[sethyrish]

I’ve been trying to design a bench top mill using a steel beam and column as I have a little welding experience and steel is relatively cheap. Also I read somewhere on this forum that many mills with linear bearings are overbuilt as the larger bearings are what’s available cheap on the secondhand market so why not! However there doesn’t seem to be much available where I am so I thought I’d design a monorail version to try and cut costs. I’m just not entirely sure one lineal bearing on each axis is up to the job. I’ll be doing mostly small work which I can centre over the bearing and the largest work will be engraving circuit boards and electronic face plates which wouldn’t place a lot of stress on the edges of the mill table (500 x 200mm).
Can anyone tell me whether this design is ok?
Any help appreciated. PS sorry for the image. I cant seem to get it sharp.
The bearings specs are:
Abba BRH25A
Lmax 4000
G 1950
Dynamic(C) 1950
Static(CO) 3200
Mx 36.8
My 22.8
Mz 22.8
Block(kgf) 3.6

[BillTodd]

Can anyone tell me whether this design is ok?

Sorry, but there will far too much 'rocking' from a single bearing block used like that

If you look at the static moment ratings (i.e. for twisting forces) for your linear bearings you will see they are a fraction (<10%) of the load ratings. They are simply not design for this use.

[sethyrish]

Thanks Bill.
I'll have to re-design with 2 rails per axis.

[cadfish]

I might put this out for you to consider, Structural Steel is hot rolled, thus the tolerance is within 1/16". How accurate do you want to be? Do you want a mill or a router?

[sethyrish]

Maybe I should get the mounting surfaces milled? I'd like it to be as accurate as possible as I'll be drilling with 0.5mm drill bits. I've a minimum of tools so may have to pay to get some things done. The idea was for a multi-purpose machine capable of PCB's as well as milling small items in brass. I know the water cooled spindles are mostly used for routers but are they a good choice for a mill?

[cadfish]

A little about myself: I have been working as a structural steel detailer, the guy who takes engineers drawings and makes new drawings so the people in our fab shop can make it, so I have to come up with a lot of these ideas to make stuff fit together and then have the engineer verify the design.

Here is one of those ideas, meaning that you will want to add another rail on each axis, maybe weld an angle on the outside of the flanges that way the rails can sit on the angles and you will also greatly reduce the twisting. You could then move your screw(s) to where your picture has your rails. And with the angles you could hold the flatness better and then use a drill press(?) to drill the holes in a straight line. Dang, its all coming together in my head, now I want one.

Water cooled spindles I have no idea, I am trying to design my own industrial quality benchtop size mill for some time now, 2 years, and would like a hydrostatic spindle and that's about as far as I got on it.

[sethyrish]

Thanks for the input cadfish.
Are you saying that the web thickness of the beam could be out by 1/16 or that it will flex up to 1/16. Won't the hot rolled angle also be out by 1/16 or would you use a different grade of steel. What size angle would be sufficient? Hope you don't mind the questions. I'd like it to work when I finish it!

[cadfish]

The web thickness could vary as much as a 1/16". I think I would suggest buying cold rolled angle, cold rolled does not have to cool down after it is shaped while with hot rolled one side can contract faster than the other side depending on the process the mill uses. I am pretty sure that if you were to take an angle and weld it to the flange, using stich welds to reduce warp and comparing the angle to a long straight edge as you are welding you could probably be within .004" + or -. Then after you are done welding get a friend with a mill to drill the holes for the rail or a drill press might work.
As for the size of angle, I would need information such as how wide the rails are, how large is the load of the x-axis and table w/ work pieces. with these facts I might be able to determine a size and thickness.

[Nzoldun]

Your 'open' section for the column will twist quite a bit when the cutting forces come in to play, especially since your cutting head looks to be a good way from the vertical centre-line of the column. This may give rise to considerable chatter and even break cutters. The commercial mills of similar size are box sections, to provide as much rigidity as paracticable

[sethyrish]

Ok Nzoldun, I'll change it to box, or maybe box with u channel welded to it. I like the idea of the piano accordion cover over the ball screw and rail. There are also welded beams which may be better than the universal. Anyway it looks as though the project is going to be more complicated than originally thought. Thanks for your help.

[red_glass]

Hi sethyrish

I am also part of the way through fabricating a mill so I understand just how complicated it can get

With the choice of steel profile etc Nzoldun is giving sound advice. What I would add is that you should think about where the forces are being applied and how they are being resisted, tension compression etc. (also rember lever and fulcrum) The deeper the structural element the greater the "structural depth" and the better it will be able to resist stresses praced upon it. I personally feel that it is worth the additional money to over engineer everything as far as your badget allows.
Another thing to consider is the resonence of the structural system you end up building, because vibration caused by the act of machining etc can cause the entire machine to vibrate causing chatter etc. This is one of the reasons why cast iron is one of the preferred materials to make mills from, and it is also one of the reasons why you will also find polymer concretes being used.

A really neat idea I saw was to use steel profiles to fabricate the mill and then cast the polymer concrete into the hollow interiour cavities where possible. This serves to not only vastly improve structural strength of the mill but also serves to improve the resonant characteristics while being comparitivly very cheap! The down side though is that it makes the mill very heavy.
regards Red

ps For what it is worth the other thing about cold rolled as opposed to hot rolled is that the cold rolled is work hardened by the rolling process where as the hot rolled isnt.

[RICHARD ZASTROW]

And cold rolled has mechanical stress "rolled" into it. This causes the cold rolled material to distort when machined or welded. Just another complication. LOL

Dick Z

[sethyrish]

Thanks for the advice red and Richard. I've decided to go the X2 route for now. I’ll either get a CNC conversion kit or use the X2 to manually make the fitting to convert it. Due to lack of funds and lack of secondhand linear rails in Australia. Postage from America just about doubles the price. I also liked the look of the epoxy and have been mucking around with some designs. Maybe epoxy granite will be project # 2.

[te-raptor]

Hi sethyrish,
I have a water cooled spindle, and are very pleased with it in my router. But I'd think it has much too high an rpm for use in a mill. I'd say about 6000 rpm would be the lowest useable speed on it. However, if you gear it down with some form of drive train you'd have an exellent motor with almost no noise to it, capable of running for hours without any heat buildt up. I would surely consider it if building a mill.