A 3025 Moving Table CNC Router Build.

[daniellyall]

that's one way to look at but you will have 300mm of space between bearings so flex between bearings could be a problem if you do something heavy and having a weight permanently on the end of table 3rd axis could cause a slight lift to the table at the other end unless you have the 3rd axis over a set of bearings.
my little emco has its rollers 100mm apart from centre under the table (yes rollers not bearings) its as flat as the day it was made, the table is only 10mm aluminium.

if you wont to do it that way that's cool its your build and all the info in this tread just makes it easier of other people to build there own small machine

[handlewanker]

Hi, yes 300mm is the table length and that means it's fully supported all the time.

I don't anticipate having any flex in the table as it will be made from 25mm thick aluminium.

I decided to go to this thickness as it makes attaching things to it easier.

I have a table in cast iron, 200mm X 400mm X 25mm thick that came off a wood working machine many years ago and was going to build the router around this, but due to the fact that inertia forces from the cast iron's weight will wear out ball screws faster than actual usage and general wear and tear, I decided to make it the lightest possible.

The table is going to have a plain top surface with just 12mm holes drilled in it for fasteners.....that is the holes won't be tapped, just straight drilled holes all the way through.

I want to use the type of fastener that looks like a rock wall Dynabolt parallel expanding type fastener you would fix things to concrete with.

Only, they will be made like them and this will enable the fastener to be inserted in a hole without having to clean out threads that get bunged full with fine swarf etc.

Dynabolts are parallel fasteners that expand when the nuts are tightened and grip the hole you drill in the concrete.

So, with this design the bolts can be inserted and removed in an instant and no need to plug the holes to keep swarf out.

More often than not the table will have a sacrificial plate of MDF board on it, so for those occasions when milling solid objects that need holding, clamps or a vice can be mounted too.

I could go to Tee slots in a table of that thickness, but this is design time and a flat top solid table with plain holes is how I see it.
Ian.

[daniellyall]

25mm will stop any flex but good question would 1/2 inch steel plate be easier as the holes wont wear out like putting dynabolts in aluminium. making you justify what you are doing will make a better machine

[handlewanker]

Attachment 253994

this is my idea for a small machine there will not be any need for welding on the frame it self, just a small plate will be needed on each upright everything else can just be bolted on

Hi Dan, it seems simple enough to build.......I like things that are extremely simple as not much can go wrong and making does not take rocket science to work out all the forces that impinge on the frame.

In the drawing you show the base plate with two rails at 350mm long....for my purpose.it needs to be 600mm long to get a 300mm table travel for rails on the base and bearings under the table and 270mm between the column insides.

I also need the bridge to be only 400mm long to cover the 250mm wide table work area.

Anything can be made, it's when you come to fit it together that the problems arise...... bolts always move when you least expect or want them to.

At the moment I have a steel build plan design that I might have a dabble with if I get the time as welding is not foreign to me.

My main interest at the moment is the aluminium build plan, and to that end I've cut out the various parts drawn to scale on paper so that they can be arranged to get the minimum aluminium plate size with minimum scrap.

The 20mm plate size for all the frame parts works out at 1050mm long X 450mm wide, but as the frame sides are the oddest shape and generate the most scrap, I might make them only from plate and buy in 70mm X 20mm stock sizes for the spacers and rail base plate as they can be cut to length at the supplier and save heaps of time......looking at the parts list it starts to make more sense doing it that way.

I think the suppliers can cut to length without me having to mill them afterwards.

All the drilling and tapping will be the biggest work load.

I could do with a right angle drilling attachment for my mill as drilling and tapping into the ends of a piece of material 270mm long will be a problem unless I swing the mill ram head over, attach the pieces to an angle plate and hang them off the side of the mill table.
Ian.

[handlewanker]

25mm will stop any flex but good question would 1/2 inch steel plate be easier as the holes wont wear out like putting dynabolts in aluminium. making you justify what you are doing will make a better machine

Hi, They won't be actual Dynabolts as used for masonry fixing, but parallel split type that work the same way.

I think a diam of about 12mm will be OK as this will give 2D for the fastener in the metal.

A steel table would be too thin for that purpose, and at anything thicker and heavier would be like moving a semi trailer back and forth rapidly.

