Let's talk about CAM

As to 'finish' and accuracy of cutting any CAM package, even the free ones can handle this well if the user knows how to use the software. We do very precision milling with Meshcam running on both windows and Ubuntu. The overall 'quality of the part has more to do with the machine than the software. Even crude CNC software can handle .001 accuracy pretty easily.
We have about $35000 invested in CAD/CAM software at our shop but actually use only the smallest and least expensive packages for production work. The high-end packages have a lot of 'options' but nobody ever uses them. It has been our experience that the so-called 'soilds modeling packages' are the least beneficial to a machine shop. They are great in the design environment but once the part is created it still needs to eventually be converted to gcode and the high end packages seem to come up short in this area.

I re-read your posts and on one hand it appears as if you're very familiar with the process of design to code but at other times it appears as if you have never actually used any of the packages you mention and you haven't yet told us about the capabilities of you milling machine. A better understanding of what you have will help a lot in understanding exactly what you want as you go forward with your endeavors. What kind of machine do you have first of all and secondly do you need full 3D or just 2.5D code for your projects.

Why do you say that?
I have worked at a thermoplastics composites company and we made far more complex parts on 3 axis machines with no trouble at all. I see no reason why that part would be the least bit difficult on a 3 axis.

Because it is very important to select the CAM Software, For the second port I would not even need a cam to produce that one on my manual mill. It is just a matter of selecting the right fixture. and several production steps. If you want to earn money, you will need a 5 axis mill, at least here in Germany.

So you can go on and ask for a CAM software, but how you want to get a recommendation without such essential information???

I personaly like to use 3 CAM

- VECTOR-CAM for simple 2,5D parts
- Alibre CAD/CAM for more complex ones
- Sprut CAM for very complex ones

All of them are worth there money in my opinion and all of them are availible as DEMO Versions, so you can check out all of them.
I do know the software from different companies and they need to earn money. I at home do not need a CAM for my hobby work, because my parts are all 2,5D and most can be done with own cycles or free stuff like gsimple.

So the question is: What do you want to do? I beleave you don not know that in detail yet. Making the first part you mentioned (the mold) on a 3 axis mill is between impossible and hopeless.

Norbert

gww250 wrote:

Meshcam is a very good product at what I consider to be a pretty low cost. We use it daily and haven't found anything yet it will not do...

I used it in the trial, and found plenty it could not do.

I own an "expensive" CAM program that is very powerful, but pretty cumbersome for simple parts. I was looking for an alternative that could get the job done faster. With MeshCAM, the material can be oriented in seconds (rather than minutes). Unfortunately, the drilling feature is brand new (as of Dec. 2013), and VERY limited. You cannot use a reamer, because it only drills from largest to smallest. (Well, sure, you can manually reorder the operations in the output file.) Also, you cannot peck drill a hole; it only plunges in one move. If you can deal with those two limitations, and with only one tool for roughing and one tool for finishing, it might work. Personally, I gave up on it.

My gut feeling was that it was more designed for wooden sign routers rather than machinists. I would probably love it if that was my goal.

gww250 wrote:

I re-read your posts and on one hand it appears as if you're very familiar with the process of design to code but at other times it appears as if you have never actually used any of the packages you mention

I have been doing basic mechanical design and fabrication for about 15 years now. My last job of about 4 years moved me up into doing CAD design, advanced mechanical design, advanced fabrication, lots of experience on the manual side of the machine shop and some experience running machines.
I have also been involved in making machines for laying up and processing thermoplastic composites. I have also been involved in some of the more basic elements of software and hardware that made those machines run.
Unbelievably CAM was one thing I never did. I asked them many times to get me involved in it but they had higher priorities elsewhere that needed my skills and I never had the time to get into it on my own.
So yeah, I have a lot of experience with everything that surrounds this one gap in my education but this is my first time trying to get familiar with CAM.

gww250 wrote:

and you haven't yet told us about the capabilities of you milling machine. A better understanding of what you have will help a lot in understanding exactly what you want as you go forward with your endeavors. What kind of machine do you have first of all and secondly do you need full 3D or just 2.5D code for your projects.

This is exactly what I am trying to figure out right now.
I am learning but still don't fully understand where the line is drawn between 2.5D and 3D and exactly what the capabilities of each are.

A lot of the work I will be doing will be easily accomplished by 2.5D. I want to cut flanges, gaskets and other things that should be quite simple.
On the other hand I don't want to pay for or get familiar with one software that will do that only to find it won't do other things I need.
I feel like I would be best off getting a 3D program off the bat so I'm not limited. I am also quite confident I want to be able to make several different steps in the process. For example face the top of the part, drill or profile cut holes and then 3d machine a shape.
There is however much I don't know and things that I may not know I need or don't need which is why I am here asking for input.

Unfortunately the machine is also something that is greatly up in the air as far as it's capabilities.
I designed it and am currently building it so I won't know for a long time what it's capable of. I am also on a tight budget so for now I am using cheap linear motion and my first attempt will be with an RC brushless motor spindle. From there I will push it's limits and upgrade parts until it can do everything I want it to do.
It is small, about X 400mm x Y 600 x Z 200 travel.

This is very much a concept drawing. The final machine will be very similar but some things may be changed like reinforcing the z axis or other changes.




The frame and gantry are made out of 6" by 2" 1/4" wall steel tube.


I hope to finish the gantry and start assembly next week.

For now I will not be machining it but it is designed so it would be easy to get the base ground, mounting surfaces machined and locating grooves or pads added.


As far as some of the more advanced stuff I can think of right now that I would like to do. That forming die will probably be one of them and is pretty high on my list of things I want to make.
I would also like to be able to machine 4 cylinder head combustion chambers and do other things like that.
Down the road I would like to get something like an a/b axis or even design a fifth axis head or something to expand capabilities further but that would be a ways down the road.

I didn't say that Meshcam does not do pecking. I have the old program and have been doing peck drilling, pocketing, counterboring etc. for several years. This program is very flexible and can do just about anything if you get to know it well. We use it for some very complicated parts without any problems so it certainly isn't a 'router' program by any means. WE also have some high end CAD/CAM packages but seldom use any of them except Auto-Cad and Igems.

OK, now it is clear:

Poor machine, hobby use leads to:

Don't by a CAM for beginning, just begin to learn with free tools. One good one to begin is:

gsimple

If you are able to make parts with that tool, take the next step and look for a more profesional CAM.

Norbert

Norbert's advice is good. To start you really don't need a cam package. Better to start with a good 3-d drafting package, not a solid modeling package, and then get good with it. Use a free dxf to g-code converter and start making parts to get the machine dialed-in. Then look for a CAM package. I still suggest Meshcam. It looks simple and elementary but it's very powerful and very flexible and works well with Linuxcnc right out of the box.

I have checked out meshlabe for a little bit and it does seem to be pretty nice all around. Seems more capable than all but maybe heeks while being much more user friendly than heeks.

It seems odd to me that all these programs use meshes. Is it really that much cheaper or easier than using something like a STEP?

I am curious what you mean by 3-D drafting. I have a little experience with drafting programs like drftsight and librecad and know that they have some ability to sketch into the 3D but I have never looked into it much more than that.
If that is the type of 3D drafting you are talking about then I have zero interest in learning something that takes longer to make, longer to edit and gives you less control and functionality than what I already know and use.