica.bh.cefetmg.br/

[Steven Rubin]

I'm happy to see all this interest in Electric's use of the SkyWater PDK. If I can be of assistance in this effort, let me know. Those who are familiar with the foundry rules will have to do the real work, but I'm willing to advise.

   -Steven Rubin

To view this discussion on the web visit https://groups.google.com/d/msgid/electricvlsi/CACog4cHv%3DnoUdq1zduOHgDqR%3D1O%3DFEqa1ebQE6QobcVchqRR%2BQ%40mail.gmail.com.

[R K Tiwary]

one can install LTSPICE for verifying the circuit developed in Electric. 

Regards

On Mon, Sep 14, 2020 at 6:46 PM Joselito Morallo <jpmo...@gmail.com> wrote:

electr...@googlegroups.com is not on your Guest List | Approve sender | Approve domain

Dear ElectricVLSI Group:

This is also a follow up in support to Sir Estevao Teixeira, 

Does anyone here can help provide the HSPICE Simulation Tool?

Schematic drawing can be done using ElectricVLSI and check compatibility else, we need open source community to help enable it in ElectricVLSI to do Analog design - - >this will be another proposal. 

Best Regards, 

Joselito 

On Mon, Sep 14, 2020, 7:09 PM Estevao Teixeira <est...@engenharia.ufjf.br> wrote:

To view this discussion on the web visit https://groups.google.com/d/msgid/electricvlsi/23204cdc-4723-473f-b1b4-f29c28a01378n%40googlegroups.com.

--
You received this message because you are subscribed to the Google Groups "Electric VLSI Editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email to electricvlsi...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/electricvlsi/CALy8%2BG6pZ80Xmeb4zr6wZ%2BDpukLxws5dU8Z7DUsb-FDr7fZcrQ%40mail.gmail.com.

--

R.K.Tiwary

Asst. Professor(off Campus), EEE
BITS , Pilani

Tower-B, H2-502

Classic Residency

Rajnagar Extn.

Ghaziabad 201003
mob-9871482697

resd-0120-2986502

rktiw...@gmail.com

Birla Institute of Technology & Science, Pilani 

Pilani Campus Vidya Vihar, Pilani - 333031, Rajasthan, India

[Joselito Morallo]

Hi RK! 

I got this link with your information about LTSPICE someone also could look possibility for analog IC Design Simulation https://www.analog.com/en/design-center/design-tools-and-calculators/ltspice-simulator.html

Best Regards,

Joselito Isaac P. Morallo

Open Source HW/SW Advocate

Republic of the Philippines

To view this discussion on the web visit https://groups.google.com/d/msgid/electricvlsi/CABe4omd6AVaJ_Bbm41Fczh2%2BYCe-vcXq-b0o9HB%2BNOXtSYzc-g%40mail.gmail.com.

[Patrick Mendes]

I still prefer NGSpice. Compared to LTSpice its main drawbacks are (i) some useful simple functions implementations (step param and param type measurement, for instance); (ii) data visualization; (iii) and the lack of a really usable GUI (there are a few, but I dislike them). LTSpice data visualization is also a problem with several bugs unsolved since it was still registered under Linear Technology. The lack of GUI would help at least for code debugging, but you can get used to working without it. LTSpice convergence is still superior, though. And the results of LTSpice are different from NGSpice for the same simulation, due to its internal solver.

NGSpice also lets you control the flux of the simulation with the use of well-known programming structures (if-else, while, dowhile, foreach). For simple simulations, you can also run it in batch mode and export the results to .raw file that can be opened inside LTSpice. But only if they are voltages and currents. You couldn't plot the resistance variation of an NTC, for instance.

But the main reason for NGSpice is to have access to transistor internal parameters. LTSpice doesn't let you use them in your calculations. LTSpice only shows you the small-signal parameters if you use .op and .ac together, and the parameters shown are only for seeing and based in the initial solution of the circuit. If you want to investigate the evolution of a MOSFET transconductance with a biasing voltage, for instance, you can't do it with LTSpice. Moreover, if your transconductance investigation has to be carried out with transient analyses, forget it.

Lastly, NGSpice it's not only Spice (http://ngspice.sourceforge.net/presentation.html). It supports another type of simulations together with the traditional spice ones, increasing the number of design types you can do with it. And, there are alternatives (that I didn't try yet) like PySpice, using NGSpice with Python and probably providing easier to use plotting tools for NGSpice simulations.

I'm still getting used to the whole set of commands, instructions, and the need for specific coding for simple tasks of NGSpice but the more I use the more I see it's better for my purposes.

Kind regards, Patrick.

