Error: Failed to post data

[jasonswolfe]

Hi - 

I'm hoping this is simple...  I'm pretty new to Fusion, but I've been using it for a few months and have posted tons of g-code without any issues.  This weekend I was putting together a batch of parts, and of the maybe dozen toolpaths, I had two that failed to post-process correctly.  I didn't even realize until halfway through cutting one of the parts when my router just pulled out of the work and stopped.

Quick background:

I'm using a very simple 3 axis router and running it with Mach3.

The parts are dead simple 2d contours - pretty much just just a couple of circles.  I exported one of the offending files - hopefully that works.

The CAM setup, toolpath setup, post processor, etc. are the same as I've been using on lots of successful parts.

The post processor is one that I downloaded - I think it is just a lightly modified version of the standard Mach3 post processor to simplify it a little bit.  I just tested out a few post processors, and this one made my machine do less annoying things than the other post processors that I tried...

Here is the log that comes up when I try to post process the parts:

Information: Configuration: Generic Mach3Mill
Information: Vendor: Autodesk, Inc.
Information: Posting intermediate data to 'C:\Users\Jason\Dropbox\CNC\16.0604 - Fixed Dust Shoe\Clamp 1_4bit_4.tap'
Information: Total number of warnings: 1
Error: Failed to post process. See below for details.
...
Code page changed to '1252 (ANSI - Latin I)'
Start time: Tuesday, June 7, 2016 7:37:17 PM
Code page changed to '20127 (US-ASCII)'
Post processor engine: 4.2.1 40504
Configuration path: C:\Users\Jason\Dropbox\CNC\0000 Setup\Post Processors\mach3mill_Chris.cps
Include paths: C:\Users\Jason\Dropbox\CNC\0000 Setup\Post Processors
Configuration modification date: Saturday, January 30, 2016 6:33:18 PM
Output path: C:\Users\Jason\Dropbox\CNC\16.0604 - Fixed Dust Shoe\Clamp 1_4bit_4.tap
Checksum of intermediate NC data: ee16a7c49d14d759892bb532e69dd982
Checksum of configuration: 17ddb58810cb48e1b38fd3aed8c5efa1
Vendor url: http://www.autodesk.com
Legal: Copyright (C) 2012-2013 by Autodesk, Inc.
Post processor signature could not be verified (error 0xfffffffc).
Generated by: Fusion 360 CAM 2.0.2087
...
Warning: Work offset has not been specified. Using G54 as WCS.
Error: Invalid toolpath point detected '(-1.#IND, -1.#IND, -1.#INF)' for 'onLinear'.
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
Error: Failed to execute configuration.
Stop time: Tuesday, June 7, 2016 7:37:18 PM
Post processing failed.

Any help would be greatly appreciated!  I've been hunting around for similar errors, but haven't been able to find this.  I'm guessing it's something simple that isn't set right, but I'm not finding where the issue is, and it's more disconcerting because all of my parts have worked fine up until now.

Thanks!

Jason

[LibertyMachine]

It's because you have "Multiple Depths" set as well as a ramping strategy specified. The problem comes in when the Max stepdown of the "Multiple Depths" is equal to or less than the maximum ramping stepdown. Tweaking it around, I found that setting the multiple depth to .150 and the ramp stepdown to .125 cured it. FWIW, you could have .126 and .125 respectively, and it would work.

I am curious though, why you are specifying Multiple Depth if your strategy is to slowly ramp down the contour of the part....To each his own I suppose

[jasonswolfe]

Gearsoup - 

Thanks so much for taking a look at this.  Indeed when I set the multiple passes to .150 and the ramping to .125 did allow the code to post properly.  The strange thing is that it seems to be the .125 on the roughing that's causing the issue irrsespective of the ramping.  Setting the ramping to .124 doesn't solve the issue, and in fact turning the ramping off entirely doesn't solve the issue.  If I set the roughing to anything but .125 it seems to work (tried .150, .124, and .126).  You definitely pointed me in the right direction, but I still don't entirely understand what the issue is - not a problem to use a slightly different value, but strange that .125 seems to be messing it up.

