I hope I'm not biting off more than I can chew At 76 yrs old, I don't have time to make a lot of mistakes or learn new tricks. I'll try to document my progress on the crucible lift, and the CNC build. The hotrod has gone on a back burner while I take care of these other projects and try to keep up with summer yard work, gardening, boat repair, motorhome maintenance, etc. etc. etc. The most intimidating part about the CNC for me is the idea of learning how to do the computer work to design projects and then cut the foam patterns. I'm a computer illiterate, for the most part. Again, thank you Kelly (and all on this forum) for the inspiration and the great information that helps all of us on our journey through the world of metal casting.
Wild Irish: The 3D CAD learning curve is VERY steep at first. It doesn't help that there are two major branches to 3D CAD, direct modeling and parametric modeling. FreeCAD, Fusion 360, and some others are parametric modeling based. Designspark3D, Autodesk's old 123D Design (You can still download this software, just not from Autodesk.), I think TinkerCAD and maybe Sketchup too, are direct modeling based. I've been doing 2D CAD for about 40 years now and I had/have a difficult time with wrapping my head around parametric modeling. There have been studies done that show some people's brains just aren't really wired for it. Direct modeling seemed/seems to make a lot more sense to me. If you are struggling with the 3D CAD, either parametric or direct, try the switching to the other style of modeling - it might ease the mind drain a little. (It did for me.) Don
You don't have to do parametric modeling. You can treat 3D cad the same as 2D. I learned AutoCad back in the 90's. (Back in.... feels like last Month!) Since then I have transitioned into Fusion360. I still do a lot of l2.75@90 on the keyboard. and placing construction lines up faces for spacing. Fusion is basically a collection of 2D drawing with extrusion (height). you draw a 2d Drawing (sketch) then extrude it (give it some height) then select a face (or plain) and create a new 2 D drawing on that. If you want to add a feature join the extrusion. Want to remove some cut the extrusion. it takes a little bit to get your head around it. I'm not nearly as fast in Fusion as some of these guys but I can get my parts drawn. (the CAM side is a whole new and different critter! Good Luck! (if you get too stuck on something ask. We'll help where we can or point you in better direction to get the info you need. Bruce
It really doesn't take much of a beam to support a 100lbs or so over an 8ft-10ft span. A couple of 2x6/8s will do. Also helps if you have a dedicated space for it. Mine is more elaborate because it's on wheels movable. An electric hoist is helpful, but if you don't have a lot of change in height, a lever arm attached to the trolley can do. Whatever you do, simplest is best and think about what can go wrong and what you would do if it did. I agree with the other inputs above. The leap to 3D modeling was the biggest for me. On the CAM side, I'd suggest you try one of the economical CAM programs. I use CamBam but there are a number of what would be considered entry/hobby level CAM programs. Almost all of them offer free trial. CB has 2D CAD capability and very limited 3D, but, you can go a very long way with its CAD and 2D/2.5D CAM.......and it does everything I need it to in 3D as far as CAM, not so for 3D CAD, but it's not intended to be a CAD program. In a few hours I was able to create and run programs. I bought CB because it was recommended by a member here. I tested it (and a couple others), liked it, and liked the business model.....one time license fee of $150 that you own forever and no maintenance and ongoing cost. No regrets whatsoever about that. Great value. You'll also want to download one of the simulation programs to test your GCode programs. I use CAMOTICS. You can do all this without a CNC Router and you'll always want to simulate a program before you drive a machine with it, or you'll endlessly crash your machine and break bits. 3D CAD modeling opens a whole new world as far as design capability, but as mentioned, steeper learning curve. For complex projects, I spend more time at the keyboard designing, modeling, and programming than actually doing the foundry work these days, and as David points out, tickling a keyboard is easier than shoveling sand. For small projects, if it weren't for the drying time of the coating, I'd have castings same day. Best, Kelly
When I say "beginner", I mean this is all like a foreign language to me I like the idea of cambam being inexpensive, and the fact you use it to make all those great patterns means that's probably the one I will try. I had thought maybe V carve pro was what I would try, but maybe not. On the hardware side, I don't understand all the different equipment and how it coordinates with different software (or doesn't). Would it work to use a SMC5-5-n-n controller with 4 closed loop steppers? All the videos I've seen show how nicely it works, and it is reasonably priced. I'm confused, though, as they show the user plugging in a thumb drive to the usb port to feed the program into the machine. Can you plug a computer into the same usb to feed a design program from your own files, or from the web? Also, I think I heard that the SMC only works with metric...can't be set up with imperial measurements ????
