Fusion 360: Reducing Cycle Time on 3D Toolpaths! FF61
Vložit
- čas přidán 5. 09. 2024
- Let's review each CAM operation in this 3D part to improve the cut recipe and, hopefully, drastically reduce cycle time!
PLEASE UPVOTE THIS IDEA! autode.sk/2iKCAAJ
Wireframe View Plugin: bit.ly/2gphjue
~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Reach us / CNC Info:
Speeds & Feeds: provencut.com
Download Fusion 360: www.dpbolvw.ne...
Online Fusion 360 Training: bit.ly/LearnFus...
Hands-On CNC Classes: www.nyccnc.com...
SMW Products: saundersmachin...
CNC Resources: www.nyccnc.com
Music copyrighted by John Saunders 5 Reasons to Use a Fixture Plate on Your CNC Machine: bit.ly/3sNA4uH
Want to REALLY decrease the time? DRILL out the spots where you are spiraling in! Clear what you can with a DRILL, they remove material faster than anything else
if you click on your account name in the upper right hand corner, then go to preferences >Manufacture> Show Operation machine time. then it'll show all the different machining times next to each operation.
I must say that even as an experienced user, I pick up some great tips from these videos and see some other ways of doing things! cheers!
I did vote for your suggestion in the ideastation, but as an aside, you can also see the cycle time for each individual operation on the setup sheet. Check it out, takes a few mouse clicks, opens in your browser, and you can go back and check the times easily if you are playing around to make the cycle faster as shown in this video. Learn a lot watching your videos, thanks.
Why not drill those hole FIRST, then use that clearance to do a straight plunge at rapid feedrate?
You just created my favorite cam video! Really felt like you let me into your head and let me understand what was going on instead of showing some kind of procedure. Thanks
Best toolpathing video I've seen.
Here is another trick for facing. Try using a 3d adaptive path but then on the heights tab make your bottom height model top. This makes a path that has constant tool engagement and always on the same side of the cut ie always climb milling. With that you can use a much higher spindle speed and feed.
wow, that part looks just like the soup dish that Brian Oltrogge did... he is the guy that builds the platform CNC
John Sheak *soap ;) XD
Thick soup, perhaps. 😋
soap.. yes... I cussed so much as a kid I get them confused sometimes ;)
If the fusion team just focused on the bugs found in the Fusion Friday vids, I'd be a happy camper.
Two suggestions:
- Maybe for the pocket/adaptive, you could replace the plunge/helix with 21 drills in the big holes of the part.
- And for the parallel, did you notice that on the long edges the surface is almost flat blue? I suspect the flatness cause it to do very small passes with the tip of the ball, and you might save a lot with a flat end mill. You should constrain the parallel to the sub-area that's really 3D (also maybe exclude the flat edges around the top holes).
I still caution using 3D adaptive paths with a ball endmill. It has bitten me a few times by gouging parts.
This might actually be a good application for plunge milling to rough it out
And instead of facing, go right to 3D adaptive to clear all the pocketing. Eliminate the face milling entirely, it's pointless as there is virtually no top Z height surface.
Why surface that flange or flat sections with a BEM? Huge waste of time there. Just use a horizontal path and nail all the flat surfaces at once easily.
For surfacing, yeah, parallel is a bad choice, ramp would be better. Also make sure to turn on additional Z step down passes to ensure the steep areas are surfaced correctly. Looks like some sections have very poor surface finish due to the big step down the tool will be taking. And limit its cutting by angle (.001 deg min 90deg max) to have it avoid surfacing the flat sections that were done previously with a horizontal op.
When I'm trying to figure out which operations are taking the most time, I select the entire job and hit simulate. From there scroll down to statistics to get the total time and look at the progress bar on the bottom of the simulation. Each operation is given a different color. So which ever color is the longest is your longest operation. This doesn't give you exact numbers but it does give you a good idea where you spending your time.
Literally just saved me 4 hours tonight. You rock!!! Liked and subbed
You had a few nice strategies for reducing cycle times on the part. Good work. Thanks for that.
