topic Machine crash not shown in machine simulation in Fusion Manufacture Forum

Machine crash not shown in machine simulation

alxtourv — Wed, 13 Sep 2023 18:28:14 GMT

Hey there, I've recently been working on a part using a 5 axis simultaneous pencil tool path to mill slots wrapped around a curved surface with a ball endmill. Essentially a giant knob with 24 ball end mill slots from top to bottom. My machine is a UMC 750. I am using the haas next Gen control post. I created one tool path and used a pattern to make the rest of the slots. In the machine simulation it shows all the retract moves being clear of the part. Then when I went go to run the program it did the first 12 slots perfectly fine. On the 13th slot it retracted and then not dived straight into the front of the part. I double checked the simulation and low and behold it is not shown. Does anyone have a solution to this or know of this issue? Its honestly terrifying to use any simultaneous toolpaths at this point knowing that the crash won't show on the simulation.

Re: Machine crash not shown in machine simulation

seth.madore — Wed, 13 Sep 2023 18:52:56 GMT

Oh that doesn't sound good?
Would you be able to share your Fusion file here or privately?
File > Export > Save to local folder, return to thread and attach the .f3d/.f3z file in your reply. To share the file privately, it's the same process, but email it to me: seth DOT madore AT autodesk DOT com

Re: Machine crash not shown in machine simulation

alxtourv — Wed, 13 Sep 2023 21:13:22 GMT

Hey Seth, I'll attach the file here. its the setup Labeled "winder-knob".

The machine crash happened while ball milling the 13th slot.

Thanks!

Re: Machine crash not shown in machine simulation

alxtourv — Wed, 13 Sep 2023 21:15:00 GMT

I noticed that the Machine crash happened between the end of the roughing pass into the first finish pass. the tool retracted fine, then came down and plunged diagonally into the front of the part. Almost like it was trying to start at the front of the part rather than back.

Re: Machine crash not shown in machine simulation

seth.madore — Thu, 14 Sep 2023 06:47:46 GMT

The issue is that Fusion currently lacks the ability to simulate connections between toolpaths, and it's a weakness we're well aware of and working on improving. Let me check with the post team to see if there's some improvement we can make to the post to provide safer positioning..

Re: Machine crash not shown in machine simulation

seth.madore — Thu, 14 Sep 2023 07:49:22 GMT

Could you share the post processor that you're using, as I see it's one that you've edited. It's possible that it's out of date on some functionality, so I'd like to compare output code between your post and a current Haas post

Re: Machine crash not shown in machine simulation

alxtourv — Thu, 14 Sep 2023 11:14:45 GMT

So I was actually just using the latest next Gen control post. I was trying to combat the problem by using a modified post I had after the crash occurred.

Re: Machine crash not shown in machine simulation

alxtourv — Thu, 14 Sep 2023 11:47:19 GMT

Honestly this is pretty disappointing considering I pay for all of the extension ad-ons. I tried manipulating the toolpath by splitting it into two patterns each from 0-180 degrees. Now it wants to crash through the part on the 3rd or 4th slot. This really shouldn't be something that's being marketed to customers if it's going to cause costly crashes. The holder smashed right into my part. Ruining my part, tool, and throwing the head of the machine out of alignment. Honestly super disappointed with fusion 360 at this point. It's strange considering powermill seems to have some pretty good 5 axis capabilities.

Re: Machine crash not shown in machine simulation

programming2C78B — Thu, 14 Sep 2023 11:59:53 GMT

To be fair, all new programs should generally be ran at 5% rapid, distance to go, single block upon first run until proven out. What is the code that's causing the "crash" issue? Is your machine new enough to have the safety run feature?

I can understand if the codes became different along the pattern, but seeing how close your retracts are leads me to believe that it WOULD crash, unless keeping the safety G53 z0 moves

Re: Machine crash not shown in machine simulation

alxtourv — Thu, 14 Sep 2023 12:02:56 GMT

To be fair, it was a simple pattern toolpath that had gotten done 12 perfectly fine slots. You don't expect it to nose dive randomly 13 slots in. And even if it was simulation in machine graphics and caught there it still doesn't provide a fix to the issue. It just shows that 'yes this will crash'. But if you can't generate code that won't crash then where does that get you? Clearly this is a huge post processing error that I'm willing to try to help recreate my issue to get it solved. I've caught the software doing similar things with 5 axis linking moved in the past as well.