Even with a steel build router to the same work dimensions I would have an aluminium table too.

Inertia is a real problem with mass, and accelerating for rapids will put lots of wear on the screws.....going to 25mm thick aluminium will give me thickness without a weight penalty.
Ian.

[daniellyall]

that pick was just a quick drawing to give you a ruff idea of what I was getting at to give you 400mm on the X axis so if you Z axis was 100mm wide you will only need 500mm not 600mm for the x axis rail, the length in total would just be X axis plus stepper. As long as the machine is not a squire it wont matter what axis is your longest but making the X axis the longest you could have a longer A axis a bit shorter on the Y axis.

A suggestion with your A axis why not put a strip of material along the back or front edge and cut it squire when the machine is built then you can take the A axis of an on and just have a couple of cam clamps to hold it in place this way you don't have to build the Y axis as long, this make it cost less so you have more to spend on a good spindle

[handlewanker]

Hi, There is a big difference when having the table as a moving item.....for stability, when the spindle is at the far right or left, you need to have the table longer moving under the bridge as opposed to wider under the bridge,

Slides work better as guideways when they are longer and not spaced too far apart.

Therefore with the design I propose the slides are 600mm long and spaced apart at approx. 200mm under the table.....this gives the table bearing configuration as 200mm wide and 300mm long at all times during the table movement.

Having the bridge supports as close together as possible gives it strength to flexing and twisting......too wide apart and/or too high and you will get flexing in both axes.

All this is extreme thinking for a 3025 router, but as I intend to make this small, rigid and taught in all aspects, the design I have at the moment is what I consider to be usable for the cutter forces envisaged.

Bridge mills are more rigid in all axes compared to moving gantry models, but can be just as deadly if the slide configuration is not carefully designed.

My opinion is this......not too wide for the X axis, not too high for the Z axes and as long as you need for the Y axis.

Expanding dimensions on two axes, the X and Z to gain capacity, especially the Z, and you get a frame that will, by nature of the law of leverage, not be able to resist deflection when the cutter digs into the metal.

If the frame deflects .5mm in any direction, then so does the cutter, and if the cutter deflects .5mm when it is cutting you get a hop skip and jump effect as it rotates and each tooth hits the job.

This design I am proposing will be as good as it can get, and if a bit more material can make a huge difference then that is how it will be.

Designing down to a cost does not enter the picture, as that is self destructive no matter which way you look at it, but there has to be a sense of realism too.

BTW, the extra 100mm table over length is not a problem as it's off the work area and serves to just mount a 4th axis on, or whatever, whenever....the other end is 50mm longer for a tailstock and also just overhangs the work area.

This extra table length from 300mm to 450mm is just a bit more aluminium.
Ian.

[daniellyall]

cool sounds like you are on to it

[handlewanker]

Hi, early days yet, still gathering information as to what it will cost and the aluminium/steel prospect.

One thing's for sure if the aluminium bits come out under $500 then I'll most probably go for it, otherwise building with steel will be much cheaper and so maybe that could lead to some more expensive parts in the build.

Just finished sketching the steel design and it appears to be quite simple and robust, apart from the very low cost which entails quite a bit of 75mm X 30mm steel tube for the frame.

In that design the base is formed in a rectangle 600mm X 300mm from steel tube, 30mm side upmost, with one internal cross tube across the middle and a 6mm thick steel plate welded to the top for the linear rails to be mounted on.

There will also be two 30mm X 10mm steel strips welded on the plate for the rails to be mounted on.

Rails will be 20mm HIWIN type and this will give 34mm + 10mmm to the bottom of the table to clear the ball nut housing.

There are two 10mm steel plates welded either side of the frame at one end to bolt the uprights to.

As I hate too much bolt work in an assembly, the bridge will be made from two 70mm X 30mm tubes one above the other spaced 125mm apart overall with the 70mm side facing upwards, and a 150mm X 400mm X 8mm plate with two steel strips welded to the front to mount the rails on.

This lot gets welded to the two uprights to form a complete assembly.....bridge and two sides.......and this then gets bolted to the base frame with 8 10mm socket head screws, squared up and fitted with two dowels per side.

Doing it this way enables the front of the bridge to be machined for the rails and ball screw mounting as it will lay flat on the mill table.

The base will be machined on the linear rails mounts also as well as the ball screw bracket seats.

