Opti-rough with ballnose endmill causes collisions

[Maclaw]

Hi All,

Has anyone encounter problems with ballnose endmills in optirough (rest) toolpaths. What I've recently noticed is that this toolpath, when used with a ballnose endmill, may occasionally cause collisions with in-progress stock. This can be seen in verify - red color or in the report tab. The collisions regard the flute (cutting) portion of the tool. I attach an MC file with a dummy project to look at this. The EXACT same toolpath (copied) is used with a bullnose and the collisions disappear. If anyone has any thoughts about this - please let me know. The file is metric.

Thanks in advance!

Maclaw.

OPTI WITH BALLNOSE.mcam

[gcode]

After analyzing the reported collisions, I realized that the toolpath was fast feeding through tiny cusps on horizontal surfaces.

I left all the settings in your file unchanged and enabled Microlift with a setting of .5mm 

Now it fast feeds over the tiny cusps and the entire toolpath runs without reporting any collisions.

[Maclaw]

Thanks gcode. Tried it and it doesn't collide. You could say case closed, but...

Anyway - from my perspective - that's a workaround. From my point of view - the toolpath shouldn't do that. Opti is a very powerful and almost automatic roughing startegy in conjuction with stock model. But the most important thing for a programmer is it's reliability. You must be able to trust it! I know, I know - you should always verify and simulate it, but if you get a problem - it could take you more to make it work than sometimes roughing the part "manually" using other paths. So my theory is to rather use bull-nose due to the flat portion that it has - it doesn't generate scallop. Seems opti isn't so comfortable with this scallop generated by a ballnose and may collide. I did this in aluminium - that is a very forgiving material. If you were to mill an exotic... - it definetely would be a problem. I also have a question to the Developers of MC : Is Opti-rough more dedicated to bullnose than ballnose? Does the above "theory" make sense. If yes - than I will do that - will avoid ballnose and the case is closed for now - hopefully it will be more bug-free in the future. What do You think? Besides if You play around with the step-over/step-down/step-up in the file in the toolpath that uses the ball - you could get even worse collision scenarios. Can the Developers make their opinion on that - that would be very useful and life-saving info for all of us. 

Thanks very much and looking forward to your opinions.

[crazy^millman]

6 hours ago, Maclaw said:

Thanks gcode. Tried it and it doesn't collide. You could say case closed, but...

Anyway - from my perspective - that's a workaround. From my point of view - the toolpath shouldn't do that. Opti is a very powerful and almost automatic roughing startegy in conjuction with stock model. But the most important thing for a programmer is it's reliability. You must be able to trust it! I know, I know - you should always verify and simulate it, but if you get a problem - it could take you more to make it work than sometimes roughing the part "manually" using other paths. So my theory is to rather use bull-nose due to the flat portion that it has - it doesn't generate scallop. Seems opti isn't so comfortable with this scallop generated by a ballnose and may collide. I did this in aluminium - that is a very forgiving material. If you were to mill an exotic... - it definetely would be a problem. I also have a question to the Developers of MC : Is Opti-rough more dedicated to bullnose than ballnose? Does the above "theory" make sense. If yes - than I will do that - will avoid ballnose and the case is closed for now - hopefully it will be more bug-free in the future. What do You think? Besides if You play around with the step-over/step-down/step-up in the file in the toolpath that uses the ball - you could get even worse collision scenarios. Can the Developers make their opinion on that - that would be very useful and life-saving info for all of us. 

Thanks very much and looking forward to your opinions.

You answered you own question and supported the answer by the way you asked it. The very nature of a ball endmill is to machine curved and varying surfaces is it not? Would you machine a 20 meter by 20 meter flat plate with a ball endmill or a flat endmill, Face Mill? You have chosen to use ball endmill and didn't use it correctly with the right amount of micro lift not taking the cusp into account like you would need to do with the example above and the software needs to now be smarter than the user? You were provided the solution to the problem you gave the software that is the correct method and process to get the results you need. On the 20 meter by 20 meter example if you took 20m step overs with a 10mm ball endmill 3 mill deep. What kind of cusp would be left? Now Face toolpath need to think about the cusp a programmer left because someone decided to use a ball endmill the wrong way and left all the cusp? Same with contour I decided to cut a flat face with a large multi pass step over leaving cusp because instead of using the correct tool i used the wrong one, but the software now need to understand the design intent and other things to hold a programmers hand who is lacking in the ability to pick the correct method for the work at hand?

