linear slide idea

[noid]

I'm new to the site but not new to cnc. I'm working on my own design for a cnc router based on existing mass produced machine ideas. I have noticed that the standard slide systems ie. rails (thk,round bars etc.) are expensive and hard to modify if need be. My design for the rails should be cheaper and easy to adapt to any length for a reasonable price. This is still in the design stage but my research tells me that I can manufacture my design cheaper.
I'm planning to use a precision ground shaft machined with a flat running the entire length and a suitable square bar drilled to match.
My question is has anyone tried this idea and has it worked?

[DDM]

so your idea is basically a supported thompson rail with and open pillow block correct? it would probably work but percision ground shaft isn't the cheapest to begin with and you'd probably want to harden it someway if it isn't already. If you could elaborate I'd be more than happy to help.

Carl

[noid]

Yes a supported thompson rail with the open blocks. I find a that in canada with the shipping and such the rails themselves are the expensive touch. Precision ground shafting is about 2.50/ foot this means of course a 10' bar costs 25.00 heat treating is about 1.00/pound so I can produce 2 5' rails for about 32.50. Now this is only a ballpark price so far but it blows the 125.00 usd/36inch length figure out of the water.

[jcc3inc]

Dear Sir:

When I built my 3 axis table I too worked toward an economical approach. I used cold drawn flats for the ways and used standard bearings with eccentrics on the carriages. Cold drawn flats are dimensionally quite good and the cost is right. If you would email me, I will sent back photos of my ways etc. My drive is rack and pinion.

Regards,
Jack C. jccinc at owc.net

[HuFlungDung]

Originally Posted by noid Yes a supported thompson rail with the open blocks. I find a that in canada with the shipping and such the rails themselves are the expensive touch. Precision ground shafting is about 2.50/ foot this means of course a 10' bar costs 25.00 heat treating is about 1.00/pound so I can produce 2 5' rails for about 32.50. Now this is only a ballpark price so far but it blows the 125.00 usd/36inch length figure out of the water.

I'll be interested to hear how the shafts turn out after hardening. Doubtlessly, they will use induction hardening methods, but I wonder how much warpage will occur, because they have to use a water quench (I believe) in order to harden precision shafting (typically grade C1045).

[ESjaavik]

With precision hardened bar I believe the method is to put in hardening first, then precision.

[Graham S]

try ebay

[DDM]

maybe the key is to find some sort of happy medium in material. I'm not much in to different metals and alloys (wish I was, would appreciate any links) but couldn't you use some kind of chrome plated rod that was machinable, do your flat and drill your holes, countersink for the screw head and go right into the base of the rail through your square stock and do away with the heat treating. This wouldn't be as good as a thompson supported rail but it wouldn't come with the price tag either.

The other thing is don't worry about using a hardened material and incorperate a lubrication system into the pillow blocks. It wouldn't be that hard if you had the time to work it out.

The other question I have is where you can get square or rectangular solid stock that has pretty tight tolerances and is out of a good quality alloy that would be sutible for a linear rail for a bushing to ride on?

Carl

[Graham S]

I think it all depends on what you are trying to do, for a lot of hobby machines they don't see that much work and if a rail is really cheap and it wears then you replace it, same with nuts etc. It is a different way of thinking about it expensize linear rail lifetimes are rated in 10's of Km but half the time we don't need that. Main thing I suppose is to have something that is flat/straight enough.

If you look for example at CNC plasma cutter kits they use cold rolled bar as the rail, why? Because it the accuracy matches the cutting process and when they wear you replace them for peanuts (not with peanuts, that doesn't work).

So in this case if the blocks were delrin the need for hardening would seem to be less and again they could be replaced if needs be.

Why not make a test pair for the z-axis

[BigDaddyG]

Some alturnatives to the standard Thompson type rails:

You may want to look into Igus Dry lin rail systems, far cheaper than the thompson and actually superior in quality. I have been using them now for years, and have many in production as well as some home shop use. (www.igus.com if your interested, I have written about these many times)

Another one, is Pacific Bearing. They make a square bearing and rail that also works great. I have many machines out there still in production after many years.

If I had to suggest one over the other, I would go with Igus. They actually fasten right to Aluminum Extrusion, have tremendous loads and capacity's and are very cheap. I have been using the DryLin W Guide System on a series of Custom Lathes (3-axis cnc with 2" Dia. capacity) and these babys work great. I hope this helps.
Good Luck,
Glen

[Moondog]

Hey BigDaddy... I am just setting up my Igus system for my build. I would like to talk to you about how you set up the paralel and tolerances etc if you wouldn't mind...

you can email me at frans@cfbs.com.au

cheers

[Graham S]

Glen, have you ever posted any info on these lathes, they sound sweet!