Fusion 360 - Lathe Post ??

Andy:

You are right. Thank you for the correction. I swear I will learn to read soon.

Hugh

Have a look at how VCS handles Fusion post editing and be prepared to be rather impressed. I'm no Javascript expert but this helped me to develop a custom post processor.
knowledge.autodesk.com/community/screenc...334842276.1561580970
This guide is also pretty helpful
cam.autodesk.com/posts/posts/guides/Post...Training%20Guide.pdf

Muzzer:

Thank you. spumco mentioned Visual Studio Code above. I found more information through Google and downloaded a copy. Have been having trouble getting it to run as advertised though.

I also just found the Post Training Manual/Tutorial you mention. In a YouTube video by Russtuff he mentioned it. I'm hoping it has the info I need to get VSC working well. I just need to find the time and energy to study it.

From videos of what VSC can do, it sure would make modifying posts for Fusion easier.

But I fear neither will answer the burning G95 question.

Thanks.

Hugh

Well the LinuxCNC documentation doesn't really seem to clarify G95 usage. Why not do some testing yourself to see if your suspicions are justified? You can always modify the PP if you think that is appropriate. Perhaps the finishing instruction section generated by the std PP doesn't reset it when it should. Presumably this issue would only happen if you invoked a specific move in MDI anyway, which may be unlikely unless you really know what you are doing.

I'm away from my home and workshop ATM but may look into this tomorrow when I get back.

As you may have noted, Path Pilot is built on LinuxCNC, so perhaps there is better documentation available from Tormach?

VCS seems pretty solid and has great add-ins. You may benefit from talking the time to learn how it is used, otherwise many of the features may pass you by. The add-ins are held in a sort of app store, if you know how to find that from the GUI.

Muzzer:

I may do a little testing. I'm curious what the Post does (regarding G95) with operations other than threading. I have little idea what I'm doing, but I use jog and MDI a lot on my mill as a substitute for a manual machine. I suspect the lathe will be the same.

I may post my inquiry on the Fusion 360 Post Forum. I searched that forum for LinuxCNC Lathe Post and did find some information, but nothing on G95. I'm sure someone there wrote the LinuxCNC Lathe and Tormach Slant Bed posts.

I'd like to get a handle on VS Code before doing post testing though. I think it will help a lot. To get there I'm working through Autodesk CAM Post Processor Guide. I'm also in the middle of converting a 10" lathe and it's not ready yet. Although, my Sherline Lathe would work for this testing.

Thank you for your help. I'll let the list know if/when I make any progress.

Hugh

Looking into this a bit more closely, there may not be much of a mystery here. G95 (per revolution feed) is pretty much the default setting for lathe operations, while G94 (per minute feed) is the default for milling. When you generate lathe g code from Fusion, most (all?) of the files start with a G95, with not a G94 to be seen. Looking more closely at the post processor code, I don't see how it would actually generate G94. Put another way, there's no reason to ever use G94 on a lathe. You were concerned that manually issuing an F feedrate when G95 is active could cause a problem if you believed G94 to be the default. Apart from the fact that it's not clear when / why you'd do that, it would also seem to be quite unconventional.

I don't know when the LinuxCNC g code set was defined and by whom but they appear to conform to the "Fanuc Group Type B" - and within that, G94 and G95 exclusively occupy "modal group 5". In other words you can only have G94 or G95. It seems you can set an initial  default by configuring the "RS274NGC_STARTUP_CODE =" line within your ini file. Mine simply has G7 (diameter mode) and G18 plane select) ie no G94 or G95. Note that the G33 / G76 cycles are not defined in the original RS274/NGC definition, so have been implemented subsequently in LinuxCNC. linuxcnc.org/docs/2.6/html/gcode/rs274ngc.html .

Interestingly, the LATHE-G76.NGC example file provided within the AXIS sim examples doesn't specify either G94 or G95. I guess feed per rev is perhaps implicit in the command itself.

On the matter of G33 vs G76, I can't seem to force Fusion to generate a G76 threading operation, no matter which options or tick boxes I change within Fusion. I wondered if I was getting G33 due to having selected variable depth of cut / engagement but that doesn't change the result (LinuxCNC calls this "depth degression"). Furthermore, if I select the "use cycle" tick box, Fusion generates the toolpath but the post processor subsequently fails to process the resulting intermediate file. I don't understand how you are generating G76 if you are using the linuxcnc-lathe post or the Path Pilot version of the same.

Bear in mind I'm far from expert in lathe operations, as I am new to it too. Until recently my experience has been limited to milling.