What do you find is the differences in designing a prototype compared to actually manufacturing the item?
We know that some prototype systems can only do certain tolerances and hold certain thicknesses, how do YOU incorporate all that from designing a prototype then going into manufacturing because of those differences? Do you just design it for manufacturing and worry about small issues when you do the prototype itself?
I can only tell you what I do, and I think a lot of what you are asking is application specific.
If what I am designing is going to be built on a somewhat large scale, i.e. batch sizes of 100+ units, Then I always design it for assembly and manufacturing and use the tolerances and fits that I've found work best for who ever is doing the assembly.
I've always considered a prototype to be a functional version of what I want to manufacture. So if it's really hard to build the prototype, then I know it's going to be a least two fold more difficult when I want it manufactured, and cost about that much more.
Now if I know I'm building or designing a one of a kind, one time item, then I know I have a little more leeway in keeping my tolerances tighter, and perhaps getting a little more fancy with the design, because I know it will only have to be done once, and then the pain is over.
So I think it's not so much a question of designing for a prototype versus manufacturing, as it is designing for single part - small quantities versus large quantities. It really changes the way you think about a lot of things.
We make a lot of prototype parts and some short run production, i.e. <100. Our machines are purchased with that business model in mind, our shop is set up for that mentality, etc.
Tolerances are really non-existent in prototypes if you go to a good prototype shop, they should be bale to be very close to nominal and it not be an issue. (I know some prototype shops are blacksmiths where you have to specify everything and they have to inspect everything but why use organizations that do not understand their processes well enough to make things work.) it is when you start getting multiples and need interchangeability that you have to worry about the tolerances.
We use many four and five axis processes as part of our normal processes so the costs of making whatever features you need in a small volume are again somewhat irrelevant whereas those features may have a large effect on larger volumes.
I agree with Josh - it is really are you designing a one out, a small batch, or large volume. If you need it on the large volume it is rare that you should have a different design for protos because you would probably spend more time designing the proto than it would cost to just make it the same.
I understand where you are both coming from. This is what I am running into. We have an EOS M270 and it seems like we are always doing some super crazy geometric dependent parts. With that in mind, I always wonder how the hell are they going to manufacture this lol. Don't get me wrong the parts are amazing and look great when it is done, I just don't understand how they will take it to the next step.
When you design a part, do you know the manufacturing process you will utilize before prototyping? Or do you rely on prototyping the part then worry about manufacturing later?
It's funny you say tolerances are non-existent when doing prototypes, it seems like some engineers just don't understand that, they have this notion that when prototyping a part it must be meeting to stringent specs ie. Surface Finish, Tolerances, Heat Transfer or more.
I am ok with that, because I know that sometimes they need to test certain aspects of the prototype, thats ultimately what a prototype is for, but will you achieve those same properties when bringing to market?
These are really good questions, and some very good things to be considering, I believe, when designing any part.
I always take into consideration how a part or product is going to be made when I am working on a design. In my mind, it is a crucial part of the design process and has to be taken into account.
I have a pretty good idea of the type of machines that will be running the parts I design, and many of them I know how to run and program myself, having spent time working on the shop floor.
My time on the shop floor have been an invaluable asset to me as I work on a design, because I know what the limitations of the equipment are, and I also have a pretty good idea of what strengths and capabilities they have. I know what size of material they can handle, how much material needs to be allocated for workholding, and how to avoid designing parts that produce an excessive amount of scrap, and won't be easily nested together.
Every design is going to require a different manufacturing process and a different part flow. The trick though is to keep things as simple as possible and still achieve the desired ends results. Better machines will generally make better parts, but if I design a part that requires 5 axis machining or heavy stamping, when the same design goals could be achieved with a part bent on a hand brake, then I feel that I am only wasting precious time and money. Which, is something I never want to do. ;)
Thank you for enlightening me on that! I am glad you actually spend time in the job shop!
Let's say you have a part that you are doing on DMLS machine...it's highly geometric and really that's why they are doing it on the DMLS machine to try it out. How can you really take that part that really can only be made doing DMLS and manufacture it? Don't get me wrong I love DMLS parts and that it really can be used as "Additive Manufacturing" or "Digital Manufacturing" but how can you justify only doing it the DMLS way?
