CAD on the go
17 March 2009
Process type: Hardware
As a man constantly on the road, Rob Jamieson knows a thing or two about mobile working. Stopping short of strapping a workstation to a trolley, he shares his top tips for CAD on the go
For ultimate power on the job then a desktop workstation is the only way to go, but if you are like me and need power on the move you have to make some compromises. But exactly which compromises do you make to fit the needs of the job in hand?
If it’s just email on the move you can use one of the myriad of blackberry type devices or now a netbook. I use a Blackberry every day but for viewing of larger files and PDFs I find it hard to see. Netbooks are good as they have a full keyboard and bigger screens, but as they generally use low power CPUs they are fine for web-based mail but full applications can tax them.
Viewing or using full 3D or even 2D applications can push them to the edge, but soon to be released updated versions will feature faster CPUs and graphics. If you think about it, PDAs died out because phones became more featured but CAD viewers have not really moved to general phones. The cost point for netbooks has also been quite low which has defined a new market space and feature set. Because of the price point you tend to find Linux is a more popular operating system but it can be harder to find CAD viewer software for. I have an older netbook I use for when I’m travelling light, on which I can use PowerPoint but its power for 3D or photo editing is limited. As a result my preferred option is a workstation laptop.
One time I got out a 17-inch mobile workstation on a plane and half of my laptop was on my neighbouring passenger’s tray. If ever there was a case for me to fly business…
Workstation laptops come in three varieties: 14-inch, 15-inch and 17-inch. This is based on screen size, but resolutions can change. There are not many 14-inch laptops available and they have fairly low-resolution screens and because they have the smallest form factor they are restricted on CPU and GPU (Graphics Processing Unit) power. They are fine for 2D CAD or viewing 3D and email but you will most probably need a desktop workstation to do serious work.
The next category is the 15-inch laptop. These are still portable, but of course have a bigger screen and can have the fastest CPUs and upper mid range GPUs. This is the most hotly contested area for manufacturers so there is quite a range to choose from. The external power supply units (PSUs) are similar in size to the 14-inch models and fairly portable.
The top size is a 17-inch laptop and they are classed as desktop replacements. They have top CPU options, mid- to high-end GPUs and very large form factors. They also have very big external PSUs to match. As they have lots of options they use lots of juice, which limits the battery life. Six cell batteries are typical in most laptops but the 17-inch models normally come with nine cells. All of this makes 17-inch laptops very heavy so they are targeted at replacing desktops - desktops that can be moved. One of the big problems with 17-inch laptops is using them on the move. Battery life is one thing but the physical size makes them hard to use on the bus or plane. One time I got out a 17-inch mobile workstation on a flight and half of my laptop was on my neighbouring passenger’s tray. If ever there was a case for me to fly business.
So what is my ideal? I’d go for a well specified 15-inch laptop with a fast CPU and high resolution screen, good graphics and a large drive. In my view this is as powerful as the 17-inch but is portable enough to be carried and used on plane etc. The screen is good enough to be used at my desk and unlike the larger format laptops it doesn’t give me a sore shoulder.
On the options list you can add a small PSU. Lenovo offers these as well as non OEM companies like Targus . Options such as extra connectors to charge blackberries and phones, can be added. Plane/car 12-volt adapters are included. My current Laptop is a Lenovo ThinkPad W500 and a PSU that is around 14mm thick, which is not only lighter, but a lot easier to carry. As a fairly new laptop it has a Display Port adapter which you can get convertors to HDMi (TV) and DVI. I store all my passwords in the fingerprint software, which saves logging on.
Rob Jamieson often gets a little over excited about Lenovo’s slimline power supply pack
Like most corporations my company does not want you installing unnecessary applications on the encrypted drive. Having two hard disks is one option but since the USB memory sticks have become so large and cheap (16GB are now £20) I use this as a second boot device with another OS on. I installed a TV tuner for local watching and use a Slingbox from home. The Slingbox sends my TV over the internet so I can watch satellite while I’m away on business in my hotel room. The only downside is that my wife sometimes changes channel half way through, typically when I’m on the Discovery channel.
