[Rhodan]

I've had a K2 15"x25" for about a decade and due to life circumstances I haven't really used it for the last 5 or so years. Circumstances have changed and I've moved to a place where I have a huge double garage and only one vehicle that barely takes up 16' of the 24' length. I've got room for all my woodworking tools plus the K2 PLUS a bigger wood router table! I'm going to dedicate the 15x25 to electronics if I get this 48x48 built.

One thing that got me excited was finding out phidgets carries all sorts of linear motion supplies now and for not much money. I realize this means its probably not what you'd call top of the line quality but I'm a hobbyist and I still haven't managed to wear out anything on the K2 which uses plastic anti-backlash nuts so I'll probably be dead before anything wears out. I'm also not worried about 0.000001mm accuracy - this will be wood and the difference in temperature between the shop and where the piece ends up will result in more inaccuracy than the CNC machine anyway. Plus just about everything I make will be decorative and not in need of high tolerances.

Anyway, what I'm thinking of using is the 25mm x 1400mm ballscrew in the X and Y (useful travel 1252mm or 49.29") https://www.phidgets.com/?tier=3&cat...=34&prodid=672. Haven't even started thinking about single or double Y ballscrews or single motor w/belts/chains or double motor. Do you think this will be heavy enough to avoid whipping yet not so heavy as to be overkill? They do have smaller ones but the price difference isn't that great so I'd rather go big if it doesn't cause other problems.

For bearings and rails I was looking at 25mm supported round rail and open bearings (https://www.phidgets.com/?tier=3&cat...=34&prodid=652 and https://www.phidgets.com/?tier=3&cat...=34&prodid=681) Of course these rails will have to be longer than the ballscrew travel to accommodate the X Gantry width and Y carriage but I'll figure out that detail later. I'll probably get extra long and cut where necessary.
For some of the framing I'd use their PG40x80 T-Slot extrusions (roughly 1.5" x 3") with connectors to build taller or thicker. I'm thinking perhaps the Y plate and X wings may have to be machined from aluminium for stiffness but I'm not adverse to trying the T-Slot if there's a decent chance it'll be strong enough.

Like the title says, I'm just spit-balling at the moment and wondering if you folks think this plan is good or foolhardy?

Edit: I mak speel gud!

[handlewanker]

Hi, as it's all been done before, re-inventing the wheel will be a tedious exercise not knowing the outcome..........take advice from those that have built, go on UTUBE and browse the various DIY builds and then design with confidence.........they're all the same really, just the fixing methods that vary.
Ian.

[Tutu]

All sound good but you are really much better off with Hiwin style rails vs SBR but SBR25 is the minimum acceptable solution.

[ger21]

Like the title says, I'm just spit-balling at the moment and wondering if you folks think this plan is good or foolhardy?

That depends on how much you want to spend, and the quality of the machine you expect.
Accuracy is not the issue. Any machine, built out of almost anything, can be very accurate. The goal is rigidity. The more rigid, the better.

[NIC 77]

One thing that got me excited was finding out phidgets carries all sorts of linear motion supplies now and for not much money.

Just so you're aware, that stuff is available all over EBay and Aliexpress. You can also get custom end machining from those places. I think you should figure out exactly how you will mount your motors prior to buying ballscrews as you may want custom end machining. I got custom end machining from the ballscrews I bought from BST Automation on Aliexpress to use with my SYK Nema 34 mounts and to have double nuts at the end.

Anyway, what I'm thinking of using is the 25mm x 1400mm ballscrew in the X and Y (useful travel 1252mm or 49.29") https://www.phidgets.com/?tier=3&cat...=34&prodid=672. Haven't even started thinking about single or double Y ballscrews or single motor w/belts/chains or double motor. Do you think this will be heavy enough to avoid whipping yet not so heavy as to be overkill?

You shouldn't be using 5mm lead for those lengths IMO. It will be slow. The link you posted is to a 5mm lead ballscrew. Actually I clicked on "other SFU25 Ball Screws" to see what else they have. Every single darn 25mm ballscrew they have is 5mm lead, even one that is 2 meters long!

You are far better off using 2020 ballscrews which can be found everywhere, like Ebay and Aliexpress. And perhaps 10mm for the Z axis so it doesn't lag too far behind on speed when doing a 3D carving. I'm assuming you will use steppers, not servos.

For the linear rails, perhaps check out automation overstock. Also there are many suppliers of square linear rails on EBay, aliexpress and elsewhere.

Probably the best thing you could do is to draw out your design on CAD, and post some pictures, and ask for advice at that point.

You could also do a total cost estimate and figure out if it's worth it to buy something like a Saturn 2 4' x 4' instead.

[wizard]

Welcome!

