[dimmaz88]

Thanks Peter, I'll have a think about it. I'm thinking of adding triangulation instead of the horizontals.

My issue with the big plywood box is figuring out how to flatten the ways for the linear rails. If I use steel I can make the "ladders" separately and machine one on my machine for the ways.

I've read quite a lot that steel/aluminium should be a minimum of 5mm, any less and it resonates too much. Have you any experience of this?

[peteeng]

Hi Dim - There is no hard and fast rule on thickness that vibrates or does not vibrate. My 3mm steel machines certainly do not vibrate in ways that upset the cutting. Large hollow sections do have acoustic noises as they act as air columns (eg organ pipes, so block the ends if its annoying). My z axis for instance is an I section made from two 3mm steel channels that are bolted back to back and it shows no sign of vibrating. In one thread I did a lot of vibration analysis on AL construction extrusions and all the small free edges where vibrating like crazy. Everything vibrates just don't want vibs that upset the cut.... Design the machine to be really stiff and it won't vibrate. If it does then solve it. Don't try to solve it before it happens. Peter

Plywood machines are fine if your aiming at 0.1mm accuracy, if you want better than that then you move to metal or concrete or composite. I have in the past lapped ply and mdf to a high tolerance, just takes a good lap and some wet and dry. Epoxy putty and a few hours of making dust..

I've been lapping some concrete surface plates to learn about concrete finishing for machine parts and I've got them to 0.01mm fairly fast. Peter

[routalot]

Hi Peter, I can't get my head around your calculations, although I believe they're 100% accurate.

In the image attached imagine a sideways force, are you saying that a simple plywood box would resist that force more than my steel ladder design?

Attachment 497996

I think I'll have to learn how to do the analysis in fusion and pacify myself


Hi Routalot, my thoughts exactly. I could just have a turntable for indexing. Although on simpler parts, imagine a pawn chess piece, it would be nice to just turn it as you would on a lathe.

I don't see the need for a turntable at all.If the gantry moves to the other side of the workpiece and the C axis has rotated to direct the tool at the job,where is the need for the complication?Am I missing something?

[dimmaz88]

"Although on simpler parts, imagine a pawn chess piece, it would be nice to just turn it as you would on a lathe"

If I'm turning simple parts a turntable negates the need for multiple setups, I can turn them in one go. And I suppose if the geometry's simple it wouldn't be using all of the axes anyway, it would work more like a 3+1 machine.

[dimmaz88]

Hi Peter, am I right in thinking this composite design will have at least the stiffness of a plywood box? I like the idea of using steel as it will make mounting rails etc more solid. However, if I can use a ply infill instead of steel cross braces it will save on complexity/money. I'd weld steel angle to the box section, then bolt the plywood to that. Maybe even use adhesive too if required.

[peteeng]

Hi Dim - Mounting parts to timber is straightforward and should not be discounted. Just have to learn about new processes and it all comes down to what you want to achieve. SHS is not flat and is either cupped or crowned so its not all plain sailing in steel world either. You will have to do some simulating to figure out the relative benefits of any of your designs. If you weld be prepared for warpage and the need for correcting the surfaces. So to answer your question ply as the bracing is a good approach and I'd glue it to the steel so the load transfer is over a bigger area. . Peter

[dimmaz88]

Thanks Peter, I tried simulating with fusion but I don't have it as part of my package.
I'm aware of the issues with steel and I'll be cognisant of it when building. I'm thinking of either using a replication method with moglice, although I'd need a 2.5m straightedge. The other alternative is to have the frame machined professionally, I'll have to look into how much this would be.

I feel happy with this steel/plywood approach, it's good to have a direction to go in finally. Thanks for your input so far

[peteeng]

Hi Dim - Hmmm just looked up Fusion hobby and they have knocked out the simulation and other bits. They are about to have the Fusion 10 year celebrations... I've been using it for 1.5 years... There are several free FE packages around. Haven't used them but if your interested in this its worth digging around. Freecad is free and has simulation. Peter

[Titaniumboy]

just looked up Fusion hobby and they have knocked out the simulation and other bits. They are about to have the Fusion 10 year celebrations...

I am taking a single community college class and was able to sign up with Autodesk as a student. The student version of Fusion 360 seems to have all the bells and whistles that were stripped out of the hobby version.

[peteeng]

Hi Dim - Thinking about your column a bit more. I like to stay in one material universe. Mixing materials you really need to know how to get the best out of them otherwise you can end up worse off. So by the time you tack in an angle to take the ply you may as well have tacked in some steel sheet. Then deal with the welding and finishing details. Peter

to close off the plywood discussion as I think your going to end up in steel world. There are a couple of things to consider about a structure and parts of a structure. Bending, which is described as an out of plane action. In-plane loading and shear loading. A 17mm thick F17 grade plywood when bent is the same stiffness as a piece of 7mm thick steel. So in bending it has very good performance as its quite a bit lighter and damper then steel. In-plane the 17mm thick ply is the same stiffness as 1.3mm of steel. So for instance in your design the sheet bracing is in-plane load so a 1.5mm would be good. The columns are in bending so you need to size for that. Local shear load is something not well understood by DIY. So currently you are about to place rails on 5mm thick material. This material when loaded deflects like a hammock (this is called membrane action) it has to deflect to take up load as its a membrane, like the hammock its droopy until loaded. Commercial mills never have rails mounted on air but on deep webs to counter the local shear deflections. When your trying to make a stiff machine every component has to be considered and especially at points where loads come in and out of the structure or transfer from one part to another (such as at carriages). So keep at it and develop the design a bit more....