Laptops are a growing market and increasing in volume compared to desktop workstations so its market value and share is key for large manufacturers. All offer slightly different features but in the end a laptop has to be robust and reliable because it soon becomes the hub of everything in our mobile world. Never scrimp, buy quality and buy what fits your needs.
Rob Jamieson is marketing manager for workstation graphics at AMD. He realises there’s a thin line between enlightening readers and getting busted by his IT department for installing unauthorised software. The opinions expressed in this article are not those of AMD.
.(JavaScript must be enabled to view this email address)
Comments on this article:
Material selection process can be a minefield
13 March 2009
Process types: Design and Manufacture
Form and function are the core tenets of product design, but by not fully understanding the effects that manufacturing can have on materials, John Tumelty believes we’re working one sandwich short of a full picnic
Material science is a topic of immense depth and complexity and it doesn’t diminish that much even when you filter out a significant number metals, ceramics or plastics. With plastics alone, for instance, there are many thousands to choose from with a vast range of different chemical, mechanical, physical, aesthetic and processing properties.
Whilst I never set out to be a talented bore, my wife continuously reminds me of this innate ability, so I plan to keep this to a relatively simple overview of some key considerations that the designer should be aware of during a plastic product development phase.
figure 1
Form & function
The shift of focus within the design and engineering field in recent years has led to a disproportionate concentration on aesthetics when really what is required is a rounded approach, one which balances aesthetics with function, with performance, with manufacturing. The number of wild and eclectic chair designs that grace the covers of magazines and are never seen again is one of my pet hates – why bother?
Taking key decisions early
At a very early stage in the design process a lot of key decisions are often made in regards to the selection of materials. This is a result of many environmental or physical performance attributes being defined early on. e.g. must be able to operate in 120°C ambient temperature environment; must be able to support 200kg; mustn’t shatter when dropped etc. What is critical to remember is that this material selection process needs to be carried forward into the detail design phase.
wild and eclectic chair designs that grace the covers of magazines and are never seen again is one of my pet hates – why bother?
Different plastics behave differently within the same geometry. Certain materials are more prone to sinking or creating voids, melt flow indices give a hint as to how well very thin sections will be filled (or not filled); fibre filled materials are prone to warping according to the alignment of fibre structure and the influence this has on shrink. If we then add in the variables associated with the design of a mould tool, then the list just grows and grows.
Variance is an issue
If we now look at various ways of achieving the same geometry in an ABS thermoplastic via different manufacturing methods, even more differences and difficulties become apparent.
Figure 1 shows standard tests carried out on a standard shaped part made via different processes (FDM, CNC machining and injection moulding) with three different gate locations for the moulded part to highlight the impact of gate positioning. What you’ll see is that even with the same material and same geometry processed in a variety of different ways, the manufacturing and processing methods can have a huge influence on the mechanical performance of the part.
In conclusion
To conclude, without waffling (too much), the material selection process is a minefield for the designer and a critical component of this selection process must be the final intended manufacturing process combined with how the geometry will influence this. The design process should not be segmented into isolated stages but should flow, continuously re-evaluating earlier conclusions before achieving completion.
John Tumelty is Managing Director of Proto Labs. His headquarters in Telford has more CNC technology than really should be allowed in one place.
.(JavaScript must be enabled to view this email address)
Comments on this article:
Design tools on the Apple Mac
12 March 2009
Process types: Design and Hardware
It’s been brewing for years, but finally a platform long dormant in the 3D design world is making a break for the mainstream. Al Dean comes clean and professes his love for all
things Apple flavoured
It started in San Francisco, just a little over a year ago. A weekend was spent enjoying the sights, sounds and if I’m totally honest, the bars, of one of the world’s great cities. Just as we were in the final stages of launching DEVELOP3D, I knew I had to buy a new laptop. The usual workflow for this sort of thing means load up Dell.com, whip out the credit card and order - it arrives in plain paper packaging a week later and away you go. Job done.
