Plasmac Post Processors

More
27 Aug 2020 06:38 #179663 by phillc54
Replied by phillc54 on topic Plasmac Post Processors
I guess it would be possible to add all of the PlasmaC specific codes.

Please Log in or Create an account to join the conversation.

More
31 Aug 2020 20:36 #180223 by ts_7
Replied by ts_7 on topic Plasmac Post Processors

phillc54 wrote: That seems like a similar error that was fixed here:
forum.linuxcnc.org/plasmac/38438-plasmac...sors?start=10#177876


Yes, I had same situation. Decimal delimiter was the cause.
If anyone else faces the same problem note your locale are to be set at Sheetcam settings as well as at your system settings.
I set both to en_US.UTF-8 UTF-8

In Debian you can see your currently set delimiter by command:

locale decimal_point

Then you can modify or add correct one by:

dpkg-reconfigure locales


Language should correspond to Language Tab of Sheetcam options settings.

I hope this may help to whoever may have the same trouble.
The following user(s) said Thank You: phillc54, tommylight

Please Log in or Create an account to join the conversation.

More
31 Aug 2020 21:02 - 31 Aug 2020 21:06 #180235 by ts_7
Replied by ts_7 on topic Plasmac Post Processors
This is related to Sheetcam - There was a discussion on Plasmaspider about small circle rules. The idea was to use set of rules and code snippets to relevant size of circle. Smaller circle diameter - slower feedrate. Within one Path rule set. The processor picked the smallest applicable size (rule relevant to that size) and calculated Feed rate to be translated to g-code F value.
This time with new processor I get all rules/size options/code snippets into the g-code. Irrelevant to evaluating the circle diameter.
For example in Sheetcam:
rule 1: for circles smaller than 10mm feed rate to 40%, code "before" sets feed rate to 40%, code "after" sets feed to 100%
rule 2: for circles smaller than 15mm feed rate to 60%, code before sets feed to 60%, code after resets feed to 100%
rule 3: for circles smaller than 25mm feed rate to 70%, code before sets feed to 70%, code after resets feed to 100%
This would produce 3 lines of g-code setting speed to 40%, 60% and 70%. All three code snippets translated to ngc.
I wonder if sequence of rules plays any role? Sorry for bothering, may be this should be discussed at Sheetcam board?
Last edit: 31 Aug 2020 21:06 by ts_7.

Please Log in or Create an account to join the conversation.

More
01 Sep 2020 09:01 #180281 by rodw
Replied by rodw on topic Plasmac Post Processors
You can still do this in Sheetcam. the codes to use will be in the plasmac docs. I did share some screen dumps of the rules to use. It will be in a thread in the plasma section.

I want to do some more experimentation on this. Those numbers probably came from a guy called SeanP. I personally think the velocity reductions to use are not as simple as that. eg. if you have different acceleration to SeanP your values might be quite different. Long term, I want to be able to calculate the smallest radius that can be negotiated without slowing down and work from that. eg your machine might only make a 30mm radius at cut velocity, my machine might be able to do a 15mm radius as it has better acceleration. PLus this radius will change based on the cut velocity...
The following user(s) said Thank You: ts_7

Please Log in or Create an account to join the conversation.

More
01 Sep 2020 09:55 #180285 by phillc54
Replied by phillc54 on topic Plasmac Post Processors
I had something similar for a few different hole sizes in the gcode filter but I removed it as I wasn't sure it was worthwhile for air plasma.
The following user(s) said Thank You: ts_7

Please Log in or Create an account to join the conversation.

More
01 Sep 2020 15:22 #180336 by ts_7
Replied by ts_7 on topic Plasmac Post Processors
Yes. Hmmm, I did not consider acceleration. Thank you

Please Log in or Create an account to join the conversation.

More
06 Sep 2020 15:22 #181018 by DenkBrettl
Replied by DenkBrettl on topic Plasmac Post Processors
So, I had the same issue with the "attempt to compare number with nil" error and was able to debug this further.
I'm using the Linux version of SheetCam and it turns out that I environment variables set that change the number format to German. This triggers the issue as noted before in the thread. There are two possible solutions to the problem

1) Use
unset LC_NUMERIC
in a terminal before starting SheetCam.

2) I changed the code of the postprocessor, I added those three lines after line 61 (after the if scale==metric block):
if not tonumber(minimumMove) then
    minimumMove = minimumMove:gsub("%.", ",")
end

This replaces the . with a , in the case where the dot doesn't produce the right result.
Maybe something for version 'G'? :-)

Klaus
The following user(s) said Thank You: phillc54, ts_7

Please Log in or Create an account to join the conversation.

More
07 Sep 2020 06:39 #181098 by phillc54
Replied by phillc54 on topic Plasmac Post Processors

DenkBrettl wrote: So, I had the same issue with the "attempt to compare number with nil" error and was able to debug this further.
I'm using the Linux version of SheetCam and it turns out that I environment variables set that change the number format to German. This triggers the issue as noted before in the thread. There are two possible solutions to the problem

1) Use

unset LC_NUMERIC
in a terminal before starting SheetCam.

2) I changed the code of the postprocessor, I added those three lines after line 61 (after the if scale==metric block):
if not tonumber(minimumMove) then
    minimumMove = minimumMove:gsub("%.", ",")
end

This replaces the . with a , in the case where the dot doesn't produce the right result.
Maybe something for version 'G'? :-)

Klaus

Thanks for that. I use a Linux version as well.
If anyone else out here can confirm that method 2 causes no issues with the Windows version then I will add it and make a version G

Please Log in or Create an account to join the conversation.

More
07 Sep 2020 06:58 #181100 by rodw
Replied by rodw on topic Plasmac Post Processors

phillc54 wrote:

DenkBrettl wrote: So, I had the same issue with the "attempt to compare number with nil" error and was able to debug this further.
I'm using the Linux version of SheetCam and it turns out that I environment variables set that change the number format to German. This triggers the issue as noted before in the thread. There are two possible solutions to the problem

1) Use

unset LC_NUMERIC
in a terminal before starting SheetCam.

2) I changed the code of the postprocessor, I added those three lines after line 61 (after the if scale==metric block):
if not tonumber(minimumMove) then
    minimumMove = minimumMove:gsub("%.", ",")
end

This replaces the . with a , in the case where the dot doesn't produce the right result.
Maybe something for version 'G'? :-)

Klaus

Thanks for that. I use a Linux version as well.
If anyone else out here can confirm that method 2 causes no issues with the Windows version then I will add it and make a version G


Probably a question for Les Newall on the Sheetcam forum (or by email).

Please Log in or Create an account to join the conversation.

More
07 Sep 2020 07:52 #181104 by phillc54
Replied by phillc54 on topic Plasmac Post Processors

rodw wrote: Probably a question for Les Newall on the Sheetcam forum (or by email).

I'll give it a shot.

Please Log in or Create an account to join the conversation.

Moderators: phillc54
Time to create page: 0.274 seconds
Powered by Kunena Forum