Umm... errrr... but you abandoned the surface finish in search of a lower cycle time. The part would probably not had a good surface finish to begin with, but you have the knowledge and skills to correct that. Try looking beyond the problem a bit.
The part is better suited to z-level finishing than planar finishing. Planar works well for parts with flatter profiles. Z-level works better with steeper walls. Yes, the final pass will probably take more time to machine with the Z-level but the surface finish will, at a minimum, be passable.
Segregating the part into a “not very flat” section ( the section in the middle) and a “relatively flat” section ( on the periphery) and treating each differently is a worthwhile strategy. Each can then be re-roughed and finished with different cutters best suited to the surface. It can be a bit more time consuming to set up, but can yield nicer results and reduce cutting times.
On a technical note try leaving more material after roughing cycles so the finishing mill has material to work with. In your example you left .005 ( or 4% of the .125 finisher ) That's probably not enough.
Larger mill wirh rounded corners and trochoidal path, use a drilling op. to remove most of the material in the through holes...then there is very much less need for plunging or spiraling...
Your retract plane issue is not in the Heights tab, it's solved in the Ramp subsection of the Linking tab.
Get rid of the facing in the beggining...all this material is going away anyway! Also the contour could be done in the end so that there is less material to be removed and more aggressive cutting data can be used (my understanding is that the customers machine has limited horsepower hence the facing with a 1/4" tool). As for the weird holes...I would go for the concave ones first and leave the convex ones internal features entirely (the helical plunge) for the second operation when doing the other side.
Safe plane should be in the last tab I believe. It gives you a reference from something I forget what but it’s commonly overlooked and skimmed over without much thought.
Safe plane is the absolute no runins possible type of plane. It can move anywhere and won’t break your tool.
Nice video John! I'm not sure if you know, but when you simulate your setup you can hover over the operation on the bar (where you scrub through the simulation) and it will show you machining time for that op in a hover window. That isn't exactly what you were wishing for but it might be a little easier. cheers
Another youtuber? This sure rings a bell as a soap dish made on a router, and at least originally had an interesting flip tab setup.
wow... I want to see already the final machining video
What did the final modified machining time come out to be with the adaptives and your modifications? I saw the times for the 2D contour and the Parallel operation, but not the final setup time.
NYC CNC I was thinking you were going in chronilogical order. I saw that but thought it was before you modified the cycles. Thanks!
im just saying that i use these feeds for Steel. If you are maschining wood or plastic u can almost allways use the maximum speed and feed !
U could safe so much time! Also for your projects!
What was the final time savings? You had the "pre tweak" number, but not the "post tweak" final number.
Plung milling would be a win on this part. I understand Fusion is very affordable BUT it produces a LOT of bad tool path. Long time programmer here, I have used many CAM packages. Fusiuon in a production shop would not be for my shops.
I am not a fan at all of parallel strategies. Not just because it can create poor finish, but because it also puts alot of wear on the Z ballscrew. I definitely prefer ramp. I would probably do the facing and stage 1 roughing with a much larger tool then come back with a smaller tool to get in the pockets. It makes sense in my head, but checking in CAM would be prudent...I think limiting the depth of the pockets that go all the way through to around 10% would reduce overall machine time since you have to rough out those pockets on the flip anyway.
Completely agree with the 3D adaptive and bullnose suggestions. With the ballnose I would consider doing a touch/avoid on any flat surfaces which could be finished with a flat bottom tool.
chronok
Parallel strategies put too much stress on the Z axis with the constant up and down movement. Z axis has a great deal of weight fighting against gravity (not so bad with a properly counterbalanced spindle). I prefer ramp because it tends to be smoother, less likely to gouge and saves wear on the Z. :)
Im starting to use largest tool you can that doesn't sacrifice dimensions and finish of the part. Id almost face mill the top save some time thier. I do like changing that ramp angle saves alot which im going to try myself
You also have flat areas around the edges that could be machined quickly away with a thick flat end mill. Isn't that a better approach?
At 16:50 I think you are talking about "chip thinning"
Ramp with 50 % off feed is a normally good.
A 63 mm face mill who'd been the best white 50% step over for the face and 3 mm cut depth.
