Benjamin Foster, Freelance Industrial Designer and Manufacturing Engineer (Australia)

The nice thing about an industrial designer is that you acquire skills that allow you to work just about anywhere working on just about anything! You have a very valuable and transferable skillset that’s only limited by your imagination to use it.

Ben graduated with a MEeng in Product Design and Manufacturing Engineering from the University of Loughborough and you can contact Ben by visiting his LinkedIn profile found at http://au.linkedin.com/pub/ben-foster/13/3a7/b91

Have you always been interested in design and building widgets or is something you discovered while looking at your career options in high school?

I first used a lathe aged 7 at my Granddad’s place, was self teaching myself in CAD on a 16mhz Apple mac when I was 11 and have pretty much been toying with all sorts of machines all my life making different things… so I guess I must have had a product design ethic before I can even remember.

If there’s such a thing, I find free-style CAD with a vague brief and with intent to either CNC or 3D print the most fun. I’m waiting for the day when I finally own one of these machines so I can start making all sorts of stuff without incurring too many costs. I do also like tinkering with 3D scanning but that’s more or less on hold as I all wait for more efficient software, higher resolution cameras and faster computers to hit the market. If I really had it my own way, I’d employ a small army of minions to work on all the projects in pretty much everything that crosses my mind.

1st is a solidworks concept CAD model of a 5cm long plastic toy designed for plastic injection. Made up of around 15 components and designed to be spread out on a Sprue for a 2 part mould it includes all draft angles, split lines and nominal thicknesses.

Not particularly actually. I used a z650 and have a friend who took a look at the z850 and they were quite alike. The z450 and z350 are almost identical (bar the size to the z650), the z510 and z410 were basically the same too only without the depowdering station and fancy ink loading cartridges. Having seen the insides of these machines you can see the evolution of parts over time but take apart one once and you could essentially take apart all the rest in he series. Even the $750,000 SLS machine I worked on for a year was quite straight forward component and mechanics-wise though with a price tag that high and it not being mine, I never dared to tinker with it.

Tinkering with these machines with no training or license could leave you with an invalidated warranty and expensive repairs. Do it at your own financial risk.

… for certain projects zCorp machines are excellent (like decently built colour vrml files). However to obtain a decent 3D model, let alone a good colour model from a medium to large architectural firm is surprisingly difficult. Your best chance in obtaining good colour 3D models to print is actually from freelancers, engineers, character modellers or students as they often have the flexibility, time, and access to the right software to eventually get it right the first time and to get it right every time there after. In one company I worked for, we processed only 3D prints for zCorp printers. About 80% of clients were architectural, and of all the models we ever printed over one year, only 17% used some colour and maybe 2-3% made the effort of properly texture mapping their model. It’s basically because texture mapping in itself is quite time consuming and not always necessary. Many architects also like nice clean and white models.

There are good and bad points to this range zCorp machines…

– cost is generally per cm3 of material used. Approximately $2-5 dollars I’m guessing. Cheaper quotes basically means either a very large company is using these machines and they have good turnovers or they have concocted their own powders and inks. I think Shapeways do this with their “ceramic” 3D print option (though I’m not 100% sure)
– there’s a lot of machine preparation work and cleaning up.
– depowdering is very time consuming. The newer enclosed zCorp machines are “self contained” but any operator will always get some powder spreading out here and there. As much as they claim it’s a 3D printer for the office, if I were to own one, it’d be in it’s own room.

ZCorp 3d models designed, printed and and built by myself for a client for a motorway proposal/bid. Data was obtained as a point cloud for terrain, 2D plans for motorway section. Everything else I did basically including image mapping.

If I were to own one, I’d also do the following:
– build my own large depowdering station with a laminar flow hood
– install decent lighting and a switch to flick between UV light and a good tungsten light
– use a good shielded vacuum on a semi rigid arm (so it holds in place whilst you are air dusting your model
– build this table at a model maker’s height and with enough space around (their machines are made for people 5ft tall I swear). There would also be space to put your legs underneath if you were sitting on a stool.
– use a decent variable foot and trigger operated air brush with fine nozzles
– have a separate cabinet for glueing and ventilation also with access from at least 3 sides. Sometimes these models need 2 people to work on
– you also have to buy proprietary materials. there are a lot of them and they are expensive for what they are but this is basically the same for most 3D printer manufacturers. The powder which sells at about $1400 for 8kg is more than ~90% plaster, ~5% talc, ~5% polymer & surfactant. I’m not even onto the print heads you churn through and the components that eventually fail from wear and tear… it is an expensive machine and before you consider buying one, I would seriously revise how much client base you actually have, how much that returns and the pay back period. You don’t really need a service contract either but if money is no object then go for it. For newcomers to the scene, get a service contract for sure. Clients want their models fast. If you don’t have your machine up and running the next day, you may loose clients or they may miss their project proposals rendering the $1000 models essentially worthless.
– there are plenty of other things to think about with 3D printers in general and much more than this email can hold. I would be inclined to simply outsource the actual 3D print and charge the client a bit more. That way you don’t have to deal with all the extra complications and time consuming manual labour that could be better spent elsewhere.
– finally, reverting back to the % of models that are made white, outsourcing the model to a nylon SLS printer will give you much better results in most types of prints with the benefit of being cost competitive and being able to drop your model on the floor and not have it shatter into a million pieces. That being said, zCorp powder prints can be better to have in some cases.

New:

Decent zCorp machine new >$50,000 + materials and service contract … so have about $100,000 in your pocket to start with and at least a $50,000-75,000 turnover on just these 3D prints to make this cost effective. You’ll need to employ a person to work about 30-40% full time to meet that return.

Used:

Working zCorp 510 <$10,000 and easily modded to use alternative materials (~$5/kg) and inks. However any broken parts are legacy components and are fairly expensive to replace, print quality isn’t quite what the newer machines offer, you need to build your own depowdering station (which isn’t a bad thing as previously mentioned) and non-proprietary materials need some fine tuning though there’s information on the internet for that. i.e., this is much more affordable and a tinkering type project type. i.e., if you abandon it then it’s not the end of the world. You might find buying 2 machines means you can keep one for spares and learning how it all works.

If you really do want non-sales pitch and work on the field information on zCorp plaster powder printers, EOS SLS nylon, Object resin inkjet, envisiontec SLA then I can give you more info on them. Basic questions if you were to get any printer of any size would be…

– who are your clients and who are they likely to be if you are to find more?
– how much money do you have for this
– how much time do you have or employee time is there available
– do you already have the software to process all this data
– can you pay back the system and all the ancillaries in 1 1/2 to 3 years?
– how does this all compare to simply outsourcing your models to companies that have already got the machine or better than the machine you intend on buying.
– can you plot it all on an excel page?
– have you contacted universities for quotes? they sometimes have the cheapest and little publicized rates (at least in London)