Oh for crying out loud! Just discovered another controller that looks even more intriguing to me...the DDCS V4.1 This one will definitely connect to a computer via ethernet cable, and it has either metric or imperial programming. Lots of other specs sound good to me, but I'm guessing, I don't even know what I don't know
Thank you, guys! So far I have not tried anything at all on a computer. It's all been reading about CAD/CAM, and it just seems very intimidating to me. I've never learned my way around a computer properly, just self-taught for emailing, web surfing, shopping, using forums, and saving some files,. Never cut and paste, photo shop, or anything else that most computer savy people know how to do. Puts me at a definite disadvantage.
Your starting point definitely can size the degree of challenge. You'll need to learn CAD, CAM, some GCode, and, since you are building your own CNC router/controller, a little bit on how a machine controller translates GCode into motion, at least enough to calibrate your machine. That's a quite mouthful. Most people that can read and make drawings can pick up 2D CAD fairly easily. As I mentioned most entry level CAM programs have rudimentary CAD capability, so you can design parts and make programs to cut them while only needing to learn one software program. That software program will need to have a post processor compatible with your machine controller. The post processor is just more software within your CAM program that creates a GCode file that can be received and correctly read by your machine controller. Most CAM programs have a host of post processors to accommodate unless you have an obscure or very sophisticated/proprietary controller. I'd say, demo some of the CAD programs and see if you like one. Also download a simulation program. The sim program creates a 3D picture of the object that will be cut by the toolpaths created by you CAM program. You have to go through these two gates before you ever really cut anything so there isn't much reason to spend time and money beyond this before you can reliably do these steps. Then you can use your programs to actually drive a machine to cut a part. It helps a great deal if you have a knowledgeable person to help you when you get stuck. I spent days sometimes trying to figure out problems that even a casual user would have been able to identify and correct. Second best, but very helpful, are YouTube Tutorials and Web Forums for the softwares. As you consider a software, go to YT, search for tutorials and see how many are available (many are provide by the producer of the software), view them and see if they make sense to you. There are YT videos that discuss how you get started and make these decisions. Look them up. Also go to the web support forum for the various softwares, see how well traveled they are. If the last post is weeks or months old, that's not a good sign. If questions are asked and promptly being answered, that's a good sign. In the case of CamBam, the web forum isn't very busy these days, but the forum admin and the handful of moderators and forum faithful are very responsive and helpful. When I was looking, it came down to CB and MeshCam. As far as when you get to the machine controller, mine is GRBL based, not because I chose it, but moreover because that's what was used in the machine I bought. It's open source and well supported. It wouldn't be the choice of pros, but is in no way limiting for me. It takes some perseverance, but you'll get there. Best, Kelly
Wild Irish: I'm not sure if you know it or not, but the DDCS is a stand-alone controller, which means that you don't need a separate computer to run your CNC router. You just feed the DDCS the g-code file and it will handle the movement, without the need for a PC to run a software interpreter like Mach 4 or LinuxCNC. Which is a nice thing if you don't have an abundance of PC's laying around and want to do something else while the CNC router is doing its' thing. There is also no need for break-out boards between the PC and the stepper controllers, since the stepper controllers hook up directly to the DDCS. I know the DDCS will handle limit switches/homing switches, and an MPG for manually jogging the router. (That's a Manual Pulse Generator - it allows you to jog your router as fast or slow, forward or reverse, as far as you want. Or as far as your machine will allow - whichever comes first. In X, Y, or Z, you just tell the MPG which axis you want to move - it tells the DDCS and away you go.) DDCS V4.1 is probably an updated version of the controller that was developed over on the MadModder forum several years ago. The original manufacturer, and there have been several, went belly up several years ago leaving a LOT of people with hardware that didn't work properly - that was either at version 1.0 or 1.1 of the controller I think. There's a LONG thread over there regarding the re-writing of the original DDCS code to a stable, open-source, and usable, product. They then put the upgraded software out there for anybody to use. I've got a DDCS V3.1 sitting in a box with a pile of parts that will become my CNC router. (That box of parts has been laughing at me for a couple of years, waiting to be assembled. I'm retiring in a couple of months so its' laughing days are just about over.) As I understand it, there haven't been any incompatible hardware changes to the DDCS yet so I should be able to get the source code and upgrade my V3.1 to a V4.1. I scored a couple of MPG's off Ebay for a reasonable price - so they too went into the laughing box. Don
Coming from someone who has modified post processors and still working on getting a 5-axis up and going, Kelly's advice is gold. Looks like you are headed in the right direction. That standalone controller would get my vote. I don't know much about it but I do know that having an operating system as another layer of complexity can be quite annoying sometimes. Is having relatively infinite computing power and speed for the task nice? Yes, but, my cnc's average way below 200 IPM most days.
The standalone controllers on my 3D printer and laser seem to get the job done. The CNC has an older version of LinuxCNC. It came with built-in setups for the Gecko controller I have so that part was very easy. Being so old, I have to turn off the processes for update notification and manager when they start 'nagging' me...
Me too.. Going to put it on my go704 now that i have the pm940 running well... That's if I ever have time..