Re: Machine crash not shown in machine simulation

alxtourv — Thu, 14 Sep 2023 12:06:17 GMT

The leads had to be eliminated to not gouge the part. There was no issue with my leads. The machine retracted to G53 Z0, then home, then came over and smashed into the front of the part. If you simulated the toolpath you would clearly see that there was no issue with the leads or links defined that would have caused the crash.....

Re: Machine crash not shown in machine simulation

alxtourv — Thu, 14 Sep 2023 12:08:54 GMT

The file is right there, if you put any time into it like Seth did, you would realize that technically speaking there is nothing that I defined in the toolpaths that would cause a crash. Don't act like I don't know how to single block a machine.... don't throw blame around. This is clearly a major post processing issue that needs to be addressed.

Re: Machine crash not shown in machine simulation

seth.madore — Thu, 14 Sep 2023 12:10:09 GMT

@alxtourv wrote:

This is clearly a major post processing issue that needs to be addressed.

Agreed. I don't have an answer for you yet, but I've raised this thread with our Post team to see if there's something that needs to get done

Re: Machine crash not shown in machine simulation

alxtourv — Thu, 14 Sep 2023 12:26:03 GMT

Thanks, I appreciate that.

If it helps at all the crash that occured was sending the tool to somewhere around the coordinates of X-.2 Y2.1 Z-4.6 while it was tilted at 89 degrees ready to finish the slot.

Definitely rough coordinated haha but right around there.

Re: Machine crash not shown in machine simulation

seth.madore — Thu, 14 Sep 2023 12:49:19 GMT

Do you have an approximate line number where this occurred?

Re: Machine crash not shown in machine simulation

alxtourv — Thu, 14 Sep 2023 21:04:36 GMT

I believe the crash occurred around line #160847.

I posted a picture for reference.

Re: Machine crash not shown in machine simulation

nick_p3dcreations — Sat, 16 Sep 2023 14:22:08 GMT

Can you grab your HAAS software version also?
In the past I have had MRZP issues giving me similar issues, which were ultimately solved for good with a combination of machine software updates and fusion post updates.

Re: Machine crash not shown in machine simulation

AchimN — Thu, 21 Sep 2023 12:36:19 GMT

Hello @alxtourv ,

sorry for the delay, I've looked into your program as well as into the postprocessor but I do not see anything wrong with the either the post or the NC program. I´ve also compared the output with other posts which use similar methods for positioning and the coordinates and methods are exactly the same.

When you say that the machine tried to move the opposite side in operation Pencil1 (7) in its 13th occurrence, that position would then basically match with the first occurrence of the that operation right?
When I then look into the nc program, I do see this for the first time of the Pencil1 (7) operation:

(Pencil1 7) // 1st occurrence N2205 G53 G0 X-29. Y-8. N2210 G0 B84.391 C-90.001 N2215 G254 N2220 G1 X4.9162 Y-0.0005 F650. N2225 G255 N2230 G0 G234 H8 N2235 X-0.0005 Y-3.4559 Z-4.6004 N2240 G1 Y-2.139 F650.

and this is the occurrence which you highlighted:

(Pencil1 7) // 13th occurrence N3525 G53 G0 X-29. Y-8. N3530 G0 B84.391 C89.999 N3535 G254 N3540 G1 X4.9162 Y-0.0005 F650. N3545 G255 N3550 G0 G234 H8 N3555 X0.0005 Y3.4559 Z-4.6004 N3560 G1 Y2.139 F650.

The difference is mainly that the C-axis is rotated 180deg. The coordinates of the initial position using G254 are the same between both operations which is correct, and when you look at N2235 and N3555, these coordinates are in TCP and XY are inverted due to the C-axis rotation which also looks correct to me.

Since you say that your machine moved diagonally when it did execute N3555, I think the main issue is that it didnt get to the desired position in N3540 when using DWO since there should only be Z movement between N3540 and N3555 which was obviously not the case on your side.

I can currently only assume at this point that the reason for the issue is on the controller side, potentially an issue with the DWO feature G254.
Would you mind getting in contact with Haas support directly and ask them if they have an idea whats wrong? As @nick_p3dcreations mentioned, it could be related to the software version on your machine and it would be quickest when you connect with Haas directly. I will monitor this topic, please just let us know when you found that we must do something on the postprocessor side.

Thank you,
Achim