I've had a K2 15"x25" for about a decade and due to life circumstances I haven't really used it for the last 5 or so years. Circumstances have changed and I've moved to a place where I have a huge double garage and only one vehicle that barely takes up 16' of the 24' length. I've got room for all my woodworking tools plus the K2 PLUS a bigger wood router table! I'm going to dedicate the 15x25 to electronics if I get this 48x48 built.

Sounds like a plan.

One thing that got me excited was finding out phidgets carries all sorts of linear motion supplies now and for not much money. I realize this means its probably not what you'd call top of the line quality but I'm a hobbyist and I still haven't managed to wear out anything on the K2 which uses plastic anti-backlash nuts so I'll probably be dead before anything wears out. I'm also not worried about 0.000001mm accuracy -

As others have pointed out there are many sources for linear motion hardware these days. I must point out though that many of these businesses are selling to the 3D printer markets where the parts may be acceptable but are pretty much worthless for a CNC router. A lot of the extrusions sold on line fall into this worthless category. I wouldnt dump money into extrusions of questionable quality.

Interestingly accuracy isnt often that important on a CNC but repeatability is. These are two different concepts. Repeatability and the related freedom from backlash makes a CNC a usable tool.

this will be wood and the difference in temperature between the shop and where the piece ends up will result in more inaccuracy than the CNC machine anyway. Plus just about everything I make will be decorative and not in need of high tolerances.

That is fine today but once you have the machine you will likely end up dojng things on it that you havent thought about yet. Making the best effort to get the highest quality possible out of a given budget makes the most sense to me. Lets face it most of us dont have the cash to buy a high end machine this we want to make something that we can afford. Thus you really need to decide upon a budget and a set of machine specifications first.

Anyway, what I'm thinking of using is the 25mm x 1400mm ballscrew in the X and Y (useful travel 1252mm or 49.29") https://www.phidgets.com/?tier=3&cat...=34&prodid=672. Haven't even started thinking about single or double Y ballscrews or single motor w/belts/chains or double motor. Do you think this will be heavy enough to avoid whipping yet not so heavy as to be overkill? They do have smaller ones but the price difference isn't that great so I'd rather go big if it doesn't cause other problems.

It kinda feels like you are seversl steps ahead of yourself here. Id try to nail down basic frame arraingements first

For bearings and rails I was looking at 25mm supported round rail and open bearings (https://www.phidgets.com/?tier=3&cat...=34&prodid=652 and https://www.phidgets.com/?tier=3&cat...=34&prodid=681) Of course these rails will have to be longer than the ballscrew travel to accommodate the X Gantry width and Y carriage but I'll figure out that detail later.

Again this is feeling very backwards to me. You really need to know what the frame will be like before you even think about linear bearings. Even things like the hieght of the Z axis impacts the length of the X. The use of round rails will require significant spscing on the X linear rails due to their lack of rigidity which can lead to longer rails.

Now i realize this is a hobby machine but do consider profile rails for at least some axises.

I'll probably get extra long and cut where necessary.
For some of the framing I'd use their PG40x80 T-Slot extrusions (roughly 1.5" x 3") with connectors to build taller or thicker.

This probably disturbs me more than anything. Im not sure what you mean by some of the framing but i suspect that this extrusion will be too light for most of the framing. You seem to onow this suggesting to use connectors to build taller and thicker. In my mind this is a very bad idea.

I probably should preface this with a discloser, im not a big fan of aluminum extrusions for a machine build especially for certain parts of the machine. The big reason from my persoective is the cost to make a robust machine. The extrusions themselves are expensive but you really get nickled and dimed by the cost of all the required hardware. The second problem is the less than ideal structural nature of the extrusions, especially the really cheap extrusions. So now that you know im biased a few comments.

1. Buy extrusions from reputable companies in size suitable for the part of the machine you are engineering. For a gantry beam to span 48" plus, you are talking a substantial cross section. Misumi, 8020, Rexroth and others may have beams suitable for the span and the perfotmance you want.
2. Do not skimp on the required angle brackets, T-bolts joining plates and whatever else they will nickle and dime you for. You want to make sure the machine does not disassemble itself while running. In the end you are primarily wotking with aluminum which is soft and subject to loosing up over time.
3. Many express concerns about the machines weight which is a real issue but the reality is a little weight in the machines bas can prevent a lot of walking around of the machine form vibrations or dancing. So focus on light eeight only if you really believe you have too.

Now to get you to consider steel which can make sense in certain parts of the machine even if you arent set up to do a lot with steel. One area where you can get a big payoff without too much builders stress is the gantry beam. If you can drill and tap holes you can leverage the structural advantages of a steel beam for the gantry.

I'm thinking perhaps the Y plate and X wings may have to be machined from aluminium for stiffness but I'm not adverse to trying the T-Slot if there's a decent chance it'll be strong enough.