On the ramping itself - I think you're totally right that my setup is weird.  It's not really what I want it to do, but I haven't quite figured out how to make Fusion do what I want.  What I'm after is a ramp on the Z plunge so that if I'm using tooling that doesn't like to plunge cut that it will ease itself into the piece.  Usually when I've done this in other software it's a little zigzag as it eases the tool in rather than a giant spiral around the piece.  I don't neccesarily hate the giant spiral, but it does look like I'm duplicating some cutting.

Anyway - thanks again for the help!  At least I can get functional code out now, even if the root cause still doesn't quite make sense.

Thanks,

Jason

[jasonswolfe]

I'm still having this problem...  I ruined a couple of parts last night by not checking that they had posted correctly.  Stupid me...  but I still would love to understand what's going on.  I thought that it was just the .125" setting in the multiple depths that was causing the issue, but I changed all of my parts to .126" last night after they failed the first time, and the second time around 2 out of 3 posted fine, but the 3rd one failed like before.

It's extremely frustrating because I'll get half the code, so it looks fine when I pull it up in Mach 3, but then it gets halfway through and then suddenly stops.  Worse because it shuts off the spindle and drives itself to home cutting a huge nasty gash on the way...

The parts were extremely simple (just circles) - it's the 2d contouring that's causing all of my problems.  It seems like it should be the simplest of all, but for some reason it's been the bane of my existence.

Let me know if there's anything I can do to help troubleshoot.  Thanks,

Jason

[LibertyMachine]

Export and share the part if you don't mind, I'd like to dig into it. I suspect it's a setting you aren't seeing, or have set incorrectly. Still using the stock Mach3 post>

[jasonswolfe]

Gearsoup - 

I'm sure you're right.  I've included the current file - it works as is, but the only change I made to get it to post correctly was to change the depth in "multiple depths" to a number different than .125".  To recreate the error, change the cutting depth in setup2 from .127" to .125" - at least on my computer it causes a complete failure to post.  This is actually better than what I've been seeing - I've mostly been getting 1/2 of the code correctly, and then it abruptly ends with "failed to post data."

I'm using a modified version of the Mach3 post processor - it's called mach3mill_Chris.cps.  I downloaded it from the forum, and have been using it as it makes my machine act more or less like I expect.  I believe the stock mach3 post processor drove the machine back to home at the end which was causing huge issues for me (my home is at a negative location relative to 0,0 for the parts, and it was crashing into guards on the way to home).

Thanks!

Jason

[LibertyMachine · Accepted solution]

Why it's failing to post, I'm not sure. I think the..."flaw"...in your process is simply having the combination of "Multiple Depths" roughing passes AND the Ramp style entrance selected. If you are going to use Ramp, you don't need to specify Multiple Depths at all. Simply give it a large Ramp Angle and then specify Max Ramp Stepdown to what you would want to see (I chose .125" for instance)

It posted out just fine for me with that choice. Move away from choosing both options. Multiple depths are fine to select if you are dropping into a pre-drilled hole and then going from there in a simple 2 axis contour operation.

If you are going to ramp down the entire contour, just turn off multiple depths. Just make sure your Heights tabs are set proper

[jasonswolfe]

OOOOOOOOH....  I totally didn't understand or notice that it has the maximum ramp stepdown option in there.  And that's what you were telling me the first time...  Sorry, coming from CAM software that only had ramping as a lead-in type strategy.  This is much better - it kills two birds with one stone by both ramping and only stepping down my preferred distance.

I still stand by my being able to make Fusion post an error, but only because I was doing something that probably doesn't make much sense to begin with... 

Thanks again for looking at this - and sorry that I didn't get it the first time around.  I thought I had things figured out with my "anything but .125" setting, but this fixes the problem and makes my code much better (at least less bad)!

Thanks!

Jason