Axis substutution issues.

[HayreAss]

Hey All,

I am having some issues here with axis substitution.
What I see, is NOT what I get...
I'm sure it's my error somehow, but I'd like to learn how to resolve it, as a LOT of my work is exactly this type of work.
Machining patterns around a cylinder...
(in this particular instance, a conical shape, not a true cylinder)
Work/Tool/Construction planes are all Top.
When I simulate the cut, all is well, the part looks just like I want it to.
However, when I actually cut it, it's reversed...

Any Ideas where I am going wrong here?
I'm sure it's simple PEBKAC

I've been successfully making parts for years, but it quite often involves making a scrap part, adjusting the drawing so that reality ends up where I want it, and accepting that simulation does not look like reality.
I'd like to resolve that.
I'm sick of scrap...

[Kyle F]

I'm definitely interested in this,.. as not too long ago I programmed a 4th axis part that did a bit of simultaneous work (usually it's just 3+1 workflow on our 4th axis) and I had the exact same issue. my rotational axis would spin the wrong way. this was on a haas vf-2 with hrt210 (IIRC) It was just on one toolpath so we ended up just going in and fixing it by hand but it's been on the back of my mind ever since. Anyways best of luck and I'll be looking forward to figuring it out as well.

[Greg Williams]

Good chance it is a post problem, simultaneous and axis subs are handled differently in the stock posts. You may have to add some reversal code into - pxyzcout1

[HayreAss]

2 minutes ago, Greg Williams said:

Good chance it is a post problem, simultaneous and axis subs are handled differently in the stock posts

 

OK, I have no problem believing that.
Where should I look for that?
I see axissub_dir and rot_ccw_pos in the post settings, both are 0
But there is nothing telling me what the options are for axissub_dir

Also, other details on the parts are placed in the correct location.
Details  indexed to angles are correct.
Other details, slots, etc... that are also cut with axis sub are in the correct location...

[#Rekd™]

If you post a ziptogo file it will be easier to see what is going on. 
 

 

[HayreAss]

Ok, I had to get back to the shop and edit the drawings a bit to slow down the counterfeiters.
No need making it easy for them!
Here we go though, 2 parts.
The 'Sample' zip file is the one that mills backwards.
The 'Sample 2' file machines correctly, as shown in the attached pic.
 

sample 2.MCAM-CONTENT sample.MCAM-CONTENT

[#Rekd™]

I looked at the files quickly and found something that could be causing the issue.

See this video.

https://www.dropbox.com/scl/fi/bhwm3gcdgsisih2cl1870/Machine-Definition_JF.mp4?rlkey=b74e7bzatki7p1u3gjy7s9qdn&dl=0

One other thing to watch on Haas machines is there is a setting to "mirror" the axis, not sure why you would want to mirror an axis (just program it correctly).

[HayreAss]

Really weird now.
When I look at them here, they are the same, as I would expect them to be.

Same machine, on the same system!

 

[#Rekd™]

Look at the local settings within the files and look at the global settings.

Also study the code to determine if the direction in correct after posting to compare what is correct vs the incorrect.

Outside of that having verification software would check the actual posted G-Code.

[HayreAss]

I wasn't even aware that there was local AND global settings.
I thought all Settings were Global.

[HayreAss]

Even stranger, when I import the zip2go that I posted, I do see the CW vs CCW like you show.
It'll be a week or two before I can test on the machine, so we'll see.