Im not sure what you mean by Y plate here. The gantry need to be a beam. If the X wings you refer to above are the supports for the gantry beam, often you want a box section for the best pay off in materials used to achive a stiff solution. Extrusions or steel tubing can be used here but the design can become more complicated if the Z acis is very high.

Again the specifics of your machine will dictate how some of these components are designed.

Like the title says, I'm just spit-balling at the moment and wondering if you folks think this plan is good or foolhardy?

Edit: I mak speel gud!

Foolhardy? I dont think that is fair. Most people go through many iterations of designs before starting a build. Sometimes things get changed up half way thtough.

Some of your questions though really cant be answered without an idea as to machine arraingement. Maybe i missed it but im not even sure what type of Z clearance you are talking about here. That is important because a very high Z clearance can impact the length of the machine. As for ball screws the lead of the screw is important in determining what RPMs will have to be reached for a given feed rate. Since this is wood working you likeky will want high feed rates.

It isn't foolhardy to build a machine out of aluminum extrusions even if i have reservations about the use of such. I do have a concern that you avoid buying junk extrusions and that they be of suitable size for the application at hand. Some of these low end extrudions are so bad that you would likely be better off with wood construction

Im not particularly concerned about the use of round rails as long as you understand the limitations of such rails. You understand the performance goals you want to achive.

Oh by the way on a cell phone at the moment all uncaught spelling errors are now history.

[Rhodan]

As I said in the OP, I've already had a machine for several years and did a lot of stuff, just not in the last five. It originally had unsupported round rails and I managed to make a new aluminium plate for the gantry and switched it to supported round rails (1/2"). I had a LOT of deflection with the unsupported rails but after swapping to the supported ones that disappeared. Still have a bit of deflection thanks to the unsupported rails on the X but some care in tool orders and plunge rate negates the effect.

My Z height will be fairly short. I played with 3D carving and stuff on the K2 and ended up just doing 2D in layers instead due to much less hassle! If I do any 3D it'll just be thin things. I'm more into vcarving than 3D so either sheet goods of 1inch or less or hardwood boards.

Yes, Y plate means the gantry beam and wings are the supports. That's what my K2 uses and I've never noticed any problems with flex.

[handlewanker]

Without a basic concept sketch of a probable layout nothing will make sense as there are too many variables.

A 1.2M X 1.2M CNC router is quite big when you consider it has to be rigid enough to withstand the forces of a cutter, even on wood.

At the same time, as a CNC router type machine, it's all been done before, you don't really want to re-invent the wheel.

Just follow the trend as has been established for many builds.....you really only have 2 choices.....moving table or moving gantry and a high speed water cooled spindle is your first choice as you won't be milling anything harder than wood.....metal is not an option.
Ian.

[wizard]

As I said in the OP, I've already had a machine for several years and did a lot of stuff, just not in the last five. It originally had unsupported round rails and I managed to make a new aluminium plate for the gantry and switched it to supported round rails (1/2"). I had a LOT of deflection with the unsupported rails but after swapping to the supported ones that disappeared. Still have a bit of deflection thanks to the unsupported rails on the X but some care in tool orders and plunge rate negates the effect.

My Z height will be fairly short.

That by itself means little. It is all about the implementation.

I played with 3D carving and stuff on the K2 and ended up just doing 2D in layers instead due to much less hassle! If I do any 3D it'll just be thin things. I'm more into vcarving than 3D so either sheet goods of 1inch or less or hardwood boards.

So a machine mire oriented towards processing sheet goods?

Yes, Y plate means the gantry beam and wings are the supports. That's what my K2 uses and I've never noticed any problems with flex.

It really is a question of design. Some designs have the gantry beam ridding almost directly on the X axis linear bearings, ither designs might have the beam 10 inches above those linear bearings.

If those beam supports have any significant length there is potential for flex and vibration especially in larger machines cutting agressively. This is why we need to understand machine arraingement and Z height. In some cases plain plates can be an issue.

In any even i failed to get a related point across, using box sections is often more economical for a given amount of stiffness in gantry beam supports. This especially the case when short pieces of square or rectangular tubing can be had at low cost as drops. The economics comes about due to less material needed (paying by the pound plus processing) but there is another economy and that is in weight control. In a home built machine, especially one with simple stepper drives mass in a moving gantry can become a problem thus you want to favor solutions that are stiff but relatively lighter.

All that said about weight control is only really valid for moving gantry hobby machines. Mass and stiffness is generally good to have in machine tools in general. On wood cutting routers id have to say that there comes a point where you dont get a lot of value in trying to make a machine even stiffer.

In any event what im trying to get at here is that the mechanical arraingement of the machine is a huge factor in how it will perform as is how that arraingement is implemented.