1st Machine, Steel ,Overhead Gantry

[bradodarb]

After countless hours lurking and learning, I am finally making some progress on my machine. It will have an envelope of about 40x30x16. I plan on mating it with a base that has a three sided column to allow for three different work tables. One will be a vaccum/t-slot flat table for panel work, one will have an 'A' axis rotary with a tailstock and the last will have a set of feed rollers parallel with 'Z' that will supplement the 'X' axis for running long lengths of carved moulding.

It is constructed from 1/4" wall steel tube, 1/2" and 3/4" plate. I have the main frame and the X carriage built and hope to finish the rest soon so I can get the components stress relieved. I have all the ABBA linear rails (20mm), and am researching good deals for ballscrews. Was considering jumbo steppers, but now am pretty sure I want 400w-1kw servos. Probably gecko drives and definitely EMC.


The machine will be used almost exclusively for carving, thats what I do for a living (CNC, not so much by hand). There is nothing quite like three or more axes interpolating to reproduce a model you have spent hours/days on.
I can't wait until I no longer have to be at work to get my fix.

[Mike F]

Just a quick observation if I may. In the second photo (side view) it appears the Z axis protrudes quite a long way in front of the carriages carrying the gantry. With a router/spindle added as well, there is going to be quite a large turning moment around the front carriages, effectively trying to raise the rear carriages off the track. Is it not possible to re-design the gantry so that the centre of gravity of the whole gantry lies between the carriages?

Just a thought.

Mike

[bigz1]

I agree with Mike's comments. But also have you considered putting another pair rails a 1/3 of the way up to stiffen the gantry? Also think the Y axis gantry rails could do with more support.

Liam

[bradodarb]

Blanket statement:


Previously posted pictures are not complete and do not show ballscrew/motor assemblies or final support gussets.




"Just a quick observation if I may. In the second photo (side view) it appears the Z axis protrudes quite a long way in front of the carriages carrying the gantry. With a router/spindle added as well, there is going to be quite a large turning moment around the front carriages, effectively trying to raise the rear carriages off the track. Is it not possible to re-design the gantry so that the centre of gravity of the whole gantry lies between the carriages?"

I understand what you are saying, I put some considerable thought to that as well. The further the spindle is from the center of the carriage the more torque on the gantry support and more power needed to overcome the leverage on the Z X plane. Another downfall to this design is that it really decreases the travel of the X axis.

My only excuse for going with this configuration is that it affords me about 4 more inches of precious Z. I have also rationalized the shortcomings of this design by observations of the industrial machines at my shop. I realize that these machines are very massive in comparison with mine, but they all have considerable catilevers within their design, a few of them even have their gantry as a cantilever.

I'm not trying to come off as a know-it-all, if you could illustrate to me a way of centering the spindle without giving up Z, i'm all ears. I will try some reconfiguring as well.




"effectively trying to raise the rear carriages off the track." The rails I am using (THK stlye recirculating prismatic ways) should support loads equally well in your scenario. -At least that is what I gathered from the documentation, I could be wrong.





" But also have you considered putting another pair rails a 1/3 of the way up to stiffen the gantry? Also think the Y axis gantry rails could do with more support."

You are absolutely right, Liam. The Y does need more support, the whole thing needs and will receive more reinforcement. I just need to settle on a ballscrew/motor arrangement so that I may place them confidently.


Incidently, I did run the assembly thru Cosmos with applied loads four times what she'll ever see and had no appreciable deflection, nor pressures exceeding the linear hardwares spec's. I know things on paper don't always duplicate well in reality, hope it works out in real life as well as it does in solidworks.

[Rodm1954]

Quote
if you could illustrate to me a way of centering the spindle without giving up Z, i'm all ears.
Unquote

You could extend your X axis frame on the top to be outside the tubular box frame. The vertical plates that connect X and Y can then be placed on the outside of the tubular box frame. The connection between these plates can then substitute for the X axis framework above the tubular box frame.

Z is moved back and is then free of obstuction to travel upwards.

This modification would mean that you cannot add another vertical to the long sides of the tubular box frame as it would obstruct X axis travel. Perhaps just strengthening the existing design of the box frame will be sufficient.

Placement of a single ballscrew for X might have you back to the drawing board.