Will & David,
Sorry it's taken a while to reply, but I've been busy. As I am a grad student I don't have too much money in the market, but I am interested and have invested (modestly) in this sector as I have direct experience with it and I think it is growing.
My thoughts on 3D vs. Stratasys. I think FDM is better when you are looking for 1 (one!) functional prototype. In all other cases I think either the Actua or the SLA has the advantage. Why? Because if you want more than one functional prototype (say 10) then it is still more economical and less time consuming to build RTV rubber molds off a single master and cast urethane models (this is a service that Compression, Prototype Express and others service bureaus offer along with RP). So making a more accurate master pays off tenfold (SLA wins). If your not looking for functional prototypes but simply visual models then the Actua is many times faster, which makes a big difference when building a large part (say a desktop phone).
I agree that SSYS's machines are simpler, cost, less, and can be more easily integrated into an office environment than 3D's SLA, but I think the Actua will be able to compete on all of these fronts. I also believe the Actua's jetting technology has more potential for greater speed and accuracy than SSYS's extrusion.
I saw the Genisis(sp?) at the Gwaltney Symposium here at Georgia Tech. I wasn't really impressed. They had some problems getting it to run properly and it seemed to take forever to build a part. Since this is a new product I can forgive some bug difficulties but I don't see any real advantages over FDM. I'm not sure about pricing but I believe all of teh office type modelers are between 50-70k (with the possible exception of Sanders which is too slow to be seriously considered for most apps).
Depending how you use the word accuracy it may encompass several distinct parameters. The most prominent difference between SLA and FDM is surface finish. To aggravate that difference, it is much more difficult to sand an FDM model vs. an SLA. I don't have any data on dimensional accuracy in front of me but I would guess that for most builds (and it does matter how and what you build) SLA is within +-.010" while FDM is within +_.015-.020". For many users the surface finish is more important. To get a real feel for that you really need to see some parts and keep in mind what, if any, post processing (sanding, etc.) was done.
My lab, the RPMI, is primarily concerned with using these technologies to get prototypes of small quantities (10-100) in the end materials. We use RP technologies primarily as a tool and not as an object of research themselves. My work has been in the creation of Rapid Tooling for injection molding. Using some of the processes we've developed we can go from CAD file to 150 plastic parts in a week. Since accuracy and good surface finish are prerequsites for injection molding we don't see the FDM machine helping us towards that goal.
I do plan to use the FDM to model a plastic housing for a battery powered sander. Using FDM I can make the housing directly and put in a motor, batteries, etc. Doing the same out of SLA would have led to a less durable part. SLA resin is actually pretty strong, it is just brittle. This brittleness is what makes it hard to make usable models for plastic housings and snap fits, etc. If Ciba or whoever ever gets a more flexible resin then that's a big chip out of FDM's advantage.
Oh Yeah- You can find out some more about the RPMI at our webpage at:
rpmi.marc.gatech.edu
My page is under faculty and staff. The whole page is about to undergo a major revision so please excuse us if it's messy.
-Brian
|
