[greybeard]

My cnc design requires a small, open-centred moving table design, approximately 4 inches wide and 12 inches long.
:wee:
I've come up with this idea which may have been done before, but I've not seen it.
I hope the drawing is self explanatory, and it's open to simple variations.
Essentially I'm using a 4 inch length of 0.5 inch screw thread as a bolt to hold the sides of the trolley apart at each end instead of ends.
Each of the eight nuts has a threaded locking stud through one face, and carries a bearing, with a nylock nut holding the bearing in place.

The outer nuts/bearings are fixed in place, washered to allow the bearings to clear the table sides, and then the inner pair are screwed outwards against a spring washer. This allows a positive alignment of the bearings to be made against the shafts, at the same time gripping the table sides. When finally adjusted, the spring washers are locked in place with epoxy or superglue.
The outer nuts can then be removed in the future, if necessary.

Any comments or improvements please.

John

[Pat2000]

John I'm not sure that I'm 100% what your little asssembly is for but I think I know

It looks a little 'complex' and I'd reckon it could be 'weak' a hard knock would kill it?

How about the 'vee groove' bearing idea see

http://www.hepco.co.uk/db_pages/prod...=0006&cat=comp

for some ideas (don't look at the prices they are a little silly although the vee-groove bearings and 'single vee' rails are not such a bad price.)

I don't think you have much access to machine shop tools? but its pretty simple !

Pat

[greybeard]

Morning Pat.

Perhaps I should have posted more of a preamble to explain why my design has evolved.
You're quite right in guessing that I have no machine shop - just a drill press plus the usual wood working tools. But I also have fairly large scrap boxes plus generous amount of ingenuity bordering on eccentricity.(see other postings !)
My workpiece is a small block of wood 8 in x 1 in x 4 in deep , on edge. I will be "engraving" a profile on the 8 x 1 surface with the cut about 1/16 x 1/16. So the load on the assembly couldn't be much smaller.
I've performed the operation by hand for several years, with templates and a small router, and have built some partially successful equipment in the past, but my recent discovery of the zone has led me to look at complete automation of the process.

As I'm on a basic pension, this has to be very much a design based on what is in the scrap box, what ebay and car boots produce, and what I can do with it. Fortunately my electronics background can cope(when dusted off), so I should be ok there.

I hope to post a photo of the built version as soon as the remainder of the bearings arrive. Please keep the comments coming.

John

[Pat2000]

umm I appreciate your situation John (as much as I can grasp!), I did have a quick butchers at your 'fan making' on your site a while ago.

one point perhaps is - if the bearing assembly dosn't require adjustment then allow for a rigid or solid fixing if you can - even perhaps if its just the top bearings? I appreciate that adjustment is needed - of course, keep that for the bearings on the underside maybe?

I'm concerned about those nuts working loose and rotating!

hey - you're just lucky with the size of your axis travel ! and them tinky tiny cuts!!

I'm penny pinching this end also, but fortunate to be able to use the works machines and scrounge bits of material !! - in my own time LOL

ATB, have fun, pat

[greybeard]

Thanks Pat. I'm thinking in terms of using superglue after assembly to lock everything up. If it ever needs to be taken apart, a small amount of heat softens it ok.
I'm still playing with the idea though, so it may still finish up quite different.
What appeals to me is that most parts do more than one thing, plus the lack of machining.

[NEATman]

Greybeard-
First, very creative-

Any reason why you didn't put two bearings into one nut, on two of the adjacent faces? This would prevent the nuts from trying to loosen, as side loads are applied. It would still allow you to add another nut to the outside and mount the table.

If you were intending to preload the bearings by rotating one nut in relation to the other, a longer moment arm with something that could lock it in place would help.

Keith

[greybeard]

Hi Keith. The idea of using two nuts is to fix one pair, then rotate the other pair as adjustments to make the table 'square' to the shafts and remove slop.
I hadn't thought about the need to preload the bearings on such a lightweight assembly ( the whole thing only weighs 2.5 pounds including the stepper !).

[greybeard]

There's also a geometrical problem with the diameter of the shafts/size of nuts/size of bearings to allow for. With the two nuts I can use a wide range of sizes, and still the arrangement can be made to work.

[jans123]

Hi Greybeard!
Using nuts to get the bearings in angle is inded a little inovative. I like the simplicity. It must how ever be BIG nuts, but they can be found. You will how ever encounter some problems. Weakness in the rail is one problem to deal with, another is the problem to get the rails perfectly parallel. I have tried a similar system with bearings in a "V" (but I didn't think of using nuts. (that would have saved med lots of work...) I had only one single rail (a tube) bolted on top of the frame and made a carriage with four bearings (arranged in two "V" after each other) that could run on top. A pair of bearings (just vertical) run under the fram just under the possition of the tube. Those two bearings where preloaded with a spring, keeping the carriage above on the track. On the othe side of the frame I had a simple flat iron standing on the edge, a pair if vertical bearings above and another pair of bearings beneath (pre-loaded to keep down the upper set.) This solved the problems for me with the moving the table. The problem with Y and Z movements where similair to yours, but it wasn't easy to get everything perfectly parallel and it had a tendence to get "soft" rather quickly. I guess the bearings had to take a whole lot of forces they wasn't designed for.
/jan

[greybeard]

Re "big nuts" - you have to realise how small my set up is, and that my posting was aimed at anyone with a simillar sized machine. Obviously it could be scaled up, but if I start with 8mm shafts I only need nuts about 23/25mm across the flats to give the right geometry. And the shafts are only 150mm long. The inner frame of the table only moves 50mm in the y axis, with the outer frame moving 300m in the x axis.
The other bearings have just arrived, so I hope to find the time to build at least the inner frame, and post photos in the next few days.

[jans123]

I didn't notice how small you was making it. - sorry. Small bearings means small balls that cant take much load.I t isn't easy to get rails stiff enough if you go down to that size. Have you considered to use L or T- bars instead of a circular one?
/jan

[greybeard]

jan - please see #7, where I state that the total weight of the inner table + the workpiece weighs only 2.5 pounds. This is just over 1 kg. Where is the load you think will distort the 8mm shafts which are only 150 mm long ?
Perhaps I should also point out that this is a design that will only be cutting wood with a 1.5mm bit making a 1.5mm deep cut.(see #3)
I've been doing this by hand for about twenty years, holding the wood in my finger tips.
I don't think the effort required to move it past the router tip will be enough to bend a 8mm shaft by any amount that I can measure, do you ?

[greybeard]

Anyone following this idea might take note of a practical detail that I've just discovered.
All eight nuts from my bin are different thicknesses. :tired:
With hindsight this is hardly suprising, and with the adjustment in the form of a spring washer as a spacer, any variation can be taken up before epoxy is used to fix it in place.

However it's worth running the nuts in turn up against one fixed one, and marking which point comes to rest against a reference facet. (I hope that's understandable ?).
It also makes a difference with way round you start the nut, as they're not a whole number of threads thick. Given the thickness of each side of the 'table' is the same, you can back off the nut a set distance, and mark the best facet to drill for the bearing stud.