--background, skip if you like --
I've got a late model Suzuki pick and place machine, the SMT-2300C. Its new to me / I acquired this machine about 2 months ago. I'll be using it to stuff boards for a project of mine. The machine was working great prior to me purchasing it, and it was working when it arrived at my place. Then, I got a serial comms error.
There is a lot of information about the repair steps that have been carried out so far, I'll skip the details for now. The general info is that there is a main cpu board, and a vision system board - the vision system is made by cognex / 3rd party to Suzuki and is integrated in the system via power supply and serial comms. I did a lot of work 'troubleshooting' the serial comms stuff on the main board side / replacing components, buying IC's from all over the world, learning about some of the old school serial comms, processing, and eprom programming stuff. This board rework was quite perilous and in the end, after all that work the serial comms system is working and was always working just fine. geeze.
-----
Where the problem is getting interesting now, is that during the repair work, having replaced all the components that I did and everything to no avail, at my wits end I blew the dust off the board on the vision system and what do you know, the damn thing came back to life. just a bit of dust? that seems strange to me. Alright, so the board is booting up and making comms and working and all that after blowing it off with some canned air. after about an hour of being powered down, the machine no longer works on power up, and gives the serial comms error again.
so I figured, alright, maybe I just didn't get all the crud out of the board, and because I blew air all over the board I have no idea where the problem was really. I blew the board off again, and the system came right back to life. naturally I decided to take the boards apart from the chassis and clean everything thoroughly. The vision system has a main board, and also a Suzuki made interface board which just has traces and fans out the I/O, no active components on that interface board. There are large 3x32 pin DIN connectors between those boards. I put it all back together and booted the machine up after this cleaning, and it works. however, again after some time, trying to reboot the error comes back. So I went about blowing off local sections of the board. several locations on the board showed no result from blowing off / the system didn't come back, then it turns out that blowing on one of those DIN connections with air does bring the machine back to life. on that connector there is only power supply rails 5v, +12, -12, gnd but they are fanned out to many locations via the DIN connector. So I tore it all down again, cleaned that connector very well, and put it back together, it works. then after some time, reboot - doesn't work.
wow. so it seems blowing compressed air on this connector (from a can, so not high pressure, and not a lot of air) does actually do something. It seems to me it is not related to particulates, and not related to contact cleanliness. its like the air dissipates charge somehow? it is a total mystery and speculation is just killing me on this one. you could imagine just about any reason for this air / charge moving / connector nonsense. Its a bloody mystery and I have trouble believing that the air has a charge removal effect even.
the board is 1990's early vintage. there are all dip components, each with its own decoupling cap, and there are very few electrolytics on the board.
any thoughts? anyone ever have compressed air bring a very complex board back to life intermittently? wtf!
Home
Help
Search
Login
Register
