Skip to content
This repository has been archived by the owner on Oct 2, 2020. It is now read-only.

SANYOU SRD series relay models #548

Closed
wants to merge 1 commit into from
Closed

Conversation

qu1ck
Copy link
Contributor

@qu1ck qu1ck commented May 10, 2019

Models for these relays are based on datasheet and an example specimen found in my parts box.
Created in FreeCad and exported with stepUp toolkit.

Imgur

Datasheet with technical drawings:
http://www.sanyourelay.ca/public/products/pdf/SRD.pdf

PR for source file:
KiCad/kicad-packages3D-source#219

Thanks for creating a pull request to contribute to the KiCad libraries! To speed up integration of your PR, please check the following items as you complete them:

  • A screenshot of the model(s) aligned with KiCad footprints is very useful
  • Was the 3D model(s) created in a mechanical modelling program?
  • Source files (or scripts) have been contributed
  • Step model submitted in 1:1 scale
  • WRL model submitted in 1:2.54 scale
  • 3D model alignment checked against KiCad footprint
  • WRL model uses correct material properties
  • Geometry check performed against mechanical models

@CLAassistant
Copy link

CLAassistant commented May 10, 2019

CLA assistant check
All committers have signed the CLA.

@antoniovazquezblanco antoniovazquezblanco added Addition Adds new package 3D to library Pending reviewer A pull request waiting for a reviewer labels May 10, 2019
@Misca1234
Copy link
Collaborator

Hi

Are you interested in remaking with scripts instead of hand making them?
I can show you how, we will use the ac/dc scripts as a base.

@qu1ck
Copy link
Contributor Author

qu1ck commented May 28, 2019

I don't understand the requirement for scripting.
Sure, a unified parametrized script to generate multiple models is best. But in absence of it how is having some models not better than having no models.

I'm not saying I will not switch to scripting but I don't know when I will have the time to figure out the apis.
Don't think it should be a requirement.

@poeschlr
Copy link
Collaborator

poeschlr commented May 28, 2019

We only require scripting for families that already have a script prepared that can make them. (For all others it is only a bonus.)

Edt: For some reason i though these where dc/dc converters when i wrote this (which would have meant that yes should use the existing script) As we do not have a relay script it might not be as trivial to script them and should therefore really not be required (The dc/dc converter script might be easy to convert but i have not looked into it. Might be best to add a generic "box plus straight lead" generator that gets the lead positions, lead diameter, box size and color plus as parameters. Might be something i could look into next.)

@qu1ck
Copy link
Contributor Author

qu1ck commented May 28, 2019

@poeschlr that totally makes sense. But I hope when contributor makes effort to make a more accurate model we will favor that over script generated one.
In case of this relay I have a slightly recessed black bottom which exists on all such relays as they are compound filled on the bottom.
In case of #553 I have modelled tiny feet on the corners that are even on the technical drawings in the datasheet (although not all their dimensions are given).

@easyw
Copy link
Contributor

easyw commented May 28, 2019

@qu1ck
scripts are very powerful and can be handled to generate even highly detailed models.
image
Anyway scripts, as already noticed, are extremely useful most when there is a family instead of singular models and are very handy for helping librarians in maintaing models up to date (even in the simple event to change file name and not dimensions).
But I fully agree with you that in absence of a 3D model, having a manual created file is better than having no models 😄

@poeschlr
Copy link
Collaborator

poeschlr commented May 28, 2019

I do not think we should prefer manual made over scripted for any reason. (assuming of course the scripted version is of adequate quality.) As said if there is a script available it simply has to be used as maintenance reduction must be one of the highest if not the highest priority for the official lib (We are not lacking contributors but maintainers. Case in point see the number of open pull requests that have not even been looked at.)


Edit: to make it clear: This pull request is in the clear as we do not have a script for these parts. But you made a generic comment which i simply can not leave standing without comment.

@qu1ck
Copy link
Contributor Author

qu1ck commented May 28, 2019

Can you add an example of maintenance needed for models? Assuming they are not broken from the start, in which case they shouldn't be accepted in the first place, I can't think of a reason to ever touch them again after check-in. It's not like physical part changes over time and it's model morphs into new shapes. Someone mentioned file name change but having a script doesn't help here.

I agree that if there is a scripted model of adequate quality it shouldn't be replaced by hand made one just because it looks marginally better. Simply time spent to check new model isn't worth it. But the other way around shouldn't be the case imho, better models shouldn't be replaced by worse ones just because they are scripted.

@Shackmeister
Copy link
Collaborator

Maintenance so far has been:
Renaming (most common)
License changes
X/Y Adjustments to footprint
WRL material changes
Reexport for new step versions

All of these are much much easier maintained by scripts. Ofcourse alot of models are not worth scripting, but for similar series it's great

@poeschlr
Copy link
Collaborator

You missed one very important aspect. When we have a trusted script then we really only need to check the data entered into the size definition description for accepting assets. Without that we would need to open the model in freecad and check its dimensions manually against the datasheet. (From footprints i know that it takes minutes to check a scripted submission but can easily take half an hour to check a handmade one. Even if it is a very similar one. For models it is the same or even worse.)

@qu1ck
Copy link
Contributor Author

qu1ck commented May 29, 2019

Rene, that is one time check and doesn't count as maintenance, but Shackmeister's answer has valid points.
Consider me convinced, I'll try to stick to scripting whenever feasible. I can't commit to when I'll get around to scriptifying my current PRs but I'll get to it at some point.

@poeschlr
Copy link
Collaborator

poeschlr commented May 29, 2019

The main work of us maintainers is the first time check.

Edit: My goal is to reduce the workload of maintainers not necessary the discussion of what the word maintenance truly means.

@Misca1234
Copy link
Collaborator

@qu1ck
If it is ok with you, I can do the work for these rely's?

There is no scripts for relays today but I will simply copy the scripts for AC/DC, remove the AC/DC from the parameter file and add your relays.

I promise you, you comment afterward will be
"wow, this is so easy, why did I not do it the first time, I am glad I was not so persistent to get my hand made models merged"

@Misca1234 Misca1234 mentioned this pull request Jun 1, 2019
8 tasks
@qu1ck
Copy link
Contributor Author

qu1ck commented Jun 1, 2019

@Misca1234 by all means, go for it. I'll look at your scripts and learn from them and others already in the repo.

@Misca1234
Copy link
Collaborator

Misca1234 commented Jun 2, 2019

Here is the push
#564

If you are ok with it, you can close this thread

The key file is the
cadquery/FCAD_script_generator/Relay_THT/cq_parameters.py

This one define the transformers

The script was a copy of the DCDC script so the script was not written for the transformers

@qu1ck qu1ck closed this Jun 2, 2019
@antoniovazquezblanco antoniovazquezblanco added Invalid Wrong place or wrong question Scripting required and removed Pending reviewer A pull request waiting for a reviewer labels Jun 3, 2019
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
Addition Adds new package 3D to library Invalid Wrong place or wrong question Scripting required
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

7 participants