What this really comes down to is...I know the technology we use, however I am very unfamiliar with the design process and how you decipher certain changes when knowing the part will be manufactured. Thank you very much for educating me in this realm.
Rapid Prototyping is an interesting industry, and sadly, one that I personally haven't had a lot experience with, other than what we build on the shop floor.
However, I am familiar enough with it to at least give you my opinion and take on where the industries current place in the manufacturing process is.
The whole propose of manufacturing is produce parts. Nobody really cares how it's done as long as they have the parts they want, when they want, at a price they can afford. And ultimately, those three criterion determine what the manufacturing process put in use will be.
I would not be surprised if the future of part manufacturing lies in machines like the DLMS machine that your company owns. In order for that to happen though, a few basic things need to happen.
-- The machines need to get faster than other existing manufacturing processes.
-- They have to cost less to own, operate, and produce parts.
-- And they need to put out a part quality (strength, surface finishes, etc..) to meet or exceed what is already being done.
Like I said before, each part or design is going to require a different manufacturing process or technique, and some designs (like perhaps the kind you see on a regular basis) may require a completely new manufacturing process to be created. It all depends upon the creativity of the Engineers, and how badly someone wants their parts, that will determine whether the design ever hits the market, or not.
"With that in mind, I always wonder how the hell are they going to manufacture this lol."
I have been training engineers and designers for years, and one of the aspects of that training is to help them understand the cost of making their design. Cost of manufacture has a great deal to do with making a profit today. I will agree that there are some parts that must be aesthetically pleasing, while quite often in machine design, it only has to meet functional and strength standards. Creating an overly complex part in a 3-D modeler without regard to how it is manufactured is a recipe for cost overruns.
Tolerancing of a part is not often understood by engineers or designers that have no shop or assembly experience.A few days in the shop for every engineer or designer would be a great cost benefit to the company, in that they could more easily understand the problems that they cause within their designs. When the design I did a few years ago, where I redesigned a machine baseplate that only exposed three of the six sides to machining, saved over $800 on a single baseplate that was originally machined on all six sides. Sometimes it's the simple things that cost the most.
One should always design for manufacturing. None of the prototyping machines hold extremely tight tolerances, and are merely used for shape and fit, not functionality. If you're talking about creating prototypes on a standard machining center, mill or lathe, then the design should follow nominal (exact) dimensions. That model can be easily toleranced later when turned over to manufacturing.
Thank you for your input on this. I know where you are coming from, however there are certain prototype methods that do not meet exact tolerances as well as can do certain thicknesses ie; wall thickness, hole thickness and geometries. Let's take for example SLA...We all know SLA is used quite a bit and we know there are limitations...We also know that most SLA parts are to see the looks of it etc but if you can not see the whole part how it will be manufactured then whats the point? What I am getting at is this...
There are certain limitations when it comes to most prototype methods, do you re-design the drawing so those limits arent reached and then design again when going into manufacturing?
I had a customer recently request a smaller part out of PEEK material meeting specific requirements, I was able to sinter the parts but having to make some small changes so the material wouldn't melt in small areas...now I don't know his exact manufacturing that will take place with the part but I often wonder how it can be done.
I'll be honest, I have CAD background and experience in Prototypes but none in manufacturing. I think i need to spend some time with a partner and see how some of these prototypes are actually manufactured ...but that probably means a trip to asia lol.
Once again Dennis thanks for your input, I really am just trying to see how design relates to prototype and then to manufacturing and if you have had a prototype made and expect the end part to be exact to it but when it comes to manufacturing it ends up being quite different then the prototype...
I believe Conceptual Prototyping is the key in untried technologies. In this approach, there are a minimum of 2 prototypes built, (1) the proof of concept model and test equipment (2) Advanced modeling design for pre-release testing.
Generally there is a third, more refined prototype/pilot equipment built in multi-unit production in which all tolerancing for manufacture issues are addressed. For single unit production, very often machinery is designed modular to allow replacement of poorly incorporated components. This is not my choice of methodology.
I recently performed work for a company that simply refused to prototype and test their new equipment lines. Designed on 3D Cad, computer analysis, Build and ship. They are no longer in business. The computer is no replacement for brute force tried and true engineering development and testing by experienced machine builders.
Take heed and lead !!
Bill Redd - 30 years experience