FINISHING 2D PARTS AFTER OPTIROUGH in MC-X7

[Maclaw]

Hello fellow CNC Mates,

 

I know there are a lot of efficient roughing technologies and HS tollpaths in MC. They are very cool (opti area, core, rest, using stock model etc.). I also use them and are happy with it :-) . My big question would be - what about finishing 2D fully prismatic parts after they have been roughed and rest machined with the use of the above techniques? I am not talking about 3D surfaces - only 2D relatively complex parts (a lot of pockets, bosses, varying depths and other "bright design surprises", sometimes on the edge of artistic ideas like digging with a flat endmill 12D's deep... :-))))))

 

I have attached a sample file - it's just something i modeled in CAD to show You (could not give any real-life examples - no permission to share them from my Customers...).

 

Lets say I rough it and rest-cut it with OptiRest using stockmodels in between to avoid air-cutting - could use 1 or 2 tools do do it in this example. What comes next?

 

Of course I could finish it also with OPTIREST (stock on walls and floors put to 0) - but this is not the answer (bad surface finish - wall cutting should be a continous path).

 

Is there any technique with which I could finish VERTICAL walls and FLAT floors with the more traditional 2D method, but do it in a more automized manner (like opti with stock models)? Instead of assining chains (wireframe) - pointing the whole solid body and boundries? To be honest, right know I do the job manually. That is, after roughing and rest-cutting i pull out wires and finish the parts with 2D Contour and pocket toolpaths. As you can see - it is quite a hassle - even in the example I given here - the programming time is not minimized and you have to be careful not to make any error (opti rough takes care of it for you - but only roughing)..... Could anyone give me a hint on that? Can MC handle that (I know about FMB even though never used it - my intuition tells that it's not what I'm looking for- but maybe I'm wrong - please correct me if that's the case)?

 

One other thing. I am not trying to find in MC a scenario like loading the solid, picking 3 or 4 tools, pressing the green button and wainting until it figures everything out for me and outputs the COMPLETE program... :-) It's not like that - things like that don't exist - there is always a place in the manufacturing technology that you have to put your hands on - no doudbt about it - you are the technology writer and you are in chargé of everything at the end of the day - not MC or any other CAM tool. - I just want to push it a step forward so I could really get some more help out of it.

 

Thanks in advance fellas. Sorry for such a big story.

2D_PART.MCX-7

[Chris Rizzo]

The h.s. horizontal finish toolpath is a slick way to finish everything flat very easily.

[Bob W.]

In my experience it is a balancing act between programming time and machine run time. If you want to minimize the programming you could finish the flats with the horizontal finish toolpath that Chris mentioned, then finish the walls with the waterline toolpath (3D HST). It will most likely run longer on the machine and you will have less control during programming but it will get the spindle turning sooner than programming every feature with traditional toolpaths. I haven't looked at your file but my preference is to program with the traditional toolpaths. With care in chain selection and good use of incremental depth selection I can come up with a pretty efficient program quickly. Either way I would utilize the HST horizontal toolpath.

[Chris Rizzo]

What Bob says there. It's all a matter of how much programming vs. running time you want.

 

 

Is there any technique with which I could finish VERTICAL walls and FLAT floors with the more traditional 2D method, but do it in a more automized manner (like opti with stock models)? Instead of assining chains (wireframe) - pointing the whole solid body and boundries?

 

You just described FBM to a T. I just looked at your file, and a guy I know who spent some time setting up FBM, would program that part in about 2 minutes. Yes it's probably 2x longer run time than picking and choosing geometry, but if you're making one part, it's a no brainer to use it. You could literally program dozens of these parts in a day. If your business consists of one-off parts like these, I'd give FBM some effort. If you do this type of part once in a while, I probably wouldn't bother and just program old school. Also FB drill is VERY VERY cool, even for just creating the operations. Again though, you have to spend some solid time setting up your FBM library. If you just try it out of the gate as-is, you probably will laugh at it. Personally, I do too many random jobs to use fbm, but I have seen what it can do when used properly.

 

Do you want a beautiful program that takes several hours to create, for about 10 minutes of cycle time, or a 5 minute program that has a 20 minute run time?

