Our company produces 100’s of polyurethane pieceparts per month using silicone tooling, which we create from stereolithography models or machined patterns. We utilise high-end rapid prototyping technology and CNC machining to produce accurate masters for tooling.
Our SLA (stereolithography) machine is a 250 which means it has a build cube (area) of 250mm x 250mm x 250mm. However, the Z height actually only builds to about 240mm. The Z height is the length of build vertically.
Here is a video of our SLA machine:
It is constraints like this that make each project we look at a different challenge as there are several routes that you can choose to go down when looking at building parts using rapid prototyping technology. For example, some parts are better built using selective laser sintering, (SLS) because this process is more economical for low volume parts if cosmetic finish is not important. SLS machines build parts from granules of nylon, which are sintered by a laser, and this is why they have a rough and somewhat porous surface finish. The loose powder acts as a support for the parts being sintered so parts can be stacked on top of each other, which means the build is more economical. By contrast SLA machines you have only the space on the horizontal platform to fit the parts because SLA machines build parts using a vat of liquid. A laser curing off the material layer by layer builds SLA’s. The laser is programmed to trace out the horizontal slice of the part so as to build the parts shape and features. We have diagrams of how the processes work on our web site at www.martello.co.uk/rapid_prototyping.htm
When building rapid prototypes we often find that there are issues over the size of features and wall thickness that can be produced using the current technology. This is due to limitations such as the layer thickness, which is normally 0.1mm and the beam width of the laser on the machine.
Our machine’s beam width is .22mm, which means features sub this measurement would not form in the build. The issue is a lot of parts which are rapid prototyped are destined for injection moulding which can form a lot thinner walls and features than traditional rapid prototyping. However, the latest RP machines take the resolution of builds into new dimensions. SLA technology can go to 0.05mm layering on Viper machines and there are also Envisiontec machines with fine resolution layering which have found a niche in the jewellery market. The Objet Eden machine is also good for fine features and their latest machine (Connex) can now build in 2 materials.
Materials are also very important in deciding which process to use. SLS machines offer a choice of nylons. There is the standard Duraform Polyamide material, which means is also the cheapest as it is the most commonly run on the machines and then there is 30% glass filled nylon which is more rigid. A more flexible material is the newly release Duraform EX plastic (released in 2007) which comes in self-coloured black or natural (off white.) EX can be used for living hinges and snap fit designs.
There are also many different SLA materials. There is only one material on the market at the moment, which can withstand high temperatures and that is Bluestone. Another good material is si50, which comes in either grey or blue and hence looks more like a true thermoplastic even though it is no stronger than most other SLA materials. Our in house machine runs DSM Somos watershed 11110 resin, which is transparent with a green tinge but when, lacquered becomes clear. A major development in the RP world recently is the launch of the Connex 500 by Objet, which can build using two different materials at the same time so you can build rubber features into a rigid part in one build.
Tuesday, 20 May 2008
Thursday, 15 May 2008
SLS or SLA?
There are quite a few things to consider when deciding which rapid prototyping process to use and these are mainly to do with the way each different process works and the materials they use.
SLS machines are different from SLA machines in that they use a powerful laser to fuse together granules of material. The loose powder acts as a support for the parts, which means that no support structures have to be removed, and parts to be built can be stacked on top of each other filling the cube. The result is that SLS parts are generally cheaper than SLA parts due to the economies of scales from the build process. The materials SLS machines use can be either nylon or metal depending on the type of machine and there are different types of materials within those groups. Regarding nylon, you can have standard Duraform Polyamide material, which is tough with some flexibility, or if you need something more rigid there is 30% glass filled nylon. A new material on the market is Duraform EX plastic, which comes in black or natural (off white.) This material is more flexible than PA and is good for living hinges and snap fit designs. The self-coloured black version is also useful if you need coloured parts without the cost of painting and finishing. One of the drawbacks with SLS is the porosity of the parts due to the powder-based material. It gives the parts a rough, granular surface finish, which, granted can be rubbed down and painted but this takes hours and is expensive to do well. This is where SLA parts should be the choice if you need cosmetically superior finishes on your parts or if you need to go onto make vacuum castings. However, the choices don’t stop there! There are many different types of SLA material and also machines. Machines can come in 250mm x 250mm x 250mm sizes (as our machine at Martello) and also huge mammoth sizes (Materialise.) and in-between. Stereolithography as a process has proved itself to be the most effective in building accurate parts with good surface finish and durable enough to be used as masters for tooling. Masters need to be finished for vacuum casting and SLA’s are by far the best parts to work on. Other processes such as FDM parts are very difficult when it comes to finishing, as the parts tend to de-laminate. This adds huge amounts of extra time and cost onto the job, which is no good for us as a bureau building masters for tooling!
