[Guldberg]

So it begins...

Im nearing the completion of my cnc router build from scratch from lasercut steel, waterjetcut aluminium, linear bearings and a lot of homemade parts.
I promised myself that i wouldn't undertake any new major projects until it is completly finished, but that doesnt hold me back from designing now.

Here is what we are aiming for:

Slant bed - to allow for cooleant drainage etc. and because it looks cool
Complete enclosed - To allow for heavy flood cooling and to make it look more "pro"
125mm chuck - with possibility to add a 160mm later on
400mm of z travel - properly no tailstock anyway
150-200mm of x travel - Sufficient for that size chuck i guess
Tool turret with 8 positions, room for 4 cutting tools and 4 drills - Will properly be a project in itself later on

And the way to get there:
220Ncm stepper on z - I may change the design to accommodate a 650Ncm
180Ncm stepper on x - Maybe 220Ncm as well
Double nut or zero backlash ballscrews on both axis
Build from scratch of 20mm steel plates - Giving it a weight of +140kg, should provide a very ridig construction
Hiwin linear profiles, 20mm square type for z
20mm flat linear profiles for x - Bought of ebay some time ago

Now for som design pictures. I really hope to get som inputs on the design before i go on with the buildning:







Based on my experience from building the router, im going to build it out for MDF to begin with, there is just to many unknown factors you dont catch in the designing process, hence the idea of doing prototypes.

Thats the beginning, all kinds of input and questions are welcome

[mcwilsonmfg]

wow! sounds like a serious undertaking...I will be watching you...If I see anything where I can help out...would be happy to....

[LeeWay]

I don't know a lot about a slant bed lathe. Not seen many like that. Not seen all that many period. It does look like a great project. I like the amount of steel your putting in it as well. I just have to say it looks unbalanced.
It appears like your turret won't even be on the rails when cutting stuff in the spindle. I like to design the load fully on both rails and use the far ends of travel only for larger parts and extreme cases. For the most part, simply centered where the tool will have the most solid base.
Why not build the whole thing like you have done for the spindle? Then you could spread the lower rail out and down so that it sits under the turret while cutting.
It would seem like a cleaner design without the notched angle of the bed.
Like I said, I don't know that much about these, so take that with a grain of salt. I will follow your thread anyway. Good luck with this as well.

[Guldberg]

I think you are right about it being unbalanced. I'll properly add an extra "leg" on the tailstock end and the whole thing is going to be bolted to a solid tabel of some sort.
With regards to the the turret you are properly right as well, i think i will stick with the idea of having free space beneath the spindle to allow for chips to fall down, but with the rails as close as possible to the center of the spindel

[Guldberg]

From this picture it looks like when turning regular object, the turret will be positioned right above the rails

[LeeWay]

I see why you had the rails off to the side some. To stay out of the chips. In the last drawing, it looks like cutting forces will be right over the blocks on that rail. That should work well. As long as it's not overhanging the end, I think you have it.

[mcwilsonmfg]

you could move your slide rail ways down more if you wanted. Just put a sliding way cover over them and hook it to the turret to make it move then your chips would slide off just like on the commecial cnc lathes.

[neilw20]

Earlier on I noticed mention of torsional stiffness of the bed.
The stiffest thing (IMHO) is a ROUND TUBE.
Maybe you can build the whole bed based on a big chunk of pipe.
Now slide all of your stiffening webs on, align them and weld to the tube.
Now weld the flat faces to webs. Very easy keep it all lined up after tacking the webs.
Any comments?

If you intend using boring bars (and a turret) , consider that the Z travel has to be twice as logn as the hole you want to bore, plus the width of the boring bar holder. Look at standard size boring bars, or every time you want to use one, it will be too long or you will only be able to do very short jobs.
A standard 300mm bar really needs a 650mm bed to be useful.
If you do not use a turret, or do not rotate it then longer bars fit a bit easier.

[Guldberg]

One of the few parts i already aquired. One of the x-axis rails

[lerman]

I would suggest that you seriously consider using gang tooling instead of a turret. Take a look at http://www.omniturn.com/ for some examples of what can be done.

Adding a turret is essentially adding a rotary axis. Gang tooling provides the same functionality for almost free. The only additional cost is that the X axis must be longer to allow for multiple tools.

Generally, the big downside of gang tooling is that you can't use it with a tail stock; but in your case, you don't support one, anyway.

Ken

[Guldberg]

You are right, gangtooling is easier. There is no reason for the turret except that i would like to make one:-) Until i've made the turret i will properly make a simpel toolplate to accommendate a turning and a parting tool.

[handlewanker]

Hi, Interesting project, I take it your bed design length was determined by the fact that you already have the long linear rails?

Otherwise I would say that the bed/rail length in the x axis is too long.

For bar work, the spindle would probably have a 30mm bore, unless you're only going to machine billets which will be decided by the length of the chuck jaws for the stick out length.