Hi as a newcomer to CNC with an older DMG 50T I have found CNCCookbooks Gwisard wonderful for speeds and feed take a look if you haven't already
Nice tips Master, thanks for sharing
Hi John. My name is José Horta from Portugal and I've been following your channel for a while. I teach technical drawing, cad, cam and cnc in a vocational training center here in Portugal and I've been learning a lot from you - thank you for that!
Considering this video, is it a nonsense to set the tool down most of the time in the linking parameters!?!? I've tryed that in one part I was working on and it reduced the time a lot in an adaptative strategy. Meanwhile I've raised the repositioning feed also to a higher value.
Did you ever use Mastercam? If no why not?
you never plunge in wood you will set it on fire or burn the cutter big no no, you ramp always.
Feed optimization is used if your machine is crap at going around courners
daniel lyall
I use feed rate optimization all the time, and my machine is far far FAR from crap.
Going around sharp corners means the tool is accelerating (acceleration is any change is speed OR direction) by slowing down around corners (using the correct settings) you stop the endmill from accelerating too fast around a sharp corner causing poor uneven surfaces compared to other sections.
It's also used extensively in 5 axis surfacing operations and pretty much a must to know how to use when doing full 5 axis surfacing.
Lol. At our school, we tried cutting into some very thick wood without realizing that. We almost set off a couple fire alarms. We had black smoke all over shop.
Occams Sawzall I know it's just one of the reasons to use it
you can always give it less ripums 'revs' or speed up the plunge speed but to make it easy, ramping is easiest
Please pardon my ignorance if this is a dumb question. Would it be faster to replace the plunge operations with peck drilling?
TL;DR: Roughing by drilling tends to remove material faster for less money than roughing with an endmill. blog.cnccookbook.com/2010/09/24/interpolating-holes-vs-twist-drills/
Were it my part to run, I'd get the biggest drill that I could chuck up that would fit in those holes. 22 drill hits (Or so it appears) is a lot quicker'n diddling about with helixing in an end mill. If you're really paying attention, you can get away with things like partial depth drill hits and just clean up afterwards with an endmill. (AKA "Plunge roughing")
You can get a crazy high MRR with a drill and some attention given to optimizing the CAM ops.
Andrew Foss
The one down side to drilling out for roughing is that drilling is one of the most power hungry operations you can do.
Judging by the speeds and feeds and tool sizes used, my guess is large dia capacity and HP are not something available.
My gripe with roughing by drilling is the multitude of interrupted cuts you have to now make with the tip of an endmill to clean up floors and such. I'd rather spend the extra few minutes helixing and running an adaptive path.
When doing surfacing your optimal feed rate changes based on the "chip thinning" effect. When the cutting edge is at a 90 degree angle to the feed direction it produces a 1:1 chip, that is, if you advance 0.001" per revolution you will cut a chip 0.001" thick. If the cutting edge is at some other angle to the feed direction, as is the case when taking shallow cuts with a ballnose, small radial stepovers with an end mill, or using a high-feed milling cutter, you're essentially spreading that same chip over a longer cutting edge, thereby making it thinner. To compensate for this you need to increase your feed rates accordingly.
Helical tool has a good writeup on it here: www.helicaltool.com/secure/Content/Documents/Tech_AngleofEngagement.pdf
So this is a soap dish designed by Brian Oltrogge. He designed the CNC router I have next to my Tormach. it's called Grunblau. Great machine.
didn't showed resulting total time. But interesting.
I think ive seen this part somwere before.
15 minutes in and im still wondering what in the world this contraption is supposed to be used for... lol.
Soap holder
I don't think you can just increase the stepover just because you feel like it. 0.060" is pretty damn big on a 1/8th tool. Even 0.030" is too much for something that size. The Scallop is too deep I think. 1/4" Ball at 0.012" stepover is kinda like the bare minimum for any kind reasonable surface finish. I see it at work all the time, and that sometime requires a little handwork to smooth out still.
In that project of buddy's, what is the size of the holes and surface rads? Could a 3/8th Ball fit in there, so that you could you use a bigger stepover, but retain the surface smoothness?
:)