[Maclaw]

Thanks Chris, Thanks Bob,

 

I'll try the hs horizontal and have a look at waterline (although it is 3D...).

 

I guess I'll have to give FBM a closer look. Although did not like it after trying it off-the-gate as-is... You're right Chris - it was a big laugh :-) Especially when I saw like a milion OPS in the OP Manager... Now try to control that.... :-) Anyway... Time to give it a bigger glance than just pressing the green checkmark - then I might laugh out of myself :-))))))...

 

Gotta get back to work. It's almost 10 am over here. I'll be in touch Guys.

[Newbeeee™]

Waterline 3d is very good.

Horizontal I have never had much success with - too much air cutting. Facing can work well for islands - click the solid face and voilà.

One thing is for rest machining you can cut your cycle time in half by running from a roughed stl model. Looking forwards to using stock models...

[MrFish]

A nice function would be to have a chook or just a "create step chains" function that would behave in a similar way to what FBM does and create chains at each step depth to aid in chain selection when choosing contouring chains.

[ajmer]

mr fish, what i have done is run fbm just to make the geometry. run fbm and delete the operations.

the geometry gets savedto a different level.

[Glenn Bouman]

That part is an excellent candidate for FBM, but I like Ajmer's method as doing the part with FBM will not give you Dynamic toolpaths.

Maybe in the future?

[MrFish]

mr fish, what i have done is run fbm just to make the geometry. run fbm and delete the operations.

the geometry gets savedto a different level.

 

Yip this is what I do too but does take time to crunch all the tool paths that you won't even use !

[metalmansteve]

I agree....it would be huge if there was a finishing option that allows you to select this solid and finish in one shot. Extracting and selecting a million chains and creating a million operations seems archaic to me.

[Newbeeee™]

I agree....it would be huge if there was a finishing option that allows you to select this solid and finish in one shot. Extracting and selecting a million chains and creating a million operations seems archaic to me.

It is compared to how easy optirough removes 90% of mtl in one click!!!

hmmm, enhancement request required for a '1-click finish pass'...

[Aaron Eberhard]

There is a toolpath that will let you do it, you just have to change how you're thinking a little bit Note that I didn't have any wireframe, in fact, I deleted it all from the file to prove it

 

If you use the Opti toolpaths (like you did for roughing), with the same step down and step up, it'll clear horizontal paths.. Check out this file for an example.

 

What I did was an OptiArea path to clear out the inside with a 10mm cutter, then created a stock model based on it.

 

I then use OptiRest based against that Stock Model with a step down and step up of 25mm.

 

That will give you what you want, with a dynamic toolpath to boot!

 

Hat tip goes out to Dave Conigliario for that one...

2D_PART.zip

[MotorCityMinion]

"Of course I could finish it also with OPTIREST (stock on walls and floors put to 0) - but this is not the answer (bad surface finish - wall cutting should be a continous path)."

 

IMO.... nO, U CAN'T. iT'S A ROUGHING OP, PERIOD. The horizontal strategy is probably your best best for finishing walls and floors simultaneously but you lose cutter comp with this strategy. If part tolerances can accommodate it, go for it. Limit your entry Z heights, say .02 above the floor, play with the entry - exit moves, extend the flat faces outside the vertical walls some. Like others have stated, sooner or later programming time will exceed the cost of a few parts. Got a significant quantity then spend time programming the wire frame.

[Maclaw]

Hi Aaron,

Hi Fellas,

 

Aaron thanks for the file. But this unfortunately is not as easy as it seems to the end... :-(

 

I agree 100% with MotorCityMinion. OPTI is not for finish. It's a HS roughing routine - when approaching tight corners - it makes loops (and VERY GOOD - that's it's job). This is a great thing 'cause it takes a CONSTANT SIDESTEP and makes the toolpath "sharp-corner-less", which in turn makes the machine tool run with no jerks at high feedrates. This is very cool and makes HSM work. Nothing to add more to this - all clear!!

