February 10, 2019 at 11:00 am #88479
great project(s). Thanks a lot for sharing!
I’m in the ‘first phase’ of becoming clear about which design to build, and maybe you can give me some advice.
My background is (kite) surf / hydrofoil boards and wings. So far I hand shape the boards (XPS) and print (CR10), sand/filler the moulds for the foil-wings which then get done as composite. I also have a quite large BOSCH gof2000ce professional spindle with which I milled composite contours and coarse-shape XPS. (Note: I do not plan to mill composite on the MPCNC; only XPS foam and possibly some MDF for moulds).
My first idea was to use that spindle, but it is quite large and might be to heavy (6kg).
The other point is I really like the design of the MPCNC with the stainless steel tubes. However, I’d like to go for Y axis of about 1.8m (~59 feet) to be able to mill the boards. Working through MPCNC and lowrider, I came to the conclusion that what I actually would like to do is a combination of both:
Y-axis like the MPCNC, but supported on the bottom – will there be enough space to do so?
X-axis inspired from the lowrider with 2 tubes in parallel, and the spindle in between. For x I’d need about 0.6m effective toolpath.
So what do you think about such a combination? And how about the tolerable spindle weight (I guess the inertia might lead to problems with the stepper motors?)?
Or should I just go for a ‘default spindle, build a large Y MPCNC and support it?
Thanks a lot!
AlexFebruary 10, 2019 at 12:57 pm #88500
You’re looking at roughly 1 by 2 meter mpcnc to get your cutting area. I’m pretty sure that’s doable with just center span supports for the long side. Your router is probably a little too heavy though. The real question is how tall do you need it to be?February 10, 2019 at 9:48 pm #88594
What do you mean by
The real question is how tall do you need it to be?
The router? The complete machine?
AlexFebruary 11, 2019 at 5:15 am #88620
Sorry, the tallest thing you’ll be milling, z depth.February 11, 2019 at 7:30 am #88651
ah, Ok. I think I should not need more than 20cm in Z. However I would like to be prepared for more.
As far as I understand, larger z-height means longer feet and the shear force will bend the feet/top-frame connection.
What about adding a plane of e.g. plywood to all sides. This would support y and stiffen against the shear forces – not?February 11, 2019 at 10:33 am #88686
That would help. The z assembly will also start flexing at that height too.February 11, 2019 at 10:00 pm #88809
Another thought is: keep it small and use a default spindle (allows to stick to the common design).
The background is that I need only 100cm x 50cm x 5cm for the mould (probably done in MDF), but would like to have a good precision on these.
The reduced (usable) Y-length to about 1m would mean that I need to reposition the foam for the boards once per side.
One could e.g. mill 3 reference points which could be used for repositioning.
Would this be worth the fuzz?
And regarding Z-height: I did not really get it: I read that a larger Z is unproblematic at the bottom, bending/precision gets worse when going up. Is this true? By this, one could mill not so tall things (like the moulds) at ‘high’ precision while having the necessary Z height for the foam, where precision is uncritical.
Looking at the pictures (of the mpcnc) I think that going all the way down for flat objects would also lead to bending in the Z assembly – not?
February 12, 2019 at 3:58 am #88818
- This reply was modified 1 week, 4 days ago by Alex.
If I understand you correctly, yes, the MPCNC is stiffer the closer the router bit is to the height of the X and Y rails. The more the gantry gets inserted, the longer the lever arm and the more possible play in the bit positioning. The LowRider, on the other hand, is stiffer at lower z heights.February 12, 2019 at 4:50 am #88821
If I understand you correctly, yes, the MPCNC is stiffer the closer the router bit is to the height of the X and Y rails. The more the gantry gets inserted, the longer the lever arm and the more possible play in the bit positioning. The LowRider, on the other hand, is stiffer at lower z heights.
BT nailed it.February 12, 2019 at 5:22 am #88823
So having larger Z-height (longer feet on the frame) we have two effects:
- bending of the frame which could be counteracted (partly) by adding some planes of plywood as discussed above
- the lever arm is longer and bit positioning will get worse when moving away from the x y rails height -> not so tall workpieces should be elevated on the table, such that the working area gets close to the ‘sweetspot’ Z-height
Did I now get it?
Thanks a lot!
AlexFebruary 12, 2019 at 6:23 am #88826
One other thing to consider is that even though you stiffen the outer frame rails with plywood, you still have the X and Y rails used to hold the gantry which you cannot easily stiffen and over distances longer than 2-3 feet can cause more deflection down (sagging).February 12, 2019 at 8:35 am #88840
One other thing to consider is that even though you stiffen the outer frame rails with plywood, you still have the X and Y rails used to hold the gantry which you cannot easily stiffen and over distances longer than 2-3 feet can cause more deflection down (sagging).
One idea would be to use 2 tubes in parallel at a certain distance (comparable to the lowrider’s x). But adding all these adaptions might make it much more expensive.
So what do you think about building smaller and reposition the larger (foam) workpieces as suggested above? Sounds like the better option to me
From one of the builds using carbon tubes (not a good idea i think) I had the idea: why not insert carbon tubes in the stainless steel. I plan to build the 25mm version, and one can get 23 with 2mm wall carbon tubes. Use some epoxy to glue the in should result in quite stiff gantry rails – what do you think?
Edit2: One usually overestimates the stiffness of carbon based composite. Its the stiffness/weight ration which is good. But the coefficient of elasticity of about 88 GPa is much lower than that of steel 210 GPa. So using a thicker wall for the steel tubes makes more sense.
However doubling the length of the tube gives us l³ =8x bend down for a given weight. Doubling the wall diameter is about linear at our sizes. So we would still end up with a factor of 4 more bend down.
I should have known it, but seeing the numbers makes again clear: shorter axis have an over-proportional effect on precision.
You must be logged in to reply to this topic.