View Full Version : stability - monolith vs. two-part assembly

15-10-2012, 04:21 PM

I'm at the last stage of design and can't make up my mind which assembly is better:


Left one: two up-side-down L shaped elements with a few rectangle pieces wedged between them

Right one: two separate 'boxes' bolted together to form the same shape

The bit sticking out on the top will be a rotating shaft (low speeds around 100RPM, the bottom part of the shaft is not shown) and I'd like that to be as steady as possible. The Z assembly will be sitting on the front with rails attached to the shown supports. Everything will be made from 15mm plywood for the prototype.

Any suggestions are much appreciated.


Lee Roberts
15-10-2012, 10:00 PM
Parts marked in red would become one part in the shape of a T, the L shape sides could then have a recess/pockets on the inside of the machine for the T shape to fit into and also allow it to be fixed from the outside of the machine. This should help any nuisances at the bottom of the spindle to be absorbed by more of the machine (L shape sides).

Front part marked in yellow, extened this to the full width of the head/machine and this will move your lines to the side view of the machine (aesthetics).


Lee Roberts
15-10-2012, 10:23 PM
Just another thought while looking, is there any need for the top of the spindle to be on show?

17-10-2012, 11:13 AM
Hi Lee,

thanks again for the reply. Here's an updated version of the drawing, showing the internals and the T section you mentioned:


As you can see, I've put slots in the sides to match the T section and make it more stable. I've added two 47mm holes to press the tapered roller bearings into, but because the bearings are normally around 15mm thick, I had to leave them sticking out a bit, as the hole is only 10mm deep. This leaves only 5mm of material under the bearing, so I'm thinking of adding another 15mm plate under the bearing, so support it better. There's also a hole for the drive screw (which will be a simple m10 threaded rod for now) which will sit in Oilite bushes, as it won't move a lot.

To answer your question, the spindle sticks out as I will try to drive the thing manually, so I need a way of adding a handle of some sort to the top. In the final version there will be a pulley mounted on the spindle, hidden inside the top section, with a toothed wheel driven by a motor, mounted in the 'pillar' section of the chassis.


18-10-2012, 12:35 PM
Only one question... how are you going to mill the rear slot? It needs square corners but you'll be cutting it with a round tool... unless you make it a tool diameter wider, or round off the corners of the mating part it won't fit...

18-10-2012, 01:30 PM
Indeed, I've realised that (it's something obvious if you've ever used a CNC, says something about me I guess) and came out with this to allow for the boards to match:


it's simply a cut into the corner until the tool bit touches the 'corner' of the original square, which takes the corner off, but leaves the sides intact. I'm sure this is an often used cut on squares.


18-10-2012, 09:33 PM
yep, thats one solution if you have the room... and its not on show :)

24-10-2012, 03:03 PM
Most of the nasty looking stuff will stay inside the box, so as long as it works, I don't really mind:)

I've re-designed the whole chassis using 18mm elements, went with the turned around L shape approach, but I'm not so sure about the z-axis assembly support. I'm using cross dowels as the box is going to be made from MDF and the previous approach is not such a good idea, as I'm struggling a bit with where to put the bolts and nuts. Here's the two ideas compared, the left one is the original one, using triangle support, the right one simply uses M8 cross dowels, with 80mm bolts going through the plate:


The load on the assembly is probably not going to go above 1kg, but I'm not sure how MDF takes bolts through the middle. The plate with the big hole has to be perpendicular to the rails, so I'm looking for most stiffness. Any tips / suggestions are more than welcome.