 

But when finishing, the vertical wall showed on the picture (as an example) is re-machined by the loop toolpath (at tangency points), which causes micro-gouges to the wall (Every end-mill is deflected more or less by the material it machines - if you remachine it with the exact same toolpath - you will still take material off the wall and see "micro-chips" fly :-). That's why when you want good finish - You have to have a CONTINOUS toolpath with as few entry-exit motions on the wall. I hope You know what I mean :-) On top of it all - You can't use cutter compensation (it thinks in the 3D world of toolpaths - and good - after all why use cutter comp for roughing...) etc...

 

I think the opti toolpaths as HS roughing routines combined with STOCK MODEL and rest machining are a great thing that make Mastercam a 21st century tool - for 3D and 2D. No doubt about it.

But on the other hand I have a feeling that if CNC Software made some 2D finishing paths that would work with SOLID ENTITIES (not chains - not counting the boundries) and stock models, MC would truly be 22nd century no 1 CAM tool in the business!!! Observing the options and functionalities that MC has now - I think it's a relatively short road to achieve it.

I like MC because I have a lot of control in the toolpaths and am not limited when coming up with an optimal technology to produce a part. that's very important for me. MC is my tool and I tell it what to do - not the other way around (like in some other CAM systems). But the 2D..... :-) ........maybe sometime....?

[mig]

mr fish, what i have done is run fbm just to make the geometry. run fbm and delete the operations.

the geometry gets savedto a different level.

Hi Ajmer,

How to make FBM (what option) to create geometrys ?

[ajmer]

Hey Mig,

if you just run FBM it will create geometry for you automatically

you do not have to set any options

look at your level manager after you run fbm

[Rob Dusenbury CAD/CAM CONSULTING]

mr fish, what i have done is run fbm just to make the geometry. run fbm and delete the operations.

the geometry gets savedto a different level.

 

You don't even have to do all that. When you are in FBM hit the detect button let it find the depths then cancel out. The chains will be blanked on level 1001 just un-blanked them,and program away.

 

BTW The information that Aaron gave is valuable. Using opticore with the step ups equal to the step down is huge. That one path will do all of your roughing. Copy it an switch it to finish waterline will take care of the finish work. This part can be done with two ops. If you don't like waterline for finish then use the old finish contour (detect the flats) and you are on your way. You could even use Rough pocket and turn of the rough and just use finish. This option gives you lead in and lead out and thin wall. Mastercam has what you are looking for, all you have to do is look outside the box,

[Glenn Bouman]

Mastercam has what you are looking for, all you have to do is look outside the box,

 

I agree, but would you expect a customer to have to buy Level 3 if they only did type of work?

[Rob Dusenbury CAD/CAM CONSULTING]

What I meant by look outside the box. If you already had a level 3, then why not take advantage of it, and do this type of work with all of it's capabilities.

 

I agree, but would expect a customer to have to buy Level 3 if they only did type of work?

 

Why not if it will save him money in the long run. This type of work is tedious, and chaining can take up a good part of the day. If a customer can simply do this without all that tedious programing, then yes it would be worth while to purchase a level 3.

[Glenn Bouman]

But when finishing, the vertical wall showed on the picture (as an example) is re-machined by the loop toolpath (at tangency points), which causes micro-gouges to the wall (Every end-mill is deflected more or less by the material it machines - if you remachine it with the exact same toolpath - you will still take material off the wall and see "micro-chips" fly :-). That's why when you want good finish - You have to have a CONTINOUS toolpath with as few entry-exit motions on the wall. I hope You know what I mean :-) On top of it all - You can't use cutter compensation (it thinks in the 3D world of toolpaths - and good - after all why use cutter comp for roughing...) etc...

 

 

I know a few customers that do these types of parts (electronic enclosure boxes) and the tolerances and surface finish requirements are very tight (Maclaw's points are valid) and it is a high production environment where seconds count. Using Level 3 cutter paths are going to give a longer cycle time.

There is an opportunity here for Mastercam to enchance the FBM function to address his issues (and make it easier for us to sell more Mastercam). Even if it was something simple like putting all the geometry slices on separate levels and using the z position as the level name.

 

[Aaron Eberhard]

Whoops, I missed the reference to it in the first post

 

Either way, valid points made all around, I'll see what people think about it

 

Thanks!