[BLOG] DIY CNC
I have been designing and drawing as long as I can remember. One thing I’ve lacked is the patience and skill needed to create them. Of all the things I’ve tried to create, I’ve failed 90% of the time. Usually the failures were do to a hard to make cut or simply trying to tackle that perfect curve of circle.
In a perfect world I would design it, a machinist would make it, and others would use it. Well I’ve yet to find a machinist that’ll live in my basement and work for food. So getting a machinist to make “it†is out.
So what is a creative mind to do? Build a CNC machine. With a CNC machine at my disposal, I’ll finally have the ability to mill out my designs with near perfect accuracy. We are talking precise milling to 01mm (0.00039 of an inch) and tighter.
Friends ask me what do I want a CNC machine for. To build a spaceship silly. But seriously, for a variety of things. Like building a custom water cooling blocks that will cover an entire mother board, mosfests and all. I’d like to take on the big boys in water cooling, like Swiftech and Zalman. I’ve been designing a true gluttony of metal, cooling, and style for a computer case named Damascus for over a year now. I want to touch Damascus. I’d like to make custom intake manifolds, lockable engine mounts, a motorcycle frame, and the list goes on and on. Some of these Items I plan to sell to a very niche market. Others will be just to scratch that itch.
But enough of the why’s and what’s. How is the big question. I’ve spent several months reading about and designing a cnc machine. One that is simple, inexpensive, and accurate. One that you could make with parts from your local hardware store. But I’d like this machine to be able to stand up to any $2500.00+ machine for a 3rd or 4th of the cost.
I’ll be documenting my successes and failures in this blog. I’ll be using Google sketchup for the designs and posting files for those who want to give it a go. When I can, I’ll video the building process and put it up on Youtube. I’ll cover things like scribing metal with digital calipers, center punching holes, possibly using a cut off saw accurately, stepper motors, servos, controllers and software.
I’d like to be done within 3 months but am prepared to go at it a lot longer. I hope to finalize and post my designs this weekend and order materials next Friday.
Now, I just need to figure how to post pictures on this blog.
In a perfect world I would design it, a machinist would make it, and others would use it. Well I’ve yet to find a machinist that’ll live in my basement and work for food. So getting a machinist to make “it†is out.
So what is a creative mind to do? Build a CNC machine. With a CNC machine at my disposal, I’ll finally have the ability to mill out my designs with near perfect accuracy. We are talking precise milling to 01mm (0.00039 of an inch) and tighter.
Friends ask me what do I want a CNC machine for. To build a spaceship silly. But seriously, for a variety of things. Like building a custom water cooling blocks that will cover an entire mother board, mosfests and all. I’d like to take on the big boys in water cooling, like Swiftech and Zalman. I’ve been designing a true gluttony of metal, cooling, and style for a computer case named Damascus for over a year now. I want to touch Damascus. I’d like to make custom intake manifolds, lockable engine mounts, a motorcycle frame, and the list goes on and on. Some of these Items I plan to sell to a very niche market. Others will be just to scratch that itch.
But enough of the why’s and what’s. How is the big question. I’ve spent several months reading about and designing a cnc machine. One that is simple, inexpensive, and accurate. One that you could make with parts from your local hardware store. But I’d like this machine to be able to stand up to any $2500.00+ machine for a 3rd or 4th of the cost.
I’ll be documenting my successes and failures in this blog. I’ll be using Google sketchup for the designs and posting files for those who want to give it a go. When I can, I’ll video the building process and put it up on Youtube. I’ll cover things like scribing metal with digital calipers, center punching holes, possibly using a cut off saw accurately, stepper motors, servos, controllers and software.
I’d like to be done within 3 months but am prepared to go at it a lot longer. I hope to finalize and post my designs this weekend and order materials next Friday.
Now, I just need to figure how to post pictures on this blog.
0
Comments
But that would be 6 months away at least.
In six months you could have your own CNC machine... Join the cnc revolution man.
I'm actually really really interested in seeing how your CNC comes out. I'd love to make my own version of it. I've got lots of ideas for things to make.