A 160mm 3 jaw chuck has about 40mm of gripping length and this will decide how long the billet material can stick out, given the 4X ratio of diam to length to avoid deflection, a safe stick out would give you 160mm.

The problem with working from billets is the amount of wastage that occurs for the holding length.

The other factor is the spindle bore size which will limit your stick out again to 120mm for a 30mm bore.
A bigger bore size means bigger bearings, but is very desirable.
This will make a good read for many months to come.
Ian.

[Guldberg]

You are right, maybe i'll saw them to size. It can be shorten with 65mm, still maintaining room for the stepper to be mounted below.

The z-rails i haven't bought yet, but i dont think ill make them much longer anyway

Im aiming for around 25-30mm bore, this will go well with the largest ER32 collect

[RotarySMP]

Nice design.

To oppose torsion in the bed, diagonal ribbing would be more rigid. Welding a resonably thin closing sheet to the bottom would also make it stiffer.

If you put a second set of rails bellow parallel to the Z axis, spaced further away from the spindle, you could run a tail stock.

Obviously it intends what you are planning to make on it, but a tail stock adds hugely to a lathes work envelop.

[Guldberg]

Here is a picture of the headstock inside and a possible tailstock



I do not plan to turn long thin parts on this lathe, but it could be possible to make a headstock i guess. A lot of things would have to be taken into cosideration, but no doubt its doable

[Guldberg]

Changed the ribbing to diagonal. This is going be some very odd shaped pieces.



An fullscale MDF model will become handy when trying to figure out how to mill these things and to weld them together

[widgitmaster]

I like your design, and look foreword to seeing it go together!
Widgit

[LeeWay]

I concur now. It's looking like it will work well.

[protman16]

Guldberg,

Do a search for salnt bed lathes to see what machine manufacturers think is good design. You might find some ideas that may help your build. Keep up the good work.

[handlewanker]

Hi Gulberg, I highly approve of the ER 32 collets for the headstock, I personally am considering ER 40 for the bigger capacity for my centre lathe mod when the time comes.

Looking at the headstock cutaway, I see that you have what appears to be two angular contact bearings at the ends of the spindle.

Common spindle design has two angular contacts, ( or taper rollers), at the front end and two axially floating radial ball bearings at the rear end, to offset the pull of the belt drive.

This configuration overcomes the variation in bearing load due to temperature variations in the spindle, which would increase exponentially when the spindle elongates as it expands axially.

I hope you've got 3 Phase VFD planned for the drive, as it's so simple, economic and affordable nowadays.
Cuts out all that gearbox, pulley and belt variations.

Just one observation in the bed design with the angled ribs.

The welder who is commissioned to do the welding, (if you do it yourself you have masochistic tendencies LOL), will find that trying to weld in the now very acute corner angles without getting weak joints, is a very difficult job, and will relegate the construction to an unstable weak mass.

It would be far better, from a making point of view, to have the ribs as you previously planned, at 90 degrees, and do the weld preperations by veeing the ribs right to a point to ensure maximum penetration.

There is no way on earth that the bed is going to flex in use using the thickness of ribs as you have previously drawn them,( if they are to scale), however the bed will walk around a bit while the welding stresses are working themselves out, and that is even after the stress relieving.
The best method would be to have the bed cast, which would also require ageing and further stress relieving after the outer skin has been removed from the bed top and bottom.

As a matter of interest, a design that had a flat section base with two raised flat sections and a further flat section for the bed to carry linear rails, without the internal ribbing, and being screwed together with capscrews and dowels, would be very straight forward to make, and is a DIY type of construction where the builder has total control over the assembly.

It COULD (?) be made from heavy alluminium slabs, which are pretty stable when screwed and doweled together to form a hollow square or rectangular box.

Alternatively if alluminium is a no no, then cold rolled mild steel sections would be used, having NO welding, but being screwed and doweled together.

This construction would lend itself to a final top and bottom surface grinding to allow the linear rails to sit truly flat.

The surface grinding would be very cheap, (outside source), as it is only really needed on the flat upper surface and one or two passes would make it flat and true for the headstock and linear rail mountings without having to resort to shims for allignment.

The lower surface would have the mounting feet, which are angled to give the slant effect to the machine on assembly.

The design if viewed at the angle of slant is of a conventional lathe shape, except it is now turned on it's side a bit to give the slant.

It is my personal belief that the ribbing would create construction problems of humungeous proportions, and would be overkill as far as rigidity is required.

When the CNC lathe is working the forces are of a relatively light nature, as per CNC working characteristics, having many light passes as opposed to the heavy ploughing passes of a conventional centre lathe which does require ability to resist torsional forces when doing heavy roughing.

The lathe is at it's best for accuracy when the finishing cut is occuring and the bed and ancilleries lightly loaded.

A lathe can only be described as accurate when lightly loaded and doing the final finishing cut and if the slide ways are initially truly alligned with the spindle etc.

Having the ability to handle heavy roughing cuts is another time and another place scenario.
Ian.