Skip to main content
Topic: What are the requirements of a prototype P&P machine anyway? (Read 1801 times) previous topic - next topic

What are the requirements of a prototype P&P machine anyway?

This site is about prototypes, after all...

I have been following Ian's P&P project with great interest. It will be extremely interesting to hear what Ian likes about his machine and where the shortcomings (if any) are, as I am building a machine for myself. I want to build prototypes in-house. The machine Ian got is not suitable for me, because of the limited number of feeders. I do analog stuff, which means many different value components on a typical design my last board had 82 different components. I know what I want for my machine - or do I? It would be stupid to build in a restriction that could have been easily avoided. So, my view for a hobbyist or a prototype lab p&p machine requirements:

- Low cost: The machine has little purpose unless it pays itself in a year or two, compared using outside services. I'm targeting less than 1000€.

- Accuracy: Must be better than hand placement. This is not a hard requirement, better than 0.1mm is relatively easy. If it can put down 0603's it is good enough, 0.5mm chips or BGA's would be a great additional bonus. Calibration of the machine needs to be well thought trough to achieve this.

- Speed: Must be faster than hand placement.  By definition, a prototype machine does not do production so it doesn't have to be very fast, but it has to save time.

- Feeders are not needed: By definition, a prototype machine does not do production, but one or two boards at a time. It is fine to use stationary slots and peel off the cover tape for required amount of components for this job. Something like the base plate of this (http:// is good enough. With vision, even double side tape on the work table could work.

- Vision (up- and downlooking) is needed: A DIY machine will not be very accurately built. It won't be precisely square, the component positions are only approximately known, the PCB position or angle on the table is not accurate and so on. I don't think I could get the machine to work good enough without help from cameras. Besides, having vision allows placing loose components: Put a component on a pick-up area, approximately 0 degrees position., the machine takes a look and finds out the precise position and alignment.

- Workflow needs to be well thought trough. If setup takes more time than manual placement, the machine is only cool, but not very useful.

What did I miss?


Re: What are the requirements of a prototype P&P machine any

Reply #1
There seems to be a lot of P&P projects permanently in the planning phase. Since it is a huge engineering challenge, I think the best approach would be for a group to hack or back-engineer an existing one. I've been lurking the taobao thread, but I'm still not convinced it's a good platform to start from (or that it even works out of the box). I'm anxious see how it plays out.

I think the Madell DP2006-2 would be a good platform to start from. Structurally, it appears to use some off-the-shelf parts. Yes, it is over twice as much, but I think in this case it is worth it.