But on visiting the Apple Store in San Francisco, with a little post-hangover/lunchtime beer buzz, I had, what alcoholics refer to as a moment of clarity. After decades of ridiculing them, of teasing the many graphics designers I’ve worked with about their one-button mouse, I did it. I took a turn to the dark side and did something I swore I’d never do. I bought a Mac.
I panicked for the first few months, immediately installing Bootcamp and XP so I could run all those ‘business critical’ applications - Word, Excel, PowerPoint and Outlook, as well as all those lovely 3D CAD apps. I needed to have my Windows comfort blanket. After all, that is how things are done in the design world if you’re serious about developing products using 3D tools. Right? However, more and more I found myself opting for the Mac OS X boot option. And slowly and surely I found myself being seduced by the shiny interface, the more elegant way of working and the delight that is multi-touch.
The majority of the CAD-using world will continue to rely on Windows hardware to make things happen, but for those that adopt the Apple platform it soon becomes much more than just another computer
Looking back, it seems odd that I, the Windows zealot, would get hooked by such things. After all, 3D technology is my driving passion and the reason I decided to get involved in DEVELOP3D. So adopting a platform that seemingly has very little in the way tools for that process, would seem counter intuitive.
But the interesting fact is, it does. And the trend of the last six months has made that move somewhat prescient. While I’ll never claim to have any form of crystal ball gazing abilities, it seems that I, along with a growing number in the industry, could see that the Apple platform offers some real potential.
Let’s put things right. There have been many 3D design tools available for the Mac for quite some time. Ashlar Vellum has been developing its range of hybrid modelling tools for the platform for decades. SolidThinking, enjoying somewhat of a renaissance following the acquisition by Altair Engineering, has been on the Mac for a similar length of time.
But things have also stepped up of late. Both old names and new are jumping on the bandwagon. Perhaps the biggest irony is that Siemens released NX on the Mac platform a couple of years ago. It’s ironic because Apple uses NX to develop the majority of its products, but historically had to use Windows due to a lack of Apple support. Many see this as the sole reason that NX is now appearing on OS X, but having just installed and worked with NX 6, and the inherent Synchronous Technology it contains, it works nicely on the platform.
Furthermore, this month alone, we’ve seen that Autodesk, after a 15 year hiatus, is jumping back on the bandwagon in a big way. Maya and SketchBook Pro have been on OS X for some time, but the company is expanding things not only with its CGI-focussed tools like Mudbox, but with a suite of industrial design tools including AliasStudio, which is to debut on the platform later this year.
Then of course, there’s Rhino. That ever present, ever popular, but strangely quiet, surface modelling system which has gained a massive following in many industry sectors. McNeel announced ‘iRhino’ over two years ago and it’s been in a long and extensive beta program ever since. However, the system is looking very stable, very slick and can’t be long off from its official launch.
The Mac Book Air: Not just any old laptop
Alongside the core 3D design tools, there are many other related applications around. Bunkspeed shipped HyperShot for OS X some time ago and it stacks up very nicely in comparison, particularly with the high end Mac Pro, eight core workstations. There’s there’s Luxology and its Modo system, which has been OS X-native for a long time.
So why now? Why is this trend growing? The answer, I suspect, is that Apple has achieved something that very few companies, let alone computing hardware manufacturers, ever have. It has got the ‘cool’ factor. The fan base is huge - the volumes that the company is shifting year on year, is growing and growing massively.
The majority of the CAD-using world will continue to rely on generic Windows hardware to make things happen - that’s economics - but for the majority of those that adopt Apple and the Mac platform, it soon becomes much more than just another computer.
If there’s one thing that irks me about the whole thing, it’s that Apple, in all its polo necked, black sweater, white out retail space, coolness, is ignoring a market that offers huge potential. Design is cool but, on the whole, engineering is not so it becomes a much harder sell. However, despite Apple’s apparent reluctance to promote itself in this sector, there seems to be a groundswell of interest anyway.
