Re: [CAD_CAM_EDM_DRO] Linear Slide Components

Lee Wenger wrote:(snip)

> Quite honestly, I don't see where all of the inaccuracy is coming from in a
> home-made (read non-comercial linear components) solution. As an example, I
> would refer to something like thk slides which must be screwed/bolted to a
> substrate surface every 6-8 inches. It seems to me that the stiffness of
> the substrate and the accuracy of the placement of those bolts will have an
> enormous effect on the accuracy of this system. Having never actually used
> thk or similar slides - am I wrong? Obviously a hand made solution has the
> same issues; however, my point is that the implementation of a given part
> looks to me to be at least as important as the part itself.

Lee, Your probably right about the bearings being good enough. It's been a long time that I saw a bearing from any manufacturer that when clamped and spun with a dial indicator on it, had major
runout. The real challenge here is the rail. That has to be as straight as possible or at least as straight as you need it. I know of nothing ready made, "off the shelf", that can meet any
satisfactory criteria that is not a real linear component. Quite frankly, even on cold rolled you can't necessarily rely on putting a bearing on each side of the material and expect it to run smoothly
from one end to the other. Even if you send the cold rolled for grinding, you can have issues related to stress making it crooked. Linear Rail has its good points, and in many cases it's just plain
necessary.

Let's imagine that you would purchase Linear Round Rail and stands, and your right, you have to bolt them to something straight to begin with. Then you could make little widgets to hold three or four
bearings in place to act as the "slide unit" along the rail. But, keep in mind you need a way to keep the rail super clean so there is nothing to keep the bearings from doing their job correctly. If
they are tight enough when all is clean, they will bind for sure when you get a little sawdust in between. Once dirt is in between, well, some accuracy is given up somewhere. Start counting bearings
and suddenly on a three axis machine you need 24 rail wipers per axis !!!

The only open bearing concept I have ever seen that can work is the "V" groove wheels and corresponding rail where the dirt gets scrubbed off by the bearing. But, don't let it stop you from trying.
Who knows, you may find something that works good for your application.

In regards the THK-IKO, well, I've held some of those in my hands where I assure you it will be the item you are going to bolt it to that's going to bend..... NOT the Rail !

When it comes to the smaller rails, you are absolutely correct that they will only be as straight as what they are mounted to. I know you've looked at my machine, and by now I'm sure you have
considered that my X axis rails accuracy is totally dictated by the cast aluminum Table top. Is it 100% Accurate ? Of course not. It is as accurate as the specifications on the material from the
vendor, but more importantly, every bit as accurate as it needs to be on the particular style of machine. Some have opted to go with THK in that location and that is fine. Yet, because the table top
dictates so much of the accuracy, The THK's may not be worth the extra money. You get the point.

Sometimes too, a particular design may have a "compromise" built in when using Linear Rails, especially a home hobbiest machine. Clearly, the "direction" the rails are mounted in, can have quite an
effect on "which way" an Axis is or is not truly straight.

Anyhow, I've racked my brain for about 3 years looking for the Ultimate non-linear rail "answer" for Linear Rail. I've never found it. Finding the fountain of youth might be easier ! There have been
numerous new products like those from IGUS using "Dry" slide techniques. (plastics) Problem is these are not inexpensive either.

Chris L