Openmoko Beagle Hybrid

[Joachim Steiger]

Dr. H. Nikolaus Schaller wrote:
> The problem is not technology or DIY capabilities, but cost.
> 
> What we want to have is a nice case achievable for everybody, not only  
> the enthusiast who wants to spend time and money for experimenting  
> with DIY hardware or commercial FDM.
> 
> So the question is how much does a SW developer want to pay to get HW  
> + Case? Let's say 50 EUR per plastic case.
> 
> FDM is at least 200 EUR (that is what we got as a quotation from the  
> rapid-prototyping shops for a simple part and not the whole case). Or  
> 700 EUR for a Cupcake. Or 5k for a protomold made thing. Or 10-20k EUR  
> for a 3D printer. A full freerunner case consists of 6 plastic parts  
> (incl. 2 buttons).
> 
> The other side is expectation of quality/robustness. I have been told  
> by experts who own a RepRap/CupCake that the precision is not good  
> enough to reproduce a Freerunner case (wall thickness 0.5mm).

true

also its much too complex.
i tried importing the 3d models into quite a lot of the free and or open
 3d and machining tools, but the shear amount of detail seems to be a
problem there.

also there are limitations of what you can do with which each
production-method:
* e.g. for reprap-alikes, all overhangs > 45deg need support structures.
* milling in 3axis means you can only 'mill from e.g. above'.. to turn
it to the side you already need a trick/mechanical help to mount it
sideways, without loosing alignment, or a 4 or 5 axis mill (i don't
think there is any free toolpath-gen for that yet)
* laser cutting heavily depend on used materials and is basically '2d
only' for the affordable machines (<50keuro)
this means designs consist out of 2d shapes.
one 'stacks' afterwards or uses creative mounting methods to hold the
shaped sheets together, like e.g. on the cupcake-cnc (makerbot)

> So if we find a method that allows to make 10 units from a budget of  
> 500 EUR or 100 units from a total budget of 5000 EUR I am happy!

tricky.

we got a cnc mill (3axis, 800W spindle)
as well as a simple lasercutter (50W) here in berlin in our hackspace.
there is also a rep-rap-like printing head for thermoplastics, but thats
not completely ready yet.

the much bigger problem than machining itself, is getting a the design done.
after that one needs to get the toolpath generated.
special sw as well as expertise in that line of work is what it makes so
expensive.
milling itself isn't a very cheap form to 'produce something'. but
still, its not the time the machine is running but the worktime of the
human which makes it expensive.

if somebody has too much free time and wanna try this, check out
http://camgeeks.de/ and visit us there ;)

free and/or open tools for mechanical engineering are still not quite
'done' (yet), but there is progress.
Still they work well when you learned about their limits (or even extend
them).


ps: what about finding some 'ready made universal case' like from teko
or boppla and do some cnc coutouts for the sockets? thats much easier.

-- 

roh