I’ll be reporting on my Mac + 3D CAD adventures regularly on the develop3d.com blog, so please drop by and let us know your thoughts. Because I can guarantee, brothers and sisters, that once you buy Mac, you never go back.
Al Dean is Editor of DEVELOP3D. With three ipods, a Mac Book Air and an obsession with iPhones, he is currently undergoing specialist counselling to help deal with his Apple addiction.
.(JavaScript must be enabled to view this email address)
Comments on this article:
Who cares about Autocad?
02 March 2009
Process type: Design
Autodesk is about to release a new version of AutoCAD. In these days of dedicated design and manufacturing systems, why should we give a damn about the perennial 2D CAD tool?
asks Martyn Day
When discussing the editorial topics for this issue I outlined my column idea - the new AutoCAD. I didn’t expect the response of my colleagues, which could be summed up in one word. Why? I found myself having to justify why AutoCAD still has relevance and has an important place in serious product design across the globe. It was strange, it was as if I had suggested I write about quill pens or a slide-rule. The fact that you are reading this is either a testament to my persuasive powers or sheer bloody mindedness, or perhaps a bit of both.
I can easily see where the AutoCAD-bigotry stance comes from. It’s the ‘dumb’ 2D software we dropped to develop our products in dedicated parametric 3D systems, featuring analysis, digital fabrication and virtual prototyping technology. 2D is just plain dumb, it’s history, it’s the inefficient way of working we left once we embraced 3D. Once the enlightened have crossed the rubicon and embraced 3D, it’s easy to have an ‘us and them’ attitude which invokes all sorts of emotions, some bordering on the religious.
There is no denying the success of AutoCAD and its cut down LT variant. Tens of millions of copies have been sold and pirated. As a 2D tool, it’s the world’s illegal and legal defacto standard and 2D still has a place in design and manufacture. Even with exciting desktop solid modelling products like Inventor and SolidWorks having double digit growth for years on end, 2D seat sales of AutoCAD have continued to accelerate. All the CAD vendors, including Autodesk have spent considerable time and money in trying to get these AutoCAD customers to migrate to the new 3D systems but all have met limited success. It seems even Autodesk has trouble cross-grading customers as AutoCAD DWG and 2D are very ‘sticky’.
Autodesk is the only company that can touch this many 2D users and provide insights into other ways of working
AutoCAD’s 3D capability has been certainly on a shallow trajectory. The first instance of solid modelling being built inside the application was the disastrous Release 13 (the company licensed the ACIS engine). At the time, Autodesk was trying to compete with SolidWorks and Pro/Engineer and quickly realised that they couldn’t compete by building Mechanical Desktop on top of a flakey AutoCAD, thus a new standalone system, called Inventor, was developed in parallel.
Then having very clearly demarcated fresh 3D offerings for its vertical markets, the limitation of AutoCAD’s 3D development was sealed to languish for a number of years. Indeed, AutoCAD had to wait until its 2007 release to get a decent 3D overhaul for both capabilities and user interface but this was still very pull and push grip-based editing with no parametrics. AutoCAD 2008 and 2009 revamped the lighting, rendering and user interface, with major enhancements to come in 2010, which is imminent.
Autodesk’s motivation has clearly been to get its AutoCAD customer base to its 3D verticals, especially Inventor, before its competitors do. I understand when it came to AutoCAD 2010 there was a hot internal debate as to what level of high-end functionality the base AutoCAD system should have. Too much and you potentially remove the incentive to move to Inventor. It would appear the platform group (AutoCAD) won the day and has developed 2010 to feature comprehensive 2D parametrics, conceptual free-form modelling and direct 3D printing. Nay-sayers will sarcastically claim AutoCAD has finally joined the 20th Century but by adding these tools to such a commonly-used application can only accelerate the use of paramatric and model-based design.
Autodesk is the only company that can touch this many 2D users and provide insights into other ways of working. 2D parametric design is at the very least a good education. The 3D capability of AutoCAD now looks much more useful to users both new and old, providing innovative surface and solids tools.
