Hi everyone! Thanks for the intro @User! As Matt said, I’m coming at this from the perspective of having a bit of knowledge of the traditional tooling, but a blank slate in terms of the open source (analogue) tools. I intend to try to document my process for getting up and running for others who join in a similar my situation to mine in the future. For my day job, I’m working on a Current-to-Digital converter based on a current starved ring oscillator. This is a scalable and relatively process agnostic architecture with no precision analog, so could be an interesting candidate for an ADC in a future shuttle. That said, I’m here as a pair of hands, and happy to contribute in whatever way the community feels most valuable!

Hi Anthony, there is also a technical committee in IEEE SSCS which wants to setup a website with useful tipps and hints how to get started with open source tools (I am one of the contributors for this site). I would happily use your experience and any document/guide you create for this site. Here’s the link, in the moment everything is just beginning: https://sscs-ose.github.io

Hi @User! I will certainly keep you up to date on the documentation in that case. I have just begun with a black Ubuntu VM, and will document the process from there.. here we go!

Hi @User I found @User’s Magic command cheatsheet quite handy and a nice compilation of commands and shortcuts for that important tool of the analog open source EDA toolflow which is Magic. Here is the link: https://github.com/hpretl/iic-osic/blob/main/magic-cheatsheet/magic_cheatsheet.pdf Thanks to Harald for sharing this resource

Hi @User! Thank you very much for your message. One of the first circuits I intend to try get working is a port of the current starved ROSC I taped out on 65nm last year. I’m still working on pipecleaning my toolchain - almost there I think. Once I’ve got a functioning flow, I’ll take a look at your repo and understand what you’ve done a little better. I’ll also come along to your webinar this afternoon. PSS and PNOISE being missing from ngspice is a pain, but using these circuits in ADCs means we mainly run long transients to obtain FFTs anyway (it’s too easy for weird effects to hide in linearised periodic models unfortunately). Thanks again for your helpful pointers!

Hi Anthony, ngspice does have PSS support just no pnoise atm. The long transient + FFT approach is one thing I considered also as a workaround but I think that could be a partial solution to the lack of pnoise but perhaps not a complete solution, as a long transient wouldn't have the intrinsic device noise contributions I guess? One could add transient noise sources on supply, Vctrl, etc and get an idea of noise transfer from those nodes to the output which would be good information but I think we still wouldn't see the intrinsic device noise contributions of the devices, so not sure what would be the best approach overall.