FWIW, I have only worked with Fusion and have not seen an option for this…
Do other CAM packages compensate (up or down, appropriately) an optimal linear cut feed rate (say, you put together a recipe that runs a machine at about 80% of the max HP/TQ of the mill) to the circular/arc paths that it generates or do they all just assume you are setting a feed rate for the overall operation that accommodates all situations (which ends up being slow enough to cut large arcs but could be faster for the other parts of the operation).
I ran some test cuts using a straight spiral facing operation to find feed/speed/DOC/WOC that are sustainable (machine could cut continuously without bogging down). The formula to compensate a linear feed for an inner diameter circular (which is a reduced feed rate) and works to predict sustainable cutting parameters for a pocket/adaptive 2D, provided that I use the largest diameter encountered in that tool path to reduce the feed. There are smaller diameters throughout the tool path where you could run a faster feed closer to the linear cut rate and its only at the end of the cut that I need to slow down (having tried faster feeds where I have to manually drop the feed toward the end when it starts to bog down).
On these size machines using 2-8mm end mills aren’t you talking about only a +/- a few % on feed rates?
You would probably be better off having a tool change than eking out a couple percent on a giant mathematical calculation.
I make stuff for my cars/motorsports, so speed optimization for all situations with the hardware you are driving is how my brain is wired to be able to drive on a race track (road course) as fast as possible.
On a PCNC 440, I have recipes to use a 0.75” (19mm) 2 Flute Mini Shear to rough aluminum.
I have one example where it could be fed maybe up to 30% higher for some parts of an op until it needs to be backed down for the large radius move. I found running the same op it would run okay most of the way at a higher feed rate but needed to be backed down when it encounters a large radius at the end of the op. It runs the whole op with no intervention at the slower feed for the large radius.
It is basic math and the computer is already doing 99% of the work to generate the G Code. Why not have it spend a few more CPU cycles at final post-processing to output the optimum feed for each move and get some extra machine performance for “free” from software using the hardware within its operating parameters? At least have the option to do so…can be left off for those who want to stay closer to a very safe feed.
I’m sure it’s out there. Probably In one of the luxury adding to fusion that cost more than the whole program. It sounds like if you can run 30% faster you could convert to more of a high feed milling where you get more step down for less Fz
Im usually at max rpm on my machine at feeds of 300-600mm/min so this is out of my territory.
All of the cam software packages I have used uses approximate speeds and feeds. If you require optimization you will have to do it yourself. Too many variables in metal composition, cutter composition, depth of cut axial and radial, horsepower, coolant compensation and rigidity all play significant factors.
I wanted to join in the conversation and add a couple notes. Feed speed is calculated at centerline of tool. When the tool changes direction the feed rate at outside diameter of tool can be much more than the programed speed depending on programmed radius and tool diameter. Haase and others have videos that explain this in depth and how their controls can be used to compensate for this.
I use Sprutcam and it has a couple ways to control feeds based on a number of conditions, like when cutter has full engagement, ramping, next feed, plunge and other tool movements… you can set the feed to a lower or higher % of the full speed normal doc, woc feed. These settings can be tool savers at times and tool wreckers at other times. It all depends on the user to check what is going on if they are enabled and used. You can also control how it cuts inside corners to avoid chatter based on different conditions. All these settings allow a great deal of control of the code that is generated for different conditions. That said there is nothing simple about using the many options available and they would require an investment in time to learn how to best use them for the best results under all the different tool path cutting conditions. I do enjoy and trust them a lot when running code on my 24r router cutting woods, plastics and composites. Makes it easy to have good code with very little effort. Often in just a couple minutes. Lots of room for error. I tend to carefully control all paths, feeds, speeds, woc and doc for parts made on the 1100 mill for best drama free results. The time spent setting up these programs can be considerable because I tend to control everything and not let the software use much of the automatic options and operations available. And of course, lathe setups and programs don’t have these options and require a whole different set of cutting strategies and options to understand, set and keep an eye on. While cam vendors claim their software is easy to use. I find the thousands of settings a serious challenge to use and fully understand. No free lunch here.
1 Like
What you don’t enjoy the thousands of fields that take the input values and insert them into mystery formulas (with variables created from other values generated from mystery formulas)?
1 Like
In this case, the 30% faster is all I’ve got to work as the Mini Shear is an insert tool and doesn’t really have more cutting length for a greater axial cut. For another tool, going deeper/more step down may be possible but with any tool, depending on the tool path radius, as I understand, you will always have a correction factor for the internal/external cutting radius. On the surface, it seems like it is possible to find an optimal linear cut for any given speed/feed/DOC/WOC where the CAM should be able to adjust from that linear rate to go up or down when it is on an arc to optimize…
Agreed and that’s where the various calculators come into play to help optimize further than the basic feed/speed calculations in the Fusion CAM package. What I’m seeing is that the calculators can help you optimize a linear cut and you can correct that up/down for an arc but Fusion CAM (without the manufacturing paid extension?) looks like it only takes into account a feed as an input that will run during the worst case for an adaptive clearing…which is the largest inner radius it can construct while the rest of the moves in the path are going to be forced to run at this slower feed…
Thanks for the reply. I think I might have to look further at SprutCAM and some of the other CAM packages trials/learning editions to see what they offer in terms of configuration over Fusion.
There appear to be at least 5-6 options among the “popular kids” to evaluate:
- Fusion standard edition
- Fusion (manufacturing extension to extend CAM functions beyond what I’ve got)
- Inventor (should be an easy switch from Fusion since the interface looks similar)
- MasterCAM
- SolidCAM
- SprutCAM
I have never used Fusion. I would guess just by the large number of people that use the package it would be the best option overall. Inventor is pretty expensive and I know you can draw sweet cad models but never seen machining cam examples. Mastercam is what they teach at the local schools but the quote I received on it years ago was over $6k for 3 axis. Solidcam was about the same very expensive.
I started with Sprutcam about 10 years ago and did not want to change because of the investment in time it takes to learn a new cam program. Over time learned to work through the frustration of years of upgrades and countless bugs. It has an open data model that allows the user some flexibility to do things with the software the authors didn’t know could be done. “the part I like the most” An experienced user can create custom behaviors, tool paths, tools and in short do almost anything you want with a tool and x, y, z, a axis complete with detailed fixtures and 1-6+ sided part programs with dozens of g5x.x offsets all in one g-code file. It just takes time to figure out a way to do it or trick the software into doing it. Newest versions are very stable and work well with few bugs that most people would run into.
But still no free lunch and this program will take time to get good results and years to really master. As noted above It’s still a complex program with thousands of settings and dozens of ways or options to get results. Side note, it has only very basic CAD ability and I use a Ironcad to produce my models. The advantage is I can create complex models, break them into part model files complete with defined stock and the machine offsets for g54-g59+ predefined and imported into Sprutcam. Drawback is the models are only one way with this cad so changes in cad model are not automatic in cam model and require reloading and op setup.
Anyway good luck
Cam software can be a big investment in both money and time to learn.