Technology adoption is like using stepping stones, the further they are apart, or the greater the difference in height, the less likely you are to cross them. While everyone has had trouble getting AutoCAD users to step ‘across the stones’ and migrate to 3D, AutoCAD 2010 drastically reduces the stepping distance and levels the heights of the stones. This is a powerful release that will, given time and an open mind, change the industry perception that AutoCAD is just about dumb 2D.
Martyn Day is Consulting Editor of DEVELOP3D. He has given up watching recession news for Lent. In addition he’s wearing rose-tinted glasses and is bidding on a rabbit’s foot on ebay. .(JavaScript must be enabled to view this email address)
Comments on this article:
Careful with those fillets
02 March 2009
Process type: Design
There is one, often over used, 3D design feature that can have a huge knock on effect on Design for Manufacture. John Tumelty asks that we think before we go crazy with the fillet command
The fillet radius is an engineering and design essential, enabling areas of stress concentration to be relieved while also aesthetically improving a component design. When cavities are CNC milled into a mould tool, filleted external part corners are also a necessity for this manufacturing process (you can’t cut a square hole with a round cutter).
With modern 3D CAD packages the process of adding filleted corners is blissfully simple; select faces or edges, click the button and hey presto the whole part can be filleted with ease. This is in contrast to the bad old days of pencil and paper, where adding fillet radii was a laborious process. However, it also resulted in more time and thought being given to what can be a tricky problem.
Figure 1
The simplicity with which corners can be filleted in modern 3D CAD packages has led to some new problems regularly occurring with component design. While the process of applying the desired fillet radii is simple, the thought process and design of said fillets should follow the same care and consideration as the rest of the design process. Below are a couple of examples of problems caused solely by fillets that I see every day.
The orphan fillet
Fillets, in certain geometries, can be a problem. Take, for example, the part shown in Figure 1. The green faces of this part (including the left and far sides that don’t show in the diagram) are created by the one half of a mould and are drafted appropriately. The surfaces shown in blue are created by the other half of the mould and are drafted in the opposite direction, which means they taper towards the top of the diagram. The problem is with the fillet highlighted in red.
Figure 2
The CAD software, recognising that the fillet connects faces drafted in opposite directions, has tried to resolve the conflict between two opposite draft directions, but has done so unsuccessfully. The reason for this is that, in reality, this fillet cannot be part of either mould half. In other words, this feature is an orphan. To clarify, Figure 2 shows what the mould cavity would look like if this fillet were created by the fixed half of the mould (the green half). There would be a problem in the area that appears in Figure 1 as a small red triangle on the base of the part. The part of the cavity that created the fillet (highlighted in red) would trap the moulded part under it at that triangle when the mould opens.
Figure 3
There are three possible solutions to the problem of the orphan fillet. The part could be redesigned so that everything was drafted toward the top. In that case, the entire part, fillets included, could be moulded in one half of the mould. You could avoid vertical fillets that connect faces drafted in opposite directions. This would prevent the problem in the first place. Or part could be manufactured as designed with the addition of a moving core, but this would significantly increase the tooling cost.
Parting line radII
Another area of moulded part design that can suffer as a result of reckless filleting is along the split line of the mould. If you look at Figure 3, you see the parting line where the two mould halves meet to form the sharp edge of the part. In machining and running moulds, there is always some tolerance, but slight movement of the parting line to the left or right will not change the geometry of the edge. In Figure 4, on the other hand, both the mould halves form the parting line edge, so any mismatch in the mould will leave a ledge, changing the shape of the part at the parting line.
Figure 4
Final thoughts
Like every stage of the design process, the finishing touches such as fillets can have a remarkable influence on the manufacturability, performance and aesthetics of the final product. However, just because the software has made the process blissfully simple, the designer doesn’t need to act in the same way.
John Tumelty is managing director of Proto Labs. He is mildly amused whenever someone pronounces fillet as you do when ordering a Fillet O’Fish.
