Originally Posted by
peteeng
Hi Dod - Personally I think it's a mistake to buy two bits of sophisticated machinery and think you can integrate them and get a good solution with no knowledge base or experience and get a commercial outcome. It is possible but expect to take years to get to a solution. I used to program robots many years ago and fibre lasers have come along way. It all sounds doable and it is but you can spend months sorting out one little glitch that comes down to proprietary info that the laser or robot supplier does not want to tell you about.
I recommend finding a commercial solution, there's lots out there. If you can't justify the commercial solution then you certainly can't justify the R&D required to develop a solution. I design bespoke machinery and systems for a living and at the front end of a project costs are estimated and we regularly multiple these by 3x to estimate budget costs. Even if you don't go the commercial way it establishes some $$$ value for the project. There are two usual reasons to develop your own machine 1) it does not exist commercially and 2) It appears you can't afford the commercial solution. If you go with 2) it is often a very poor choice. They don't cost that amount because they are all driving gold plated BMW's they have to get their costs back and so will your company.
Having said all of that there are robot control programs around that are free to low cost and doing what your saying/intending can be done. Good Luck finding out what you don't know you don't know and more stuff you don't know. I've been at it some 40 years and still come across more stuff I don't know. Keep us in the loop always interested in robot projects...
Oh some advice - start as small scale as you can. Even if this is desktop type stuff. The learning is the same whether its desktop or massive and the software solutions and integration will be the same but the asset costs will be much less. The first decisions you make are more likely to be wrong than close. You actually should not buy anything until you think the solution(s) is fully costed and resolved on paper.
Projects where you buy something in the hope it will work and buy something else then try to integrate them, then change it, then repair it etc. There are heaps of router threads for instance in this forum built that way and they take years to sort (and that's all been done before and well documented). Much better to keep it on paper for as long as possible. I think it was Einstein who said take 90% of the time to think about the solution and 10% to solve it...
Scaling up is easy once you know the design space, the issues and solutions. The first specialised factory I set up took 3 years and $500kAUD. Then someone asked me to do it again and I did it better in 6 months and $200k and we made money as soon as the electricity was connected... I expect you'll do something similar...Peter