SLA materials come in quite a few varieties all with certain advantages. The watersheds are great for holding their shape, which was an issue with SLA’s originally and can be lacquered to produce clear parts. Si50 is a good resin, which comes in grey or blue and produces very plastic looking parts but is not any stronger than the watershed resins. More technical information is on our web site on the datasheets page.
The thing to remember is that there will be a process by which you can get your parts built cost effectively and accurately and the best thing to do is ask around for some advice if you’re not sure. We have referred many projects onto other companies because we recognised that our polyurethane resin vacuum casting process wasn’t the most suitable for certain parts. For instance, very thin parts without features such as ribs or bosses parts maybe cheaper using vacuum forming or thick walled large mouldings maybe be better using RIM (resin injection moulding.)
SLS machines are different from SLA machines in that they use a powerful laser to fuse together granules of material. The loose powder acts as a support for the parts, which means that no support structures have to be removed, and parts to be built can be stacked on top of each other filling the cube. The result is that SLS parts are generally cheaper than SLA parts due to the economies of scales from the build process. The materials SLS machines use can be either nylon or metal depending on the type of machine and there are different types of materials within those groups. Regarding nylon, you can have standard Duraform Polyamide material, which is tough with some flexibility, or if you need something more rigid there is 30% glass filled nylon. A new material on the market is Duraform EX plastic, which comes in black or natural (off white.) This material is more flexible than PA and is good for living hinges and snap fit designs. The self-coloured black version is also useful if you need coloured parts without the cost of painting and finishing. One of the drawbacks with SLS is the porosity of the parts due to the powder-based material. It gives the parts a rough, granular surface finish, which, granted can be rubbed down and painted but this takes hours and is expensive to do well. This is where SLA parts should be the choice if you need cosmetically superior finishes on your parts or if you need to go onto make vacuum castings. However, the choices don’t stop there! There are many different types of SLA material and also machines. Machines can come in 250mm x 250mm x 250mm sizes (as our machine at Martello) and also huge mammoth sizes (Materialise.) and in-between. Stereolithography as a process has proved itself to be the most effective in building accurate parts with good surface finish and durable enough to be used as masters for tooling. Masters need to be finished for vacuum casting and SLA’s are by far the best parts to work on. Other processes such as FDM parts are very difficult when it comes to finishing, as the parts tend to de-laminate. This adds huge amounts of extra time and cost onto the job, which is no good for us as a bureau building masters for tooling!
SLA materials come in quite a few varieties all with certain advantages. The watersheds are great for holding their shape, which was an issue with SLA’s originally and can be lacquered to produce clear parts. Si50 is a good resin, which comes in grey or blue and produces very plastic looking parts but is not any stronger than the watershed resins. More technical information is on our web site on the datasheets page.
The thing to remember is that there will be a process by which you can get your parts built cost effectively and accurately and the best thing to do is ask around for some advice if you’re not sure. We have referred many projects onto other companies because we recognised that our polyurethane resin vacuum casting process wasn’t the most suitable for certain parts. For instance, very thin parts without features such as ribs or bosses parts maybe cheaper using vacuum forming or thick walled large mouldings maybe be better using RIM (resin injection moulding.)
Wednesday, 14 May 2008
Blue Sky Ideas for Rapid Manufacturing
I have been working in the rapid prototyping industry now for nearly 6 years and have really enjoyed the challenges facing today’s technology, but what of tomorrows?
With web applications becoming ever more sophisticated and the advent of rapid manufacturing on the horizon I foresee a convergence of these processes to provide the customer with bespoke design and production solutions straight from their desktop/laptop or even PDA.
This may seem far-fetched but I believe that it will soon be possible for end user customers to make the decisions on how their product is going to look and behave in a way that is tailored just for them. This could be achieved with the integration of virtual reality or 3D web/CAD design and rapid manufacturing technology. With an efficient back end systems bespoke orders could be taken from the web and created in days in many numbers. I can see that there would have to be a template product with basic functionality but then add-ons could be chosen by the user for functional features, colour, and size could also vary.
I would welcome other peoples thoughts on this issue.
With web applications becoming ever more sophisticated and the advent of rapid manufacturing on the horizon I foresee a convergence of these processes to provide the customer with bespoke design and production solutions straight from their desktop/laptop or even PDA.
This may seem far-fetched but I believe that it will soon be possible for end user customers to make the decisions on how their product is going to look and behave in a way that is tailored just for them. This could be achieved with the integration of virtual reality or 3D web/CAD design and rapid manufacturing technology. With an efficient back end systems bespoke orders could be taken from the web and created in days in many numbers. I can see that there would have to be a template product with basic functionality but then add-ons could be chosen by the user for functional features, colour, and size could also vary.
I would welcome other peoples thoughts on this issue.
Subscribe to:
Posts (Atom)