I'll still go for the 25mm aluminium table.

I like this design so much that I might do a parallel build with the steel version as it's so cost effective.

There is so little actual machining to be done, apart from drilling and some tapping, that even if someone were to send it out it would not be a humungeouse cost factor......... or a skilled worker with an angle grinder and some file work with a good straight edge could do the same.

The bright side is that when the assembly is finished it can be sand blasted and then finished with a 2 pack epoxy paint job.

This design is getting to me so a model might be on the board before long.
Ian.

[awerby]

The cost of the aluminium plate that AARGGH posted, has me rethinking the material specs.......it will still be 20mm thick but I'm now contemplating acquiring a load of scrap offcuts of ally and casting the sections in steel moulds then machining them where needed.

This will save a load of time (and money) in cutting out from plate stock and also the offcuts thereof.

I've melted aluminium using propane and a steel ladle some years ago with good results, and casting in open steel moulds is also not a problem.


There is also the bucket of sand and a polystyrene pattern method that could be tried.


[I like the idea of casting a monolithic frame. But having done some casting, I'd avoid the open steel molds. The problem is that they won't allow for shrinkage, and aluminum shrinks a lot, which can ruin your casting. Go with the lost styrofoam, and use a heavy sprue and risers. You can also use traditional sandcasting techniques, but that would require making patterns. If you're melting in a steel ladle, make sure to use a ceramic coating, or the steel will contaminate your aluminum. Aluminum can also corrode steel and make holes in your ladle. Also, using ingots of aluminum intended for casting, or at least remelting pre-cast parts, will work better than offcuts of aluminum sheet, bar and rod, which are formulated differently than cast aluminum. ]

There is no reason why the build can't be done using steel and welding where necessary.

Once a frame is welded it WON'T move......unless you then go and cut into it and allow the stresses to re-pull the structure out of wack.

Most of welded structures just get machined on the various mounting faces to true them up, and this does not translate into serious metal removal.....done that many time with no problems.

However, with steel, although it is 10 time as strong as aluminium both in tensile and sheer, the sections at 20mm thick would make it stronger than an Abrahams tank and need a fork lift to move it.

That creates a problem, for if the side frames are reduced to 10mm thick and would be amply strong, they would have a tendency to spring without being able to resist the side loading, unless the edges were turned to form right angle buttresses, and that starts to get complicated.

I went to 20mm thick material as a design concept, as there is a lot of edge drilling and tapping when joining items to one another at right angles.

The base frame has 4 spacers at the bottom and two more at the top behind the bridge......and there is a 20mm plate on top of the bottom spacers to mount the Y axis linear rails on.
Ian.

Steel can work, and would certainly make for a more massive frame, which has its advantages. But you'd have to design it for ease of machining, without places you can't reach without an extra-long cutter.]

[daniellyall]

If it was me I would not weld anything I have access to a big router and manual mill so I would be able to machine what I made drill and tap as well would not be a problem

Its good when you can access big machines that are relabel

I have seen a different way to make a small machine what is a very simple way to do it as it uses big round rails I will see if I can find it, it uses big bits of aluminium as well and its able to be expand up to a certain size where you just just have to expand one axis as the other axis can use the replaced axis.

Its just some more good info to help you build a good machine

[daniellyall]

just seen this
First Nomad CNC Mills Have Been Delivered - CNCCookbook CNCCookbook

[handlewanker]

Hey man, that Nomad is quite a neat mill design....not quite my way to make one as I'm a dyed in the wool metal worker, but it performs and that is the main thing.

I have a big Ajax Mill, something like a King Rich etc, so machining is not a problem, also a complete welding outfit and a 10" swing lathe.

Welding has to form part of the making as too much bolting together is tedious and prone to moving without you knowing it.

I'm aiming at having two parts to the steel build one, the base as a separate entity and the bridge and side supports as a complete assembly with the face of the bridge having a 10mm steel plate bolted on, and that is necessary to allow aligning the X and Z slides when the assembly is finally finished.

A lot is dependent on the getting the structure square and parallel to the various slideways as the build progresses.......if the slides are out of square in any way you don't have a router worth working on, and that is un-compromisable.

Once the base is finished and the bridge and side frames are bolted on, the 10mm plate gets bolted to the bridge face, having had the two linear rail seating strips machined before hand.

