Yet another cheap CNC build

[grumpygeek]

I have been playing with hobby CNC stuff for almost 12 years now and though I know most people have moved on to cnc bearing and rails from offshore at the cheap end I thought I would share yet another cheap diy bearing idea I had.

I am in the process of build a larger (42" x 54" cutting area) machine to replace a smaller (22" x 44") machine I built about 7 years ago, the motors, screws, and controller and even the table from the old machine are being used in the new one (only had to order 1 new lead screw) so the actual cost is fairly modest. The reason for the new machine is I am actually making a bit of money cutting HPDE parts for a friends business and need a machine that can cut 36" x 36" for a new part and the fact that I had a new idea using 608 bearings and steel angle for linear motion I wanted to test. I was actually playing around with a printed version before Covid hit but decided to use aluminum angle because I have a tendency to overtighten things and the plastic doesn't really like that. I am just going to throw up a few photos and update it as I can.

Similar Threads:

• Newbie- First Design Build (doing it on the cheap!)
• Build Thread- X2 CNC Build the cheap way
• Need Help!- 2'x8' or 4'x8' as a first, cheap build
• Build a cheap, SIMPLE, CnC foam cutter.

[grumpygeek]

So my friend really needed me to get that part that started this whole project done, so I made I push this week to get all of the get all of the lead screws connected, router mounted, and reconfigure LinuxCNC for the new machine. I wanted to get the hold down table and spoil board on and leveled but didn't have the time so with minimal testing I just put some Styrofoam under the 4x4 1/2 ABS and went for it. Cut 2 parts 1st one was a 30" disc that is used to stiffen the lid of a 36" round ABS barrel so a pump can be mounted on it, the other was a 35" grate that goes in the barrel to support filter media for industrial water treatment. The grate took about 2 hours 15 minutes to cut using 1/8" spiral bit @60 ipm with 1/8" passes, the bits I am using are really cheap ($1.50) and think it is pretty much spent. The only real issue I think have is that when the carriage is in the center of the X axis I can lift up at the router collet and get about an 1/16 to an 1/8 inch of movement which I think is the board of the X axis flexing/twisting, it is rock solid near the sides. I am going to try adding 5 to 7 pieces of aluminum angle or other material front to back to stiffen up the board and see if that helps (any suggestions are welcome), if not I guess I will redesign the gantry.

The bearings themselves worked well, when adjusted so that all bearings are touching and do not slip when held while trying to move them, they are a little tight but I can still move them with just a little pressure and can move the gantry about 1/2 the distance with a modest push, I have no reservations about using them in any orientation once they are locked in. Next steps will be to stiffen the gantry and put the hold down table and spoil board on.

[routalot]

It may not be the last word in sophisticated machinery,but you got the job done and that puts you in a different league to many home builders.It looks like the machine is sufficiently complete that you could use it to cut additional parts for the next upgrade and that is a very useful position to be in.It looks like a LinuxCNC axis display on the monitor and that also works well on my homebrew machine.I hope this thread will encourage other builders to go ahead with building and using machines that aren't just a collection of the parts that are regarded as the ultimate, as ingenuity is a vital part of the homebuilder's repertoire.

[peteeng]

Hi Grumpy - by the time you buy all those bearings screws and extras you could buy supported round rail and have a much better machine.. Peter

[grumpygeek]

Thanks routalot, I agree. When I first started things were a lot more expensive and there was a lot more emphasis on finding cheaper ways of doing things, there were certain things you had to spend money on like lead screws, motors and controllers but everything else was up for grabs in an effort to build something that a hobbyist could afford. I now also tend to stick with designs that you can build with a table saw and drill press as a starting point so anyone can built it. I think if I can stiffen up the gantry it will suite my needs perfectly. I set it up so the router actually extends past the table by about 3 inches so I could experiment with a cheap rotary axis, looking forward to trying that.

