Hello from Idaho
 

I just recently started gaining interest in the possibility of designing my own general CNC system. I was thinking mistakenly that it would be something that relatively few people would take on. But WOW! The CNC building community and the flattening power of the internet is really empowering people with information! I am really excited to learn as much as I can about general CNCing (milling, plasma, Oxy-acetylene, MIG, TIG...).

Pipe Coping with a plasma cutter
 

Does anyone have designs for an apparatus for cutting/coping pipe with a plasma cutter?

I could make one myself, but I would guess that someone has already made one and I would like to see anything you all got as far as dxfs pdfs,or other general CAD format drawings. It would help me to brainstorm how to do it with my table that I am designing right now. Thanks in advance.

Welcome Jared,
We focus primarily on the Mechmate for wood routing here. For more info on the milling, plasma etc you can also try www.cnczone.com. The Mechmate makes a really good wood router, and some people have also cut some light aluminum with it as well. :)

MechMate CNC Finite Element Analysis Possibility
 

Hello everyone. This is a great resource you all got going. Recently, I became interested in trying to design my own CNC. At the time I had no idea that the "hobby" of CNC making (perhaps businesses for a lot of you) had grown so large a following.

I was reading a post by Gerald G. I am under the understanding that he is the fellow that got all the MechMate plans out there. If my understanding is correct (this tells you how new I am to this site and the information contained herein) I would like to thank him. Probably he has already gotten a lot of thanks from people and so I will move on. As I comment please note that I am writing as much to learn as I am to teach. I have much more to learn then teach so if I say something that might not be correct feel free to correct me.

I was reading a post by Gerald and recognized the fact that it is very difficult to compare the MECHMATE CNC with other CNCs. In it I remember some posters mentioned that the cost (for expensive 50grand on up machinery) is actually just a measure of how much efficiency, accuracy and precision you think you are getting (or it should be). Sometimes, the more the money doesn't equate to any of the three attributes mentioned above. So what attributes do we pay attention to and why? What affects efficiency, accuracy and precision?

What affects efficiency is largely a combination of the software and the loading unloading system which the manufacturer makes. A lot of the loading efficiency boils down to the design of the table. That said it seems like for the most part (until you get into the 200 grand range) there are only a couple different general designs (referring to a Static Gantry design, versus a Mobile Gantry design like the MM).

Accuracy is a non-issue when speaking about CNCs in my opinion (IMO).

The only other issue then is precision. I am coming at this from a Mechanical Engineers perspective so bear with me (I am graduating from a BSME program in December '09). The precision can be lost in two primary places; one is in the structure of the table itself flexing and the other is in the flexing of the tools or gantry/motion system and associated parts....or the floor might be giving, but hopefully that is not your problem ;). Also, what I say in the following paragraphs should be taken with an assumption of a straight aligned movement system.

The flex in the gantry and table /motion system can be modeled and shown visually via FEA. However most people don't have such software available to them. But seeing as this is an "Open Source Application" and also seeing as I am a student and have access to NX 5 and ANSYS (expensive finite element analysis software), perhaps some of you might have an importable file that I can upload and then run it through ANSYS or NX 5 FEA solver to generate a visual representation of stresses/deflections of mechanical components under simplified load conditions. I could then export the drawings to PDFs or whatever and make them accessible to everyone. With the drawings I would need to know exactly or generally what everything is made out of, what sort of operation you are doing on what material, feed rate and speed, etc. to come up with a decent FEA model for you all. Additionally, if you all were able to give me multiple assemblies of various sizes of MM CNCs then I could do all of them (assuming I have enough time).

With that information you not only would have an idea of deflections in the X, Y and Z directions that happen under a particular load condition, but you can have a general estimate of what the tolerances would be given a certain deflection rate. Thus, the precission could be roughly understood quantitatively.

That said, if a person doesn't know what tools to use, what rpms to run the spindle, or how many ipm to set the travel for, (or perhaps doesn't know the touch-off distance/Wattage/Voltage setting for a plasma cutter), or uses deficient tooling, you are going to get bad cuts no matter what the machine is capable of giving you.

That's my 2 cents. I assume there might be a place to download the files that I mentioned above. If someone could point me in the direction I would appreciate it.

