Shapeways print length ultimatum (yeah you told me so Bryan :-))

[Lemosteam]

Well, it's not as new as you might think and everything they are doing now was proven then.  just because it is making things possible for us now does not mean that it is changing rapidly, in fact the only thing that has really advancesd are the materials used. 

I was trained on SLA machines in California at 3D systems over 30 years ago with similar resolution then using lasers with mirrors and hazardous fluid in a 500mm x 500mm x 500mm vat. A $1m machine equal to Formalabs Form 1 now, lol.

Process control has been perfected in the same time frame even though the manufacturing processes are quite old. I learned about SPC and tolerance management also at the beginning of my career.

One thing that has not been considered, is the fact that they are making profit from my intellectual property, so there really is a form of implied contract, and yes I feel they are not meeting it.

@peteski, no. Lengthening the model is no more hit and miss than their lack of repeatability.  How many trials would I have to try, only for them to recalibrate without my knowledge, so the next print is off a different amount.

In the real world, folks design things to fit other things within a range. this is referred to as Design in Context, to create the promise that although the digital world is nominal, the amount of process variation is always accounted for, say, to avoid squeak and rattle issues in your car, or a piston having enough clearance in the cylinder block to perform correctly or the gear tolerance for near perfect mesh to avoid gear whine.

Parts can never repeatably be manufactured to nominal, variation ALWAYS occurs.

@lashedup , I was not upset with your response at all, I hope I did not come off to harshly, but it is still a very sore point with me.

And yes, @C855B , I could use a pint.

[peteski]

@peteski, no. Lengthening the model is no more hit and miss than their lack of repeatability.  How many trials would I have to try, only for them to recalibrate without my knowledge, so the next print is off a different amount.

That's what I meant. Simply lengthening your design by x-percent will not consistently guarantee that you'll get a printout which will have correct length, ever time.  Basically, that is not a valid workaround. 

[lashedup]

Well, it's not as new as you might think and everything they are doing now was proven then.  just because it is making things possible for us now does not mean that it is changing rapidly, in fact the only thing that has really advancesd are the materials used. 

I was trained on SLA machines in California at 3D systems over 30 years ago with similar resolution then using lasers with mirrors and hazardous fluid in a 500mm x 500mm x 500mm vat. A $1m machine equal to Formalabs Form 1 now, lol.

Process control has been perfected in the same time frame even though the manufacturing processes are quite old. I learned about SPC and tolerance management also at the beginning of my career.

One thing that has not been considered, is the fact that they are making profit from my intellectual property, so there really is a form of implied contract, and yes I feel they are not meeting it.

@peteski, no. Lengthening the model is no more hit and miss than their lack of repeatability.  How many trials would I have to try, only for them to recalibrate without my knowledge, so the next print is off a different amount.

In the real world, folks design things to fit other things within a range. this is referred to as Design in Context, to create the promise that although the digital world is nominal, the amount of process variation is always accounted for, say, to avoid squeak and rattle issues in your car, or a piston having enough clearance in the cylinder block to perform correctly or the gear tolerance for near perfect mesh to avoid gear whine.

Parts can never repeatably be manufactured to nominal, variation ALWAYS occurs.

@lashedup , I was not upset with your response at all, I hope I did not come off to harshly, but it is still a very sore point with me.

I wasn't sure and I understand your being frustrated. I do realize the technology has been around for 30 years plus, but a lot was tied up in patents for a number of years which usually results in some lack of innovation depending on how restrictive those patents are. I do think things have been improving and evolving over the last five years quite a bit. Pricing is dropping, resins are getting better, screen resolutions and projection methods are getting more consistent. I have no doubt that I could get a print out of my printer that will fit within your tolerance, but it may take a print or two to get there. Resin prices are falling, but still not cheap and the length of time required to wait and see what the final outcome of the print is going to be makes testing arduous. If I can get repeatability out of it reliably then I'll consider printing directly from it. Otherwise I may look at producing a master and then casting from there. I have two passenger car test projects I'm working on - one has a fixed roof as a complete shell, the other is designed to use the Kato dome roof, so fitment will be key. Soon as I figure everything out and complete some testing I'll post up results.