A good craftsman never blames his tools for his quality of work. The software is a tool for us to use and if we chose to use it different than it is intended to be used then we have to adjust our way of thinking to use it that way. I will use Opti-Rough with Ball endmills from time to time due to many reason. I know that with that I have to make certain adjustments to my normal way of programming. One of them the most important is taking cusp height in relation to step over into account. In this case it was not and the craftsman is blaming the tool he is using for not getting the results he wanted verse stopping humbling himself and saying I need to adjust my thinking and use my tool in a different way to get the results I am after.

Now before you and many others go getting all butt hurt you did ask for opinions and it is just that my opinion. I have my fair share of issues I want addressed in the software, but this one has the solution to the problem answered and makes sense if someone stops long enough to humble themselves to understand it.

[Maclaw]

Crazy^millman, wow! Thanks for your help and good advice but unfortunately it did not go in the right isle...

I think you misunderstood my question and my problem. That's propably my fault - English is not my native language But to be honest - I don't really understand why you are talking about metetrs of flat areas milled with a ballnose leaving cusps etc. Have You seen the part I attached? If not, be informed that the part is a sphere with 4 vertical technological bosses attached to it which are filleted to the sphere with R10 mm (394 thou radius). From my 15+ year experience, a ball cutter was the first choice to clean up the fillet radii from material left by the 25mm bullnose that I roughed the thing out with... "Beat me up Scotty" - but unfortunately that was my first choice! Sorry it wasn't the face mill nor the flat or bullnose... I admit - the microlift was my overlook which gcode fixed quickly - thanks again gcode

Remember when I said "...if You play around with the step-over/step-down/step-up in the file in the toolpath that uses the ball - you could get even worse collision scenarios..."? Well here is one :

The same toolpath that gcode fixed - the change is stepover to 45% (3.6mm), stepdown to 6% (0.48mm), step-up not used, because stepdown is small enough (I wanted to use a constant-z like approach with additional side steps if needed). And guess what - you get collisions! This time the microlift is 1mm and the maximum cusp height that is derived from the stepover is:

 

c = R - sqrt[ R^2 - (s^2 / 4) ] = 0,428mm   (the microlift in the path = 1.0mm - so it should be safe)

where:

c = max cusp height (provided you mill off a layer >= the cusp height)

R = radius of ballmill (4 mm - it's a 8mm ball)

s = stepover (3.6mm)

 

So why are there collisions? Can you explain? Bad craftsmanship? Bad program to do the job? Bad toolpath/tool usage? Bad day? Bad week? Bad what?

Well - I don't know the answer - that's why I asked people that are smarter than me - maybe they will come up with something, maybe they had issues like that in the past... I think that this question is also adequate to people who created the software - they know best their intension with this toolpath. If they say - try avoiding or take special care to this opti toolpath with ball endmills - that's fair enough - that's some info at least. But in the end - I think this should be fixed - it might be in 2025, 2028 whatever, but something is said about it and some secrets revealed and some foundation built for the future of this magnificent - after all - software! Step by step - that's how it looks in my world...

I attached a file - it's operation no. 5 that generates these collisions. Operation #6 is for comparison, that a bullnose does not have these problems. It has the exact same parameters - only the tool is changed.

Crazy^millman I really respect all your knowledge, experience and professional expertise - really. You rule on this forum! But I don't like misunderstandings, hence all the above. Also don't worry about me getting "all butt hurt" - I don't have time for such soap opera BS - I'm here to get rid of a problem - pure and simple.

Anyway, my last thought is that this problem DOES need answers and a remedy.

All the best to You all!

OPTI WITH BALLNOSE.mcam

[crazy^millman]

Your English is better than mine so no worries there.

