IP3 QFP Problems - Has anybody had the same probs?| 2 June, 2000
I am currently experiencing several problems with QFP's on one of our IP3's. The first problem is with QFP240 devices. We are experiencing a high reject rate with these parts when using vision type 100. What appears to happen is that the vision systems thinks that the part has been picked up skewed in theta. When it tries to process the image, it looks for the leads in the wrong area and fails. This happens to about 30 to 40 percent of the parts picked. I was able to get the part to place consistently by using frontlighting processing, but I have to sacrifice proper lead inspection. Fuji has been unable to resolve this problem to date. The other problem which I have is that I get random shift in both the X, Y , and theta directions on all fine pitch devices. I have had fuji in to look at the machine on 4 occasions. I have also swapped X, Z, and Q servo amps, checked the Z-height calibration, and verified program data and fiducial coordinates. Sometimes the shift is 25mils or more. Also verified proper board clamping. Has anyone seen either of these problems or have any ideas on how to fix them? Any input would be greatly appreciated.
Re: IP3 QFP Problems - Has anybody had the same probs?| 5 June, 2000
...well, if I had Fuji in on 4 occasions and the problem is not resolved I would wonder what exactly I paid for...
Here is what I would do.
1. Run the glass plate program. This will eliminate variables in your production parts and boards. Additionally, by standardizing your carrier you can troubleshoot your problem more easily and determine mount accuracy at the same time. Fuji should of done this to dtermine if that machine is truly calibrated to within published specs. (somewhere around +/- 0.1 to 0.025mm depending on your camera configuration).
2. I would check the backlash in the Q axis. This is a tricky axis to calibrate. If done incorrectly your mounts would be all over the place.
3. Verify your camera(s) are securely mounted to the machine. Sometimes its something simple that is easily over looked. Also check the fid. camera
4. There was a service bulletin some time ago about wandering mounts. It seems by adding a phantom fiducial the problem was solved. Check their web site for details.
5. What are the error codes your getting on this qfp? This will point the way to solving your problem. What are you seeing in trace mode? Are you getting to element check? If not there is a fundamental error which, if not true, will automatically stop the VP of the part and report an error code. This code will help you (in theory).
6. Is the part in focus? If not the scale of the part will be altered and this will be most noteable on fine dimension parts like lead elements. Verify the Z dimnesion is correct in your program for the part.
This machine is more than capable of placing this part accurately. Keep at it and keep us up to date...