Filtering a 5-axis Flowline Toolpath

[Justin Beebe at Folsom Tool]

We are using a 5-axis flowline toolpath and it seems to be generating a lot of extra code. I was wondering if there is a way to filter it. It seems like it's spitting out a lot of unneccesary A-axis moves. Here is a sample of the code.

 

N44150 G94 X-13.3292 F60.

N44160 G93 A-61.2 F1000.

N44170 A-59.357 F1000.

N44180 A-57.48 F1000.

N44190 A-55.568 F1000.

N44200 A-53.622 F1000.

N44210 A-51.642 F1000.

N44220 A-49.628 F1000.

N44230 A-47.582 F1000.

N44240 A-45.502 F1000.

N44250 A-43.39 F1000.

N44260 A-41.247 F1000.

N44270 A-39.073 F1000.

N44280 A-36.87 F1000.

N44290 A-34.638 F1000.

N44300 A-32.378 F1000.

N44310 A-30.093 F1000.

N44320 A-27.782 F1000.

N44330 A-25.449 F1000.

N44340 A-23.093 F1000.

N44350 A-20.718 F1000.

N44360 A-18.325 F1000.

N44370 A-15.915 F1000.

N44380 A-13.491 F1000.

N44390 A-11.055 F1000.

N44400 A-8.609 F1000.

N44410 A-6.155 F1000.

N44420 A-3.695 F1000.

N44430 A-1.232 F1000.

N44440 A-358.768 F1000.

N44450 A-356.305 F1000.

N44460 A-354.279 F1000.

N44470 G94 X-13.3173 F60.

 

I'm wondering if there is a way to make all the A-axis moves just one A-axis move.

Using MCX4 with a custom Fadal Post from In House Solutions.

 

Thanks in advance.

[Oscar R.]

There's a couple of ways to do this. If your willing to sacrifice part quality try opening up your filter tolorance. If reducing part quality is not an option then try using your Tool Axis Control "Limits". Play with the min.-max. angle setting that will work best for your part. Sounds like you may have some flat features on your part that don't really need continuous full 5 axis. Using Tool Axis Control Limits will only allow your rotating axis to the limits the you specify and then the linear axis will kick in to compensate.

 

Hope this helps.

[Dtm]

We are using a 5-axis flowline toolpath and it seems to be generating a lot of extra code. I was wondering if there is a way to filter it. It seems like it's spitting out a lot of unneccesary A-axis moves. Here is a sample of the code.

 

N44150 G94 X-13.3292 F60.

N44160 G93 A-61.2 F1000.

N44170 A-59.357 F1000.

N44180 A-57.48 F1000.

N44190 A-55.568 F1000.

N44200 A-53.622 F1000.

N44210 A-51.642 F1000.

N44220 A-49.628 F1000.

N44230 A-47.582 F1000.

N44240 A-45.502 F1000.

N44250 A-43.39 F1000.

N44260 A-41.247 F1000.

N44270 A-39.073 F1000.

N44280 A-36.87 F1000.

N44290 A-34.638 F1000.

N44300 A-32.378 F1000.

N44310 A-30.093 F1000.

N44320 A-27.782 F1000.

N44330 A-25.449 F1000.

N44340 A-23.093 F1000.

N44350 A-20.718 F1000.

N44360 A-18.325 F1000.

N44370 A-15.915 F1000.

N44380 A-13.491 F1000.

N44390 A-11.055 F1000.

N44400 A-8.609 F1000.

N44410 A-6.155 F1000.

N44420 A-3.695 F1000.

N44430 A-1.232 F1000.

N44440 A-358.768 F1000.

N44450 A-356.305 F1000.

N44460 A-354.279 F1000.

N44470 G94 X-13.3173 F60.

 

I'm wondering if there is a way to make all the A-axis moves just one A-axis move.

Using MCX4 with a custom Fadal Post from In House Solutions.

 

Thanks in advance.

 

 

 

All of the 5-axis work I have done has alot of code as well and I never could filter my program. I was using a custom post from in house as well, and was also using inverse time. If you figure out how to do it without a post mod. please let me know.

[Colin Gilchrist]

Have you tried using the Filter control in the Toolpath itself? You can use it to filter out position moves before the post processing occurs.

[Greg Williams]

I have found that the more code I have the better the result

[Guest]

^^^^^^^^^^ +1

 

I agree

[Aaron Eberhard]

I'm assuming this is just a positioning thing and that's why you want to filter it? Either way, there's probably a filter set in the post to break down rotary moves to every 2* or so. Some controls (mostly older, now) will have some problems with rotary axis control if they're not given small enough vectors during a 5 axis cycle, you'll get crazy accel/decels on the motion, so the solution to feed them a vector for every X *. Without knowing your post at all, I can't help you find it, but you might want to see if you can search for max angle or something like that inside of it (if it's not encrypted).

 

Any reason you're not using inverse time? Your motion will be a LOT better if your control will support it.

[Justin Beebe at Folsom Tool]

Colin,

 

We tried to switch the filter tolerance from .001 to .010 and it makes the toolpath look less accurate in in backplot. We didn't want to try to run it on the machine because of the way it looked in backplot.

 

Aaron

 

We are using inverse time notice the G93. Also the post is encrypted so we will have to send it back to In House if we cant find the setting in mastercam.

 

All this toolpath is doing is milling a round diameter by rotating the A-axis while avoidng some bosses that stick up on the part. Instead of making one A-axis move to index we get thirty lines of code for the A-axis to end up at the end position for that rotation. Since the A-axis is the only axis moving it seems like there is a lot of extra code. It seems to me that we could eliminate the A-axis moves from N44180 to N44450 and still get the same result. I'm not sure if this is the case because this is our first attempt to use this toolpath.

[Aaron Eberhard]

 

We are using inverse time notice the G93.

 

Doh! I missed that.. Too focused on the angle steps.

 

Also the post is encrypted so we will have to send it back to In House if we cant find the setting in mastercam.

 

I've been playing around with it a bit, and my post does the same thing. It's based on the MPGEN5X_FANUC post, so I'm suspecting yours is as well. The post code I was thinking of was the "brk_max_ang" option, but I dont' have that set. Without it, it will output every 2.x*, just like yours. I can only get it to output smaller rotary vectors, not larger.. Hmmm...

 

What's odd, is that the angle between vectors of the toolpath I made (with a .001" tolerance) are ~.02* apart, where as the output is ~2* apart. My guess, and I'm completely talking out me arse, is that it has to do with rotary calculation code taking place in the encrypted part of the post..

[Justin Beebe at Folsom Tool]

Aaron,

 

Thanks for trying to help I really appreciate it. I think I'm going to call our reseller and have them get in contact with In House Solutions. They are the ones who made our post. I'll let you know what I find out. The good news is we got the entire program to fit in the machine. The bad news is it took about an hour and twenty minutes to load.