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

Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.


Component Loss in CP643

Views: 8637

Paul Bendemire

#36442

Component Loss in CP643 | 31 August, 2005

Hello,

We're experiencing about 4% loss of parts (primarily 0402 caps and resistors, few 0603s and 0805s), and the error codes are mostly 12040403 and 1CA0A009. The help manual on error codes is cryptic at best. Can anyone help me interpret these losses, and give me a starting point on what things to check/fix to reduce the loss?

Thanks, Paul

reply »

#36452

Component Loss in CP643 | 1 September, 2005

The 1CA0A009 would lead me to believe that the PD's may not be as accurate as possible. If the PD is bad then every location that uses that PD will have about the same error rate. The 12040403 would lead me to believe the feeders are haveing trouble. Usually a few feeders cause most of the problems. This will show up as a few feeders with large error rates You may have 2 problems at the same time but with both of the codes I would first check the feeders. Your feeders need to be almost perfect for 0402's. Try to get a feeder calibrated or buy a new feeder and switch this feeder into one of the problem device slots. If this device starts to work much better then it's feeders. Jerry

reply »


Rob

#36455

Component Loss in CP643 | 1 September, 2005

Or some muppett hasn't put the correct height in the part data, or... The machine has been moved and not calibrated/set up properly - becoming evident as more accuracy is required.

reply »

Paul Bendemire

#36465

Component Loss in CP643 | 1 September, 2005

Thanks to both of you for helping out a newbie.

reply »

#36478

Component Loss in CP643 | 1 September, 2005

We're looking into getting a feeder calibration jig. I know the Fuji is about $14k... Does anyone know about aftermarket? Is anyone using a jig currently for their CP643 feeders?

Thanks

reply »


Rob

#36481

Component Loss in CP643 | 2 September, 2005

Hi Paul,

http://www.alternativesmt.co.uk/productdetail.asp?p_id=2&st=es&c_id=9&m_id=0

http://www.abtech-europe.com - search under equipment - it's called the GAM20.

The GAM20 was about 1/2 the price, but the alternativesmt one could do multiple feeder types.

Hope that helps,

Rob.

reply »

Mika

#36500

Component Loss in CP643 | 2 September, 2005

Check the dimensions of the components in question with a caliper and put the correct numbers in Your part data. Check that the correct nozzle is being used.

Also make sure that you have the right pcb height in Your program. This is somewhat critical for placement of small devices like 0402, 0603 & 0805 etc. If the pcb height is set to big, then the component will not hit the paste on the solder pad and you will end up with loose components anywhere on the pcb. If the pcb height is set to small, then there is a risk of smashing the components to hard during placement and causing them to crack.

reply »

#36859

Component Loss in CP643 | 27 September, 2005

Thanks Mika,

After a couple of months looking at this problem, I have come to the conclusion that mispicks occur at multiple levels. So far, we can account for large errors (> 20%) on particular parts (e.g. 0402 caps) by looking to the part data. After this, we have another threshold (10-20%) that I am continuing to look into. Feeders are suspect, though part data could also be involved, just to a lesser extent than with those parts over 20%.

-Paul

reply »

#36860

Component Loss in CP643 | 27 September, 2005

Ceramic caps are denser than most other components and I"ve found the tops can be slightly rounded. You may want to slow down the turret speed for some parts. Do you ever get a location where there should be a part but there is not? If the part flys off the nozzle shortly before placement the machine might not know the part was not placed.

And be sure to regularly check your vacuum levels, filters and nozzles.

reply »

Mika

#36868

Component Loss in CP643 | 28 September, 2005

Try vision type 12 for capacitors, this will give you much better acceptance in the vision inspection, since those are slightly rounded. Resistors works normally fine with vision type 10. If you have a component that is very hard to get pass the vision inspection, try vision type 254 irregular shape. Also set the pick-up tolerance x-y to 0.5 in your part data. You can also set the body length/width tolerance to 0.3 This is what we do and it works great.

reply »

#36873

Component Loss in CP643 | 28 September, 2005

I am not sure about 643's, but on a 642 machine, the pickup height @ station 1 is critical. Especially on 0402's. The should be a spec, and instruction on how to verify in your maint. manual.

Bad pickup height can cause you all kinds of headaches on these machines.

reply »

#36879

Component Loss in CP643 | 28 September, 2005

I didn't mean where the vision looks. I meant where the nozzle meets the component. I"ve seen a slight doming of these components. Although now that I think of it, Fugi nozzles would be less affected than some other machine nozzles. Assumming a symetrical dome. In any case usually a less than full speed is best for these parts. At least on a high speed turret machine.

reply »

dougs

#36886

Component Loss in CP643 | 29 September, 2005

for 0402's you should be able to run the turret at full speed, if you need to slow it down it means that your machine isn't set up properly. use the vision number that the manual tells you and make sure your feeders are properly calibrated, poorly calibrated/maintained feeders WILL be the issue here, make sure you have the correct nozzle/tape leaf size ( 0.7mm ) for 0402 ( 1mm ) for 0603 if you want to know if it's definetly the feeders thats causing the problem watch the parts go by on the camera at the front when the machine is running you'll see straight away if the pick-up is wrong ( components on their edge, missing, on their end etc etc )

reply »

Mika

#36887

Component Loss in CP643 | 29 September, 2005

Hi Stephen, my mouse have its own will so the wrong [Replay] was pushed. My intention was to share some general programming info. /Regards

reply »

#36938

Component Loss in CP643 | 3 October, 2005

Regarding vacuum levels, how do we check that? Can anyone suggest a procedure?

Thanks, Paul

reply »

#36941

Component Loss in CP643 | 3 October, 2005

Another problem I have seen on CP's is mixing suppliers on the Anti-Marking Illumination Disks. Some after market suppliers disks do not reflect the light the same as Fuji's disks. It is nearly impossible to get the lighting level good enough for reliable vision processing.

I recommend to use ALL of the AMID's from the same supplier. Then perform the center and brightness test and adjust gain and offset as necessary to obtain the proper settings.

Jerry

reply »

#36966

Component Loss in CP643 | 4 October, 2005

The 1st thing I would do is go into trace and start debugging the part this will show you pickup condition

mail me if you want help with this process

reply »

PCB Soldering Tools

JTAG Live PCB Debug Software