It already is!
A client approached us around 18 months ago with a very challenging problem.
They needed to produce small volumes (20 per year) thin walled domes
(bounding volume 1.5m cubed), which have an array of very complex features
over the entire surface.
Historically these domes were machined from solid lumps of aluminium, a
process which was very expensive (over one hundred thousand pounds
sterling), wasted an obscene amount of material and took many many weeks.
They asked us to look at the problem and come up with a cheaper / quicker
route to the final part.
We first examined the functionality and discovered that the domes did not
need to be solid metal, but could be a metal plated polymer.
The geometry of the part meant that moulding the parts was not an option,
and so at first sight it seemed we were back to machining.
However, a thin walled dome with complex but small features over the entire
surface is an excellent candidate for additive manufacture. Far more
material would need to be removed during machining than would need to be
added in layered manufacture (even considering the difference between rate
of removal and rate of addition, the volumes of material concerned make
additive fab a more rapid solution)
In addition the machining cycle was very complex - not an issue for additive
We performed a feasibility study which addressed accuracy/resolution,
material properties and time/cost, in order to determine whether RP could be
a feasible manufacturing technique.
SLA was the only process to give a good enough surface finish.
We built sample sections on our SLA-350 to determine whether we could get
could enough accuracy/resolution on the small features (even at 0.1mm
layers). We could.
The resin alone did not satisfy the performance requirements of the part,
but we found that electroless/electrolytic plating of SL 5190 with a series
of metallic layers (the nature of which I can't reveal) met all performance
requirements. Sorry to be so mysterious but I am bound by confidentiality!
The only issue remaining was whether or not we could compete on speed/cost.
Domes of this size obviously cannot be fabbed as one piece and so we spent
some time splitting the CAD into manageable sections (a job made difficult
by the surface features). Once this was done we found that, by nesting dome
sections, we could build the SLA in around 10 days.
Sections are then reassembled around a machined mandrel and zipped up with a
light pen, before minimal hand finishing.
The plating process takes 2 weeks.
We haven't completed cost / time analysis yet but we have reduced timescales
slightly, reduced manpower enormously, and have reduced cost by about a
factor of 10.
NOW FOR THE BEST BIT!
Once the design team had seen the new parts, and UNDERSTOOD HOW THEY WERE
MADE, they saw the potential for an even better design.
The surface features on the dome are critical to its function. The design
of these features had very much been constrained by design for manufacture
issues, thus compromising functionality to some extent.
BUT NOW additive manufacture allowed the DFM issues to be lifted, and the
design team had far more scope in which to work
The parts we are currently making exhibit features which simply could not be
manufactured by any method apart from additive fabrication. These features
have taken the product into a new range of functionality, and most
importantly have ADDED VALUE.
SLA will now be used as the standard manufacturing technique for these
parts, and having been plated and lacquered they will go straight into
A formal case study will be presented following project completion (subject
to non-disclosure agreements).
I appreciate that this case study deals with small volumes of very complex
parts, and the discussion being initiated here addresses whether RP could be
used in larger volume manufacture. However, this case study does illustrate
the very important point that RP as a manufacturing technique is already
improving product functionality.
Dr Matt Murphy
Head of Practice - Rapid Manufacturing
Tel: 44 (0) 1664 501 501
Fax: 44 (0) 1664 501 589
For more information about the rp-ml, see http://ltk.hut.fi/rp-ml/
This archive was generated by hypermail 2.1.2 : Tue Jun 05 2001 - 22:46:11 EEST