PS, I have some slight experience, but mostly I am a bookworm. <----Not totally a bookworm, but just realize that I could miss the boat a lot of the time because I haven't spent a huge amount of time in the shop. Oh yeah, and I should explain too (most of you realize this I am sure, but) the FEA images can be used to improve the design, by finding weak points / high deflection areas and allowing us to do some corrective design work to strengthen the general design in that area.

One of the awesome advantages of making a homemade CNC is the fact that you don't have to "SETTLE"! If you see something that can be better, you can make it better for everyone. And not only that, but the CNC isn't "cost engineered" to have parts that are just the right size or a little bigger. They can be overdesigned to last a lifetime. AWESOME STUFF!

PSS: Sorry for the book I wrote up there! I just get excited (ADD and all you know).

Welcome aboard Jered, I enjoyed reading your book !
I’m no engineer, actually a studio furniture maker...but will assume your input & contribution will be a nice addition to the diverse group we are.
I just can’t wait to see some of those flex studies / results !

Good luck with your future built, Robert ;)

Welcome aboard! It'll be great to hear your perspectives.

You are correct. Gerald is the guy that designed the Mechmate and he's also a mechanical engineer with a few decades of experience under his belt. I'm betting he will welcome some help from you in answering the millions of questions he gets about substituting different sized channel (or whatever) for what is called for in the plans.

You can find the plans in the download section near the bottom of the forum page.

Welcome aboard.

Jared,

This is a great place to learn about CNC. I actually took a couple of CNC classes at a local vo-tech school and had been reading about it online for several years but it wasn't until I found this site and actually built a Mechmate that I started to understand what was happening.

Gerald has done an outstanding job of setting this site up so that people from all over the globe can build these machines using materials and parts that they can obtain more or less locally. Before I found this site, I noticed that the other forums expected you to use their parts or nothing. I wasn't able to figure out what function those parts performed so I couldn't fully understand how their machines functioned. Gerald doesn't require that you order a particular part that has some weird name and an unknown function. He (and the other members of this group) help everyone understand how the parts fit together so you can substitute if you are willing to study and take chances.

Like Heath said, we're dedicated to routing wood rather than milling or plasma cutting so I wouldn't advise trying to start an indepth discussion on this forum for how you could convert your vertical mill to CNC. However, that process would accomplished fairly simply using the knowledge you'll gain here.

Doug and Heath,

Thanks for the welcome. I had been doing some CAD to get my own design up and running, but I am now really looking at making a MechMate. There is so much support and the drawings are very well organized and done. That said, I will probably look at how I can change it to make it a Multi-CAM CNC that can deal with some thicker metal.

Differences between a woodworking MM and a medium-heavy steele working MM
 

So I was told that the MechMate is primarily meant for wood cutting and some light aluminum machining. I was wondering what makes a machine better suited to wood cutting verses say, milling some 304 SS stock or some similar harder metal. I have an idea of the places that I would need to change the table structure and the gantry in order to accomodate heavier metal work, but I would like to here what some of you think. What would need to be changed in order to make the MechMate a heavy duty steele working CNC? I assume just beefing up the rigidity in general and increasing the step motor size that is used. Is that correct? How much more to it is there?

Are you talking about 2 1/2 dimension cutting of 304 SS thin sheet metal or full 3D milling of 4 inch thick 304 SS bar stock. If you are talking about sheet metal, I would simply make a plasma cutter setup. If you are talking about 3D milling of thick bar stock, it isn't really practical to beef this machine up. It would be much cheaper to convert a Chinese vertical mill to CNC. Besides increasing the thickness of the gantry and rails, you would need a spindle rather than a router and also ballscrews rather than stepper motors.

I'm not a mechanical engineer but I have my own Bridgeport mill so I have some practical experience with clamping and cutting forces and the effect mass has on the quality of the cuts you're making.

Jared, I have combined your multiple threads into this one. I am not supporting various discussions on beefing up a MechMate to work metals instead of wood.

Great thanks Gerald (For joining my posts). I was thinking about it over the weekend and I came to the realization (partly due to other comments) and partly due to just the realization that a mill for metal is really a totally different beast, but perhaps I can combine MIG, Plasma and its outstanding routing capabilities to do 90% of everything that I need done done. Trying to mill metal it seems would be overkill. I was just too naive to see it right off.

Thanks everyone for your input and patience!