No flame! I could not agree with you more about the time advantages of quoting
and building molds from a 3D model (if it includes draft and parting line
locations as you stated). A few short years ago I was one of the "unbelievers"
and did not think that it was all that practical to go directly from model to
mold without creating 2D drawings of all the mold details. I was wrong. As you
point out the Autocad parts won't always go together, the Pro/E parts usually do.
Another recent mailing I completely agree with was from Todd Schuetts pointing
out how fast the newest CNC machines and CNC software is reducing the time to
make accurate hard tooling. Many customers in rapidly moving markets like laptop
computers and cellular phones need dozens or hundreds of accurate very thin wall
parts in the actual production material for shows, field trials, product testing,
limited production, etc. The usual RP processes don't work well here.
I think the RP future for these markets will be the desktop concept modeller
sitting next to the designer for instant feedback as the part is designed. This
will be followed by rapid hard tooling (mainly by CNC and EDM) for production
quantities within a few weeks.
Anything else, such as SLAs to make silicone molds for limited numbers of
simulated parts in materials that really are not the same as production will be
too slow for many markets. The various systems to make rapid molds (Keltool,
DTM, MIT, Sandia Labs, etc.) are probably not going to give the finish and
accuracy available from a good CNC/EDM approach. For major markets any extra
cost of throwing away a incorrectly designed prototype mold is more than offset
by the improvement in overall time to market.
A major focus in the future is going to be rapid testing. It frequently takes
longer to prove that a design works and meets its required specs than it takes to
design and build the testable parts! Fortunately the capability and usability of
non-linear FEA and mechanical simulation modelling is improving rapidly. The
first design has a much better chance of working to specification, which again
reduces the need for prototypes.
> Several years ago when people sent out for mold quotes providing 2D drawings
> to build molds from, it was beneficial to have SLA parts along with the
> quotes. However it has been a couple years since we have tried to quote any
> jobs at all from 2D drawings. Most companies have stopped producing complete
> 2D drawings of their parts ( a very large waste of time). Instead 3D models
> are produced and 2D reference drawings are made to show engraving, mold
> texture, design notes, and misc info which can't be represented in the model.
> You could not build a mold from the information on these prints. We quote a
> couple thousand jobs a year and 99% of the jobs come to us as 3D solid or
> surface files. If we do recieve 2D only data we are normally leary and pass
> on the job until they can provide 3D data to us. The mold industry standard
> today is ready to go ProE files with all the Drafts and parting lines
> defined. If the files are anything short of that, it's time and material to
> bring the files to that level before we even start talking about a mold. All
> molds are firm quoted to a set price and delivery. We cannot afford to play
> with peoples half finished Autocad part drawings. We have been burned by that
> so many times in the past it is no longer funny. This is the reality of the
> production mold building business today. Im sure there are a few small shops
> out there still working off 2D prints but they need to get serious if they
> want to survive. The only other notable exception is the Japanese who are
> admittedly a few years behind the US and Europe in this area. I would rather
> someone Email me an STL file to quote from rather than a part or a print. I
> can do much more with it. I know im going to get flamed big time for this, so
> everyone LET ME HAVE IT.
> Bob Morton
> Fusion Engineering
> For more information about the rp-ml, see http://ltk.hut.fi/rp-ml/
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 - 22:51:33 EEST