Having a separate plate bolted to the bridge face gives you freedom to move it if required to align the X axis slides square to the table and also the Z axis slide in the vertical plane.

I agonised for hours on a way to get the final alignment in 3 planes once the welded assembly was all together, and came to realise that some form of adjustment for final set-up was compulsory......you can't just weld a structure together, bung on some slides and hope they all magically get square to one another.

Most of the build for the steel model base and bridge will comprise of 80mm X 40mm steel tube or whatever stocked item is carried, so I think it will work out extremely cheap to make.

That makes fitting the best slide ways possible, and HIWIN 20mm type are on my must have list without breaking the bank.

If I could get the ISO 20 spindle like the one I want for the SVM-0 mill I have on order, that would make a lot of possibilities for tool changing without having to unscrew the ER chuck and reset depths between cutter changes, but that method relies on a separate belt drive motor for the spindle, whereas this build will go to the 2.2KW water cooled model.

The Kitmill design has a 3 step pulley drive to allow torque variation with a separate motor, so it can be done if really wanted.....the price for the ISO 20 spindle is about $350 from Defeng in china.

However, the ISO 20 spindle revs to 7,000 rpm max, and that is mainly for milling work, whereas as a router for carving and engraving etc you want to get up to the 25,000 mark......I assume.

That will be the last acquisition, so lots of browsing to find an ideal one.
Ian.

[handlewanker]

Steel can work, and would certainly make for a more massive frame, which has its advantages. But you'd have to design it for ease of machining, without places you can't reach without an extra-long cutter.]

Hi, I think with the steel frame build the best way to get at everything is to make it in two parts....base and bridge with sides attached.

That way there are only a few flat surfaces to skim after welding.

I don't want to make it too massive, and I think 80mm X 40mm tubing will do the trick without sinking the ship.

Having built many fabricated structures over the years, getting at the machined areas is part of the design requirements.

The biggest problem with most mills is to mill and/or drill in the horizontal plane....up and down is OK, but sideways is difficult without a right angle drive, and they cost a bomb.

You can swing the head over and mill/drill off the edge of the table with the job against an angle plate, which is what I have done in the past, but getting everything square to one another is also a pain.
Ian.

[daniellyall]

yes that sound very sound don't forget about racking that's a pain to deal with if, if you can fit the bass on your mill I would cut to dados for the y axis risers to sit in so then you know right from the start that your Y axis is squire.
my emco has never had to be adjusted for racking sinces I have had it (3 years).
the big machine I uses I have to do it every 90 days it always out by around 1 - 2 mm over 600mm this also depends on the weather and how bad the earth quakes have been.

the size of you machine it should not be of that much of a problem as you are using quilty rail but you never know

[handlewanker]

Hi Dan. I don't think racking is going to be a problem with a moving table type, so it will be driven from the centre single screw.

Racking can be a problem if the linear bearing blocks are too close together on the rails and the rails are spaced too far apart, so giving a slide way aspect ration oversquare.

I'll have the bearing blocks spaced 300mm apart along each rail for the Y (table travel) and rails spaced apart at approx. 210mm.

As I have a steel build model in mind too, I had another look at the drawing you posted, and that together with a collection of square tubing offcuts I keep to enable ideas to work when you can see the material in the flesh gave me some more insight as to how a steel built model would go together.

First and foremost, aligning the bridge X axis to the Y and Z axis rails will be a difficult exercise....so I had a look at your drawing as it is fabro based and decided that if the build was constructed like a layer cake, that is from beginning to end by progressively adding bits on and welding them in place as you go along, the build can be squared up and aligned in one go.

However, this does not give you the accuracy you'd want with CNC machining in mind.....OK for carving and engraving, but when it comes to making things square in 3 planes.......that's a whole different ball game.......there has to be room for adjustment to get the last .01mm .....or close enough at that.

So I decided that the steel frame build will be all in one piece except for the bridge facing plate that the linear rails get mounted on.

It will be bolted onto the welded bridge assemble ......this will give scope to adjust the bridge for any out of squareness to both the X axis and the Z.

If the bridge hangs down one side a few thou or so you can adjust it with the bolts to get it dead square across the whole 250mm travel, and if the Z axis is leaning forward or back it too can, by a bit of filing and scraping, be adjusted by adjusting the mounting face pads that the bridge will be bolted to.

