Re: So far OT it's over the horizon: Anyone into 3D modelling?


Catman wrote:
> Rope wrote:
>
> > Catman spoke:
> >> >>> Anyone involved in anything like this, or can give some pointers as
> >> >>> to search terms?
> >> >>
> >> >> http://www.torchmate.com/overview/index2.htm
> >> >>
> >> > Interesting, ta. That might well be ideal
> >>
> >> Except it's not big enough. Cheers
> >
> > Eh?
> > !"Torchmate machines are available in capacities
> > ranging from 4' x 4' to 10' x 40'.!"
>
> Yeah, but sod all depth. Needs to be 2mx2mx2m, shaped in all axes.

and for a tenner a pop???

you have LOTS of problems here.......

1/ x + y + z axes of _any_ precisions use something known as ball
screws, 2.5 metre ball screws have a certain minimum cost, even if you
cut out ALL the middle men you'll be looking at IRO 500 quid per axis
per rail, and you'll get a precision of the order of a thou for
that....

yes, you could use precision threaded bar and a couple of nuts to
eliminate backlash, but it's gonna wear quickly, take more power to
drive, and "finesse-ing" it so it works well is going to put the cost
beyond a cheap ball screw

2/ while the "screw" bit provides movement, that movement needs to
slide along something, so you'll need some sort of rails, again you can
use the pukka stuff and spend about 500 quid an axis and not destroy
your precision, or you can try and DIY from stock sections and spend
the same money getting it to work and destroy your precision...

any machine with axes of the order of 2.5 metres is going to have
issues with thermal expansion affecting precision

3/ motive power, cheapest option here is stepper motor and driver
boards, your machine is fairly beefy so these need to be fairly beefy
too, say 100 quid a stepper and up to the same again for the driver
circuitry

only gotcha here is stepper is no feedback, eg "move x amount"
instruction is sent out, there better be enough servo power to overcome
any loads induced or that movement won't happen, but the machine will
simply carry on as if it has.

you can get around this with feedback loops and servos for a closes
loop system, but lots more money an not necessary as long as you don't
try to "drive" it too hard

4/ control, this is the easiest bit, any old PC and a copy of EMC which
is linux with real time extentions designed especially to control CNC
machinery

5/ "work" material and "tool" is difficult, say your material is
aluminium of some grade, a one horespower tool driving motor driving
say a 6mm end mill is gonna take FOREVER to chew through a 2 metre a
size block or work, you want faster production then before you know it
you have a 40 bhp 3 phase motor, and your gantry just had to get 3
orders of magnitude heavier to accomodate it, and so on and so forth...

I can put you in touch with people who have made 2 metre by 2 metre by
not that much machines to do things like router wooden doors and
panneling, which will get you part way there, but....

off hand the only materials I can see you getting and man handling in 6
cubic metre blocks are "man made" foams like polystyrene, the "oasis"
stuff flower shops use, and stuff like that

right away your precision is down to something like 1mm

the size makes me thing you're trying to model something like
motorcycles, life size?

will you want to paint them when done? if so this affects material
choice too...

certainly a tenner a pop means you aren't going to be using anything
more structural than something like a ploystyrene compound

you could go the "add" instead of "remove" material route with a nozzle
instead of a router but that is going to slow you down even more and
make even 1mm precision real hard ("hard" is another way of saying
"expensive and / or slow") to do

let's assume you are wanting to make motorcycles, accurate
dimensionally to 1mm, to be finished by painting...

a/ I'd say you'd be right to be very pleased with yourself to build the
CNC machine for 5000 pounds excluding work head, whether this be router
or extrusion or hot wire or whatever

(you prolly have 200 hours work assembling such a machine and setting
up too)

b/ I'd say you'll be looking at making your own blanks on site from
some sort of foaming plastic, so blanks are prolly gonna cost 30 quid a
pop

c/ I'd say you make about 5 models before you figure out how to make
one properly, just pray this doesn't entail a fundamental machine
redesign, I mean here mebbe cut two female half moulds (if the model
has appropriate symmetry) and vaccum from a thermoplastic in there

d/ I'd say your finished models are still going to come out nearer 100
quid a pop in materials than ten, and they will still need extensive
hand finishing, and labour ain't free

e/ I'd say you might get one model in 24 hours of machine time (you
aren't going to get any "rapids" with axis movements when you've
mastered "roughing" and so on and so forth, so that's 24 hours of
machine attendant watching it.

f/ NOWHERE in this have we gone anywhere near ancilliaries, such a
compressed air of vacuum for chip removal, lubrication, which all
requires significant amounts of power, thermal management, which can be
a bitch and very expensive, and most overlooked, power consumed by the
machine itself and machine maintenance in between runs, which could
scare the crap out of you

------------------------------------------

Don't let me convince you what you want cannot be done, it can be, but
commercially it is vastly expensive, so if you want to cut the cost
down you'd better be prepared for backbreaking work and ingenuity and
sacrificing a lot of functionality here and there, and approaching a
compromise solution, if you do not accept this them you will fail and
waste a lot of time and money doing it, and be turned off the whole
thing CNC wise, which would be a shame.


HTH etc

.

Relevant Pages

  • Re: So far OT its over the horizon: Anyone into 3D modelling?
    ... > screws, 2.5 metre ball screws have a certain minimum cost, even if you ... > cut out ALL the middle men you'll be looking at IRO 500 quid per axis ... you could use precision threaded bar and a couple of nuts to ... > eliminate backlash, but it's gonna wear quickly, take more power to ...
    (uk.rec.motorcycles)
  • Re: My Homework: how to calculate natural log, e?
    ... > So GT's claim of 1ulp precision are demonstrably false. ... with pwr equal to a power of two) throws away a lot ... couple of hundred ULPs when the input is around 700. ... no deficiencies and the other way is to make it so complicated ...
    (comp.lang.fortran)
  • Re: For those who missed it -another PC milestone
    ... The processing power of the GPU is now available for general purpose ... do you mean "single and double precision" or "single or double ...
    (uk.politics.misc)
  • Re: What sports car are YOU?
    ... > You are a Ferrari 360 Modena! ... power, passion, precision, and style. ...
    (rec.sport.football.college)
Loading