[grumpygeek]

Hi Grumpy - by the time you buy all those bearings screws and extras you could buy supported round rail and have a much better machine.. Peter

I am going to disagree a little bit, the steel angle for the rails and parts for the bearings including the screws and nuts work out to about $80, I think a set of cheap SBR16 or 20 rails and bearings would probably be around $200 and I am not sure about the quality. I do know that the only thing I might have to replace in my cheap bearings is a 25 cent 608 bearing. Would it be worth it to spend the $200 on the SBR rails and bearings, most likely given the amount of time I spent coming up with the design of the bearings. I don't think just that would make this machine all that much better. I have an issue with flexing of the gantry that I need to addressed, and I don't think the rails would fix that. This is really just me experimenting with trying to find a very simple/cheap design that works and anyone could build... maybe there is a better way, maybe I should just give up and move on.

[routalot]

Developing a design that anybody can build is a fairly noble task but building a machine that does useful work is nothing to be sneezed at.It would be quite a generous gesture to carry out the improvements you think will help and to post results where people could learn more about what did or didn't work.The type of machine that would make clock parts wouldn't be needed if you were making dog kennel kits and the spectrum of projects can accommodate a wide range of machine capabilities.If we are honest about what we actually need and perhaps not focused on assembling a collection of the "ultimate" parts there might be more machines working and more projects completed.

[jckstrthmghty]

If you are building a machine for the experience and knowledge then time well spent. However if you are a building a machine for more than casual use then the rail's price is inconsequential. There is nothing wrong with 608 bearings however the hardware store rail itself is not anywhere near even the cheapest of SBR in any aspect. If you use the machine enough to wear out a bearing then you've gotten tremendous value. All those projects with better tolerances, and it would be many. I'd even say spend a bit more and get profiled rail. 12mm rail would be suffice in sbr or the profiled stuff.

I don't want to discourage you or anyone from trying but I've been down that path and if I had to do it again I would just start with the profiled rail.

[grumpygeek]

I don't want to discourage you or anyone from trying but I've been down that path and if I had to do it again I would just start with the profiled rail.

If you look at my website you can see I have been down this path a few times Just a glutton for punishment. I think if I had I gotten to this bearing design sooner I might have moved on by now.
That being said stuff like this https://www.ebay.com/itm/31326624603...YAAOSw8zpfjrVM seem like an incredible deal and maybe a great starting place but I would not know until I ordered it. I had a friend order 'cheap' rails ($800 for a 4x8 machine) about 7 years ago and after about 3 months of moderate use the bearings had worn grooves into the SBR round rail so I guess I am a little leery.

My other thought is for hobby woodworking and sign making my machine should be fine, if I were running a business were time is money I would probably just buy a kit machine, find something used, or just go to probotics, cammaster, or shopbot.

[grumpygeek]

Got around to actually measuring and trying to address the flexing of the gantry today. Verified it is the twisting of the plywood that is allowing the router to be moved up to 45 thousands when in the center position (28 inches) and about 15 thousands of 3 inches from either side with moderate pressure applied.

After installing 7 pieces of 3/4 aluminum angle along the underside of the gantry, I am getting about 15 thousands of movement in the center and 5 thousands at the sides, so about 1/3 of the previous values. Still not thrilled, but happy enough that I am going to do a few projects before deciding if I need to take further action.

[routalot]

I think I'd be looking at doubling up on the clamping blocks that hold the router to the gantry before too long.That single,comparatively narrow piece is having to resist a lot of force.Even so,0.015" isn't a bad level of accuracy for most woodworking projects.

[grumpygeek]

I think I'd be looking at doubling up on the clamping blocks that hold the router to the gantry before too long.

Agreed, that was the plan, after I get the spoil boards on and leveled I should be able to move it up a bit. I may need to recut the Z axis face plate to be a bit longer, the current version worked in my head but not quite in the execution. I guess I could move it up and use a shank extension to do the leveling with. Just one of a half dozen or so things I didn't get quite right.