[dimmaz88]

Yes Peter, you're right with the steel idea. I may as well just weld sheet steel instead of using angle and ply. I assumed the steel would have to be thicker than that, but I can handle 1.5mm.

With regards to local shear, I had thought about it but didn't know what it was called
This was a reason for liking the idea of a concrete column. As the diagram, I could create deeper sections on the corners to accept fixings and re-bar. I like the idea of concrete as it would be a one piece component, and the weight would be similar to steel if I use the sizes in the sketch.

[peteeng]

Hi Dim - With a casting like that you should try to keep the thickness similar everywhere. The sharp internal corners will crack over time (as the thick bits and thin bits expand and contract differently with thermal changes). So make the outside shell the same thickness and have a radius (or cove as its called) on the inside corners. Peter

[dimmaz88]

Hi folks, long time no speak! Life gets in the way of these personal projects!

I've not really been thinking much about my machine build, I've just had it percolating away in the background.

I did come across the Tarus Claymill C, which is a very lightweight looking machine compared to the one posted earlier in the thread. From the pictures it looks like the post is aluminium, I wonder how stiff it is side to side, there's only a brace front to back. The "ram" is also very spindly, although the spindle does look tiny.

I'd be interested in your thoughts on this one Peter

Attachment 499816

[peteeng]

Hi Dim - Its a clay mill and clay doesn't have much pushback so its fine for its purpose. Keep at it. Peter

[dimmaz88]

My machine is to cut polystyrene Peter, so I imagine even deep cuts would give less resistance than clay.

I'm so indecisive it's untrue, I just need to pull the trigger on which material I'm to build the frame from. Maybe aluminium bolted/riveted is the way to go.

[peteeng]

Hi Dim - Personally I'd go plywood. I've just been thru a costing exercise and ply came out cheapest. I pay $35sqm for 17mm F17 ply and 3mm AL will cost me $86sqm. 17mm ply is equivalent to 10mm thick al in bending. Its structurally damp and easy to work with... It will make a great machine for cutting polystyrene. Seal it with good paint or epoxy and its stable. 17mm ply is 12kg/m2 and 10mm al is 27kg/m2 so its structure is lighter... Have a think.

You have a lot to sort in terms of CAM and operation, get a quick ply machine running (even MDF its half the price of ply) sort the electronics, logistics, CAM then design the next machine in metal... or not if it does the job. Your first machine always has bugs ....Peter

[dimmaz88]

I am very tempted by ply Peter, I just wonder how I'd flatten the ways for the linear rails on a column that might be 600 x 600 x 2000mm. How did you surface your machine for the rails?

[peteeng]

Hi Dim - I built the surface referencing a straight edge as I went. Then I lapped the surface with a long lap. You will have the same issues with metal, how are you going to create flat surfaces in metal? Ply is easier to lap then metal... Peter

lap 1 - long board on top of walls
lap 2 - again long board - keep checking with straight edge as you go
Scoot 1 - Result - In this case I epoxied a 10mm thick aluminium plate to the top of the wall after lapping but I have just used the timber surface "as is" on others
Scoot 2 - this machine cuts aluminium easily and is very strong. Even though its MDF. When setting it up there was a fault in the controller and one Y axis went one way the other A axis went the other. The machine visually twisted and one sides feet left the bench. The ballscrews can supply 200kgf of force the load must have been huge. But it survived and has not shown any issues in its 5 year life. About to retire it building a 8x4 machine in ply this time.... The mdf is screwed and glued together with epoxy. Roughen the melamine before glueing if you use it....

Scoot now has a 1.5kW ER16 spindle on it and it chews through material really well...

https://www.facebook.com/ScootCNC/vi...4304636450486/ making spoons 10mm bit

[peteeng]

Hi Dim - I usually use 17GPa for the bending modulus of F17 ply. I just measured its modulus using a beam test and it tested at E=23GPa. Thats really stiff for its weight.... Plys really good stuff.. Peter

[dimmaz88]

You may have finally convinced me down the plywood route Peter, I hope I don't find out you own a construction grade plywood company here in the UK ??

I think I may have to make a suitably long straightedge, as they're ridiculously expensive at over 2m. I could also use a deep profile aluminium box section, I think that'll be close enough for a machine like this. When bolting down your rails did you use a torque wrench? I imagine it's easy to bow them if you tighten one bolt more than the next, as the plywood can squish easier than harder materials.

Fresh excitement anyway! Material chosen, let's get a working drawing done ??