I did look up a itb on a 6m. It looked like a tight squeeze. At least the one I saw. Have you thought about doing a dual plane manifold? Split the intake to two plenums and then 2 throttle bodies? For tuning and low to mid range power they usually are ideal. When you go completely itb, high rpm tuning can be tricky. All that air going bouncey bouncey.
I am not sure what you have been reading but I think you may be in for a shock. The cheapest way to get into a cnc mill is to retrofit an existing mill with cnc controls. I do not think you can build one that will actually cut metal with any precision for $2500.00 let alone less. I am a machinist and do have a CNC mill in my basement. I've got about $12K into mine and it is not big enough to do a lot of the stuff I would like. I am not trying to discourage you from starting your project but want you to know that your expectations are high. My favorite CNC Forum is CNCZone.com , they have sub forums for everything ! including DIY and "Hardware store" builds. Most of these are router based and are no good for metal cutting. If you really need .0004" accuracy in metal $10K is not enough to get you there. If you only need.001"-.002" then $5K might do it. Check out CNC zone it is really a great website for the machinist .
I hope I didn't pop any bubbles , just want you to know what your getting into.
respectfully
Scott
.001 would be great but I can live with .01. At least on my first build.
If you get a chance, check out the second page of my design thread on cnczone. I'm currently working on reinforcements and finalizing the design. It has changed a little since then but it is roughly the same. Would love your input.
http://www.cnczone.com/forums/showthread.php?p=535689&page=2 I'm FandZ there.
Just out of curiosity, would you be up for milling out a tool out of a couple of pieces of 4x1/4 steel? The plan is to make a tool to hold a center punch squarely and to have a few holes as a guide for my drill punch. My desighn calls for a ton of drilling and I'm just trying to come up with a way to get everything as accuarate and as speedy as possible. I gladly pay for your services if I can afford them.
Thanks for posting. I hope as I'm going along you can give me a little guidance if you see something not coming out right.
Chris
I read your threads at the zone and have some questions and comments.
I am assuming that the different colored stock in your drawings are to indicate aluminum/white-steel/black. As in, your bearings are riding on steel and not aluminum. Is this precision ground steel or just cold rolled plate ? If it is just CR plate the mill finish will preclude you from any precision when trying to adjust for a slop free fit. Your design looks well thought out , but I do not think it can be much more than a router for cutting wood and plastic. Removing metal ( even aluminum) is a whole different story. What spindle are you planning on ? A router is way to fast for metal, if you use a speed control you loose all your torque.
I really think your bearing surface needs to be precision ground hardened steel. You should look into some 2nd hand machinery shops like HGR and try to find some used linear bearings like Thompson makes. Hgr is about an hour from here and I visit a couple time a year. The last time I was there they had a shelf full of different kinds of linear bearings and shafts.( and ballscrews too !) If you could get a good deal on them it would greatly simplify and strengthen your build. It may end up costing a little more but would be soooo worth it.
I stay pretty much in theTormach Forum at the zone , my name there is Scott_M.
Do you have a .dxf or .stl file of the plate you need made ?
Scott
Not yet, I’m still working out spacing and bolting. I’ll get one up as I get the design finished though : ) .
I was planning on using cold rolled steel. There have been some documented success using it like in this design http://www.glacialwanderer.com/hobbyrobotics/?p=17. I’ve thought about using aluminum, since it is more leveled throughout, but I worry about wear between the bearings and the aluminum. Ground steel would be ideal but the price is a little daunting. No point in spending 300+ on rails when I could go with a superior linear motion for a ton less like you are suggesting.
I’m going to give the cold rolled a go and see how it goes. I may lay down some light paint on the cold rolled steel and then sand it with a flat metal block to see any higher ridges and then hit it with a grinder to get it more uniform.
I plan on using a Hitachi router at first and then building a custom spindle later on to get lower speeds.
It seems to me, the best hobby cnc machines are simply very ambitious engravers, because they are not cutting metal, more like precisely scraping it. A job that’ll take one of your machines 5 minutes to do will probably take a hobby cnc machine 5 hours. But I have seen some good work done on a hobby cnc machine. I'm hopeful.