This goes for the alignment of the Y axis to the X, and this too is done by adjusting the pads behind the bridge.

Having only one item to move about means the rest of the build can be to boilermaker's limits and that does not mean higgledy piggledy or bad workmanship.....probably to the nearest mm or two before final setting up.

The aluminium build is a toss up between buying a sheet of 20mm aluminium 1200mm X 900mm or making MDF patterns and getting a quote for casting them.

Castings give you a more composite design with previously bolted together parts cast in one.

This gives you 4 parts.....the base, shaped like a shallow box with one or two ribs across it and with linear rail mounting pads cast in.....only needs machining on the pad faces and on the sides at the ends for the uprights to bolt onto.....the bottom only needs a light clean up at the thickened corner points to make it sit level.

Two uprights with cast in ribs on the face to reduce the side ways deflecting forces......machining here will be along the bottom face to mount to the base and along the top face for the bridge to bolt against. both done with the same set-up.

The bridge is formed like a long U shaped channel or box with the ends closed..... only the front face, having the linear rail mounting pads, need machining, and the end faces where they bolt to the uprights.

As soon as I get a cost for the aluminium sheet material I'll know which way to lean to.
Ian.

[daniellyall]

your thinking sounds very sound making a model does make life easier so you have something to hold on to

[handlewanker]

Hi, that's true, with the model in front of me I can see where the main points of hard to weld areas are, like in the corners between the bridge parts, but as the model stands in front of me in the aluminium plate design, it starts to become easy when you can see it in the flesh.

Now to collect all the tubing length dimensions and work out how much tubing I'll need, but at a quick guestimation, the base needs 3 metres of 80 X 40 X 2.5mm tubing and the uprights and bridge need 1.35 metres of 100 X 50 X 3mm tubing.

Add to that a piece of 600 X 250 X 4mm steel plate for the linear rails to mount on the base, and a piece of 400 X 150 X 8mm steel plate for the face of the bridge to mount the linear rails on, and that is about the complete material outlay for the basic frame.

At a rough guess I'd say the complete frame build would work out at a couple of hundred bucks, which allows plenty for more exotic items to make it work.

It's extremely important with frame work of this nature, where you have many parts being joined by welding, to not do too much welding in one go.......heat being the biggest curse of welding means you get expansion that will distort the frame.

So it will be weld a bit and allow to cool and add bits more progressively while allowing the welds to go cold......this way when the welds are cold the contraction won't pull the frame out of wack as you can bump it back a bit until it's completed.

Having welded gate frames from tubing and learned how to work with the effects of expansion, I never tack a lot of entities together in one go as this will restrict the ability of the tubing to move and then move back when cold.

You have to let the tubing move and between fit-ups re-align it before adding further bits and closing the frame work in a tight loop.......if it can't move freely it will twist and distort and it's impossible to straighten it afterwards.

One big advantage with a composite steel frame build in one piece is the reduction in machining, drilling and tapping which alone also makes for a more rigid framework without adding undue mass.

The aluminium plate build won't have these problems, but the cost could push it out of the window.

I need to get some prices for casting the various parts as a further cost saving.
Ian.

[daniellyall]

If have seen what a weld can do to a big steel plate a 1m x1m x 12mm bent like a banana

[handlewanker]

Agreed, It depends on how the plate was welded and prepared etc......confined uncontrolled expansion will bend anything if the material can't move freely and return to it's original position.

Most if not all welded structures of any large size would not get heat treated in any way whatsoever to stress relieve them, but they still function as designed.

In my opinion, once the frame work is welded, the small amount of machining will not affect the final machine or throw it out of wack to any degree, and any distortion that may occur can be progressively controlled as you go along.

I've just completed the design for the steel framed router to the same working dimensions as the aluminium one, and am now making another model to see how the layout will compare to the aluminium one previously described.

For ease of construction, I would go to the steel tube model by choice, as it's quite economical and quick to build and it practically falls together in the initial build, only needing some machining on areas where the off the shelf items need to be attaché to.

Until the model is on the table, the practicality of the layout cannot be truly assessed, and what seems good on paper often turns out to be a monstrosity in the flesh.
Ian.