Hello Harry, please make a video about the change of nozzles. Interested in the process of their change

I just placed over 50 tsop16-ep with component threshold to 85 and exposure 9. The outside case edge is reliably detected every time (LT3081). No black screen. I also placed in same run many 0.5mm pitch qfp64-ep and even some wson8-ep 4x4mm tiny dual reg parts with two exposed pads and weird exposed pad shapes. none needed tweaking, even though I did notice a consistent small camera offset error to the right and a few degrees rotation. with that said I have not tried to align the placement offsets in any way, this is straight out of the box. I understand the argument for aligning to the exposed pad.... but perhaps you shouldvsey it up to align to body instead? Seems to do the job.

@aikendrum1518

7 жыл бұрын

note also that there seemed to be a thresholding issue with some light coloured caps on ""quick"" that required I set the threshold back to 80. maybe a shroud would fix this. We placed over 6000 components on 4 large boards (they were near the maximum the machine supports) with it with a variation of 158 component types (nightmare) that were double sided reflow. It was a trial by fire, there are some subtle issues with prick alignment, and reel feeding that cause issues, there are some thermal variations that effect prick accuracy and I plan to buy a dynometer to set the strip tension up consistently or maybe look at improving the clutch method somehow. with that said, happy with my purchase and results. Note we settled on 90% speed and dual head for passives. plastic tapes can cause some grief with prick accuracy, we have not worked that out yet but think it was because we got ours without the rod that goes through ther reels, and discovered late in the build that the reels sort of jam and jolt if resting on the fixed rod supports. The inability to control the leaf spring tension under the tape leads to issues with higher devices, increasing tension and causing feed issues. We have decided to stick with the tray for such parts. Also note the large tape feeder on the back "B"" feeders is unusable as it feeds directly into the metal pcb support on the left hand side. Its a great little addition to our workshop.

after many boards: the black shield does not solve all probems. I am still looking for a better solution, but I believe this can only be done by a better software from QiHe.

Question - does this happen with device oriented 90 degrees? Is this an asymmetric issue or does it also happen with vertical edges?Thoughts - these issues happen with detecting package edge and not pad edges - It kind of looks like the brightness is different from top to bottom - LED intensity not balanced. Perhaps if you installed a series resistive pot on a few of the LEDs, you could play with the intensity and get a better balance.I wonder if your black paper circlet should be matt black and more of a cone so as to not reflect. The circlet is angled - perhaps on the boards that fail the circlet is angled the same way?I will have a machine in the next month - and will let you know if I find a solution.

Can a BGA be placed, even manually via the computer, using the two cameras Harry?

what if you replace the camera with better one ?