Well first I should say I have had well over a dozen of prototypes designed by BatchPCB and they’ve always worked… Until the last one I guess.
I noticed the 3.3V plane was somehow not continuous. After investigation it turns out a via is not making the proper connection from one layer to the other one (this is a 2 layer design).
Here’s a picture of the incriminated via (via 1 in the attached file)
While light seems to go through most of the other vias, this one looks more or less blocked.
Again I have nothing particularly bad to say about BatchPCB, I am pleased with their service (if the manufacturing could be quicker that would be awesome though).
I just want to be sure I understand what happened here, because ordering new PCBs always puts me behind schedule and I can’t afford this anymore.
What do you think happened here? Obviously this board passed the BatchPCB’s DRC bot.
Is this a design issue, or a manufacturing issue?
Any particular solution to this issue you would recommend, so I can still use this prototype board?
leon_heller:
It looks like a manufacturing fault. It can be worth paying the extra and using a supplier that tests their boards.
I’d put a drill through it, put a piece of wire through it and solder it at both ends. Use wire-wrap wire if it needs to be insulated.
Is there such an option with batchPCB to get your PCBs checked?
As for drilling, it looks like that would be the cleanest solution, though I’m not even sure I can find a tiny drill bit like this one. There’s always the option to run a wire from a 3.3V pin that’s powered I guess.
I’ve got drills down to 0.6 mm that I use on my home-made PCBs, smaller ones are available. I don’t think that Batch PCB offers board testing. Gold Phoenix probably does, like most suppliers.
leon_heller:
I’ve got drills down to 0.6 mm that I use on my home-made PCBs, smaller ones are available. I don’t think that Batch PCB offers board testing. Gold Phoenix probably does, like most suppliers.
Here’s the yellow wire modification that made the PCB usable.