Printed Circuit Board Assembly & PCB Design SMT Electronics Assembly Manufacturing Forum

Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.


Programming help for GSM

#27847

Programming help for GSM | 31 March, 2004

Minor problem with rejected parts on a GSM C4 single flex head. UIC support was helpfull but I thought some of you may have encountered this problem and know what's the matter. We are rejecting 1 out of 3 SOT-63 parts. The problem is occuring to parts presented to the upward looking camera with the leads pointing up in the monator. The first pick is presented this way and rejected. The next two picks are presented with leads pointing down and are accepted. Placement on boards is at 90 degrees for two locations. Pre orient is turned on. With pre orient on these parts should only be presented to the camera with leads down in monator. Why is the first part being pre orented backwards. Is there something I'm missing. I will call UIC again tomorrow but I thought sombody may know what's up.

reply »


RDR

#27859

Programming help for GSM | 1 April, 2004

Are the parts coming out of the same reel?

reply »

#27865

Programming help for GSM | 1 April, 2004

Make sure that the image of the component in matches the way the actual part is on the feeder. To do this just go to your feeder status on the production screen, then click on the feeder that you want, in this case the one that has the sot-63 and now you should be able to see the way the GSM is expecting this component to be on the feeder. It must image must match the actual orientation on the feeder.

How many reels with the same part are you running?

.

reply »

paul pmd

#27877

Programming help for GSM | 2 April, 2004

Yes, The feeder graphic shows the leads at 180 degrees. The part is in the reel at 180 degrees when feeder is mounted on front side. The component graphic has the leads at 0 degrees. All parts are comeing from the same reel.

reply »

#27880

Programming help for GSM | 2 April, 2004

One thing that it doesn't make sense, I have never seen the GSM picking up one component leads down, present it to the camera leads down and reject it. Just so it can pick two more components from the same reel and present it leads up and accept it. This is what I would do, I disable the pre-rotate. I would then teach the component with the camera. After this, it should not reject anything. Assuming once again, that the image on the feeder status matches the orientation on the part itself.

Good luck and let me know if it helped.

.

reply »

#27892

Programming help for GSM | 2 April, 2004

You may also want to start on the component library. Hypotetically speaking, the GSM shows the component as follows in the component library: One lead to the right, 3 leads to the left. This is the feeder default orientaion and it will be 0 degrees. Now if the actual part on the feeder is one lead up, 3 leads down then you need to change the feeder rotation by 90 degrees until it matches the actual part. Once the graphic shows the exact same image as the one that you have on the feeder then the GSM will pick it up and present it this way. I am really curious to know how it was that you resolved the problem.

.

reply »

#27899

Programming help for GSM | 2 April, 2004

Jaimebc is right. You must match the orientation in the Component ID (Default Rotation in Feeder) to the rotation in the Feeder List and the actual orientation in the feeder itself for the vision to properly recognize the component.

This still doesn't explain why it's accepting 2 out of 3, but I'm sure many of you have also seen the GSM's do some strange things.

reply »

#27957

Programming help for GSM | 7 April, 2004

Thanks to all of you. As it turns out,(pun), the problem was with optimization. The SOT-63 Was being picked with other parts also set to pre orent when orignally this part was not. The program was then optimized. The other parts on the other heads caused this part to pre orent with them. Later pre orent was turned on for this part and the program was not re optimized. The GSM looks at the first nozzle for pre orent and if it finds a part that is, then all heads rotate to that orentation. If it dose not then it looks at nozzle 2 and so on. The problem was differant parts with diferant orentation were set to pre orent within the same mount block. Thanks again.

reply »

Boundary Scan

Large PCB Dispensing System