45% step over with an flat lands using a ball endmill? Sorry 2024 is not installed on this new hard drive so I cannot take a look at your file. 1mm should be safe depending on the angle of micro lift. We don't have the ability to change that angle so depending on the complexity of the shape near where you are cutting the micro life might be trapping in such a way it cannot get up and out of the way enough. By changing the process and sing a bull endmill you are creating a safe process and method to allow the elimination of cusps. I offered my opinion and it can be worthless or it can be worth something.

Good job on the math and process to determine the cusp height, but there is a lot more going on in any part that just a cusp height and each one presents its own challenges. I would only use a ball endmill in a shape where I had no other choice for Optio-Rough. I had just a scenario the other day. I needed to get in a deep v section with a small radius where I have about 15XD if i try going normal to the floor or wall with a tool. I decided to go with a 4XD ball endmill in a 5 Axis operation using Unified and Guide along with a stock model. I flipped on the dynamic motion and in essence created a 5 Axis opti-rough toolpath. I kept my step over reasonable and got no unexpected cusps because I went with 20% step over. There was no flat anywhere so I went as deep as I could with a bull endmill. i then came back with the ball. I got step downs and had the step over been to great then i would have got some nasty cusps. I made the decision to reduce the step over to not have that as an issue.

Quote

I attached a file - it's operation no. 5 that generates these collisions. Operation #6 is for comparison, that a bullnose does not have these problems. It has the exact same parameters - only the tool is changed.

Again where I was just trying to point out that this might be the best solution since it is giving you trouble. I am in a different situation where 99% of time I am not there when my program is run. I have to provide error and trouble free files in any material on any type of machine. In my experience using the bull or changing the step over provides that method. Do I want to do more aggressive things more often; OH YEAH, but then people get scared or unsure and then I may not get paid for work because of that.

Sorry about the comment coming across the wrong way. I have been attacked so much over the years giving an opinion just wanted to make sure it was understood as such. We are all grown ups and can like or dislike an opinion. It doesn't mean that person is meaning will ill because I am never meaning that it was just my thoughts.

All the best to you also!!!!

[Maclaw]

I went with 45% sideways because of a small stepdown. Did not want to waste that much time. A ball can handle 45% sideways while 6% deep in aluminium no problem. I would also need to trim the path with some "dummy" stockmodel to make it stay off the sphere that was previously roughed with the big tool, but that's a different thing.

I know you didn't mean ill will. Me neither. I try to focus on the problem and then tackle it. I rely on Mastercam everyday, even though I do not consider myself a power user. But if something goes wrong, I do worry and - if time provides - try to dig upon the subject more deeply to solve it and explain what was the issue. Since time is my enemy - a lot of times I just go around the problem and at the end of the day come out the same door only from a different corridor... And that I love in MC because it gives you the freedom to do so.

I will try to test this "ballnose opti thing" deeper and will let you know what my conclusions are if they will be worth it.  Hopefully, in the meantime will also read what others say.

Thanks and take care!!

[Maclaw]

I tested the thing and have one conclusion when using a ballnose endmill in opti-rough:

Use it with deep cuts and shallow stepovers - you should be safe - at leas from what I've tested a few times. If you try to use it with shallow DOC and big stepovers, you might get trouble with it colliding. Maybe not always - but be ready for it

What I tried to do before was using opti-rough with a ballnose with shallow DOC and big sidestep (I had a 25mm ballnose insert cutter - not solid endmill) - similar to a "constant-z" like approach but with opti's great stock awareness and the possibility to add additional stepovers if needed to take out excessive "lands" of material. I considered that it shouldn't be a problem, but it seems I was wrong.  This approach works out great though with bullnose endmills and what's more important - with insert cutters whose nature of milling is shallow DOC and big stepover. I use opti very often to take out material with insert cutters/high feed cutters etc. The approach is to go down in Z by the "millimeter" (+-) and take huge, but consistent sidesteps. No up-step usage because DOC is small. But with solid tools the thing is opposite - big DOC and small sidestep + upsteps - to fine rough out of the angles on surfaces. Anyway - beware with ballnose with this approach.

I hope this helps.

Take care and all the best!

-Maciek.

 

 

[jpatry]

On 9/4/2023 at 3:39 PM, crazy^millman said:

Do I want to do more aggressive things more often; OH YEAH, but then people get scared or unsure

I even get that when standing at the machine when the guy is running the new program.