To view this discussion on the web visit https://groups.google.com/d/msgid/electricvlsi/CALy8%2BG5bebc1vi%2BkEZB3PZsux_QcQCZ%3DPaJyeBZ7qRgoPDAY4A%40mail.gmail.com.

[Adam Joseph]

I will be releasing an open source Electric technology for SkyWater130
soon.

In the meantime, here's a preview: a GDS import of
sky130_fd_sc_hs__dfstp_1 from SkyWater's digital cell library. The 3D
view is (I believe) "to scale" horizontally+vertically.

When I heard about the SkyWater PDK in late June it was missing all the
design rules, layer definitions, and mask tables. So I wrote it off as
being unserious. From the git history it looks like they added that
stuff two weeks after the public announcement, but I didn't find out
until two days ago. Wish I'd have known earlier!

All the essential stuff is there [*]. Barely.

Does anybody know where I can get the sealring+pad cells? Those will
answer a lot of my questions about how the I/O FET gox+implants work --
and make prettier eye candy, too.

- a

[*] Except the transistor SPICE models -- I guess I'm not looking
in the right place for them.

[Adam Joseph]

On Sun, 20 Sep 2020 19:42:36 -0700
Adam Joseph <ad...@westernsemico.com> wrote:
> Does anybody know where I can get the sealring+pad cells?

Never mind, found them, see below. I think the second image is the
output driver for the digital pad.

> [*] Except the transistor SPICE models -- I guess I'm not looking
> in the right place for them.

Still looking for these.

Got all the width/space rules in, crude DRC checks work. More tomorrow.

- a

[Joselito Morallo]

Hi Adam!

I noticed that skywater-pdk has only the latest files at the moment after cloning it.

1. sky130_fd_sc_hd/		1.8V digital logic (high density)
   sky130_fd_sc_hdll/		1.8V digital logic (high density low leakage)
   sky130_fd_sc_hs/		1.8V digital logic (high speed)
   sky130_fd_sc_ls/		1.8V digital logic (low speed)
   sky130_fd_sc_ms/		3.3V digital logic (medium speed)

some files are not in the repository, do you have ways to get these?

         sky130_fd_sc_hvl/                          3.3V digital logic 

         sky130_fd_sc_lp/                           1.8V digital logic (low power)

     sky130_fd_io/		   Standard I/O

     sky130_fd_pr_base/		   Primitive devices w/fixed layout
     sky130_fd_pr_rf/		   RF primitive devices w/fixed layout
     sky130_fd_pr_rf2/		   Additional RF primitive devices w/fixed layout

Thank you & Best Regards,

Joselito

--
You received this message because you are subscribed to the Google Groups "Electric VLSI Editor" group.
To unsubscribe from this group and stop receiving emails from it, send an email to electricvlsi...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msgid/electricvlsi/20200921044356.7e2fa69f%40ostraka.

[Joselito Morallo]

Hi Adam!

nevermind for the other libraries, got it already.

sky130_fd_sc_hd/		1.8V digital logic (high density)
sky130_fd_sc_hdll/		1.8V digital logic (high density low leakage)
sky130_fd_sc_hs/		1.8V digital logic (high speed)
sky130_fd_sc_ls/		1.8V digital logic (low speed)
sky130_fd_sc_ms/		3.3V digital logic (medium speed)

sky130_fd_sc_lp/                1.8V digital logic (low speed)

sky130_fd_sc_hvl/               3.3V digital logic

also,

sky130_fd_pr

sky130_osu_sc

sky130_osu_sc_t18

some files are still not in the repository, 

Do you have other sources of previous publication and have ways to get these?

sky130_fd_io/		       Standard I/O

sky130_fd_pr_rf/               RF primitive devices w/fixed layout
sky130_fd_pr_rf2/	       Additional RF primitive devices w/fixed layout

Thank you & Best Regards,

Joselito

[Gavin]

According to the webpage at

https://skywater-pdk.readthedocs.io/en/latest/known_issues.html

it seems that they plan to release them at a future date such that it doesn't look like they have released them to the public yet.

[Adam Joseph]

Hi Joselito!

The only two sources I have used are this (clone *recursively*):

https://github.com/google/skywater-pdk

... and for the pads, I found them here:

https://github.com/RTimothyEdwards/open_pdks/tree/master/sky130/custom/sky130_fd_io/gds

The license for RTimothyEdwards' repository is set to "Apache-2.0
License".

I don't have anything else for SkyWater130.

I found that the CSV files in the skywater-pdk under docs/rules have a
lot of useful information.

- a



On Wed, 23 Sep 2020 10:33:54 +0800

> > 1.