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

Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.


Cracked ceramic chip capacitors.

dst

#24085

Cracked ceramic chip capacitors. | 10 April, 2003

Hi,

My company is currently experiencing a high feild failure rate on a 100nF chip ceramic that we use in various applications across our product range. One of them being de-coupling which means we use millions of this device every month.

The failures we are getting back were all built in a two month period where we had changed supplier of the part in question as part of a cost cutting exercise. No verification was done on the part as it was deemed to be a "straight replacement". Sound familiar?

All field failures are with boards being manufactured between October 02 and after Dec 02, exactly the time we used the new supplier.

I have checked profiles, de-paneling methods, z-axis force, z-axis heights, all to no avail. Incidentally, the failures are all over the product range and on different places on each board so tooling hits are not the solution either. I also sent some part off to a local university to be analised. The findings were interesting, they noticed after sectioning that the capacitors displayed high levels of porosity. This to me was a dead give-away, the porosity seen in the new capacitor against the porosity in the old "good" capacitors meant that the new devices were inherently weak, meaning that they could be damaged as they were placed, or as the boards were snapped out of their panel. I was also interseted in the pop-corning effect that is possible with chip ceramics. This would tie in with the porosity I am seeing.

The confusing thing about all this is that the failures are not seen at test. We have full ICT test and then full functional burn-in on all of our products. No failures have been seen in house. Does this mean the popcorning or the micro cracking is causing the device to breakdown over time?

The supplier insists they have not seen the problem with any other customers. We only see the problem over the two months we used this supplier. I am convinced the parts are inherently weak, baring (bearing???!!!) in mind that no other parts we use show have problem. None of them. The nozzles we use are soft rubber and the z-axis is spring loaded. There is no way we can/should be damaging these parts.

Any past experiences of this out there? Also, how do my purchasing department deal with this supplier?

Cheers, Dougie.

reply »

#24086

Cracked ceramic chip capacitors. | 10 April, 2003

I had a similar problem with cracked caps. I checked all my processes and finally found that the caps were cracked in the tape. Check the condition of the caps in the tape. This was unique to one mfg.- EPCOS.

reply »

steve moore

#24107

Cracked ceramic chip capacitors. | 10 April, 2003

Are you gluing the parts to the board? We had a similar situation where the glue beneath the component would add a 'fulcrum' point to the center of the device and when the board flexed the part would fracture.

reply »

Ray Chartrand

#24109

Cracked ceramic chip capacitors. | 10 April, 2003

I've also seen "as received" defective devices. A tier 1 EMS I was working with put down an entire shift worth of these devices before it was noticed. Their rework costs were $$$. Just goes to prove that there's no free ticket and when bean counters drive a copmany, quality goes out the window. Look around and see how many EMS have switched back to battery acid OA flux to overcome solderability issues when cost was the only consideration. Ray A disgruntled process guru

reply »

Joe

#24110

Cracked ceramic chip capacitors. | 11 April, 2003

Hello,

Regarding the failures not showing up at ICT. Failed decoupling caps are notoriously difficult to diagnose within ICT. Check with your test engineering folks and ask them how they've coded for these parts. I doubt that they will say those parts are 100% covered.

Joe.

reply »

Brian W.

#24114

Cracked ceramic chip capacitors. | 11 April, 2003

Cracked capacitors usually are not found at ICT. Unless the cap is completely cracked, the capacitance may not change much, and the usual failure mode is a leakage current. This is usually only seen in ESS type testing, when temperature and humidity fluctuations enter the equation. The test engineers where I have worked have never been able to get an ICT to find leakage current. You might also want to check the ramp rate on your profiles. Ceramic caps do not like high ramp rates. It's not the most common cause I have seen, but I have seen it. I have also found damamged components in T&R. Several years ago I had test failures on some resistors. We couldn't see the cracks. We tested components, analyzed the data, and sent it to the supplier with a CA. The supplier found that their equipment for packaging the components in T&R was causing the problem.

Hope this helps.

reply »

MA/NY DDave

#24117

Cracked ceramic chip capacitors. | 11 April, 2003

Hi

A lot of good answers.

"No verification was done on the part as it was deemed to be a "straight replacement". Sound familiar?"

Yes this sounds familiar, yet I am never sure it is a new supplier or component problem. It could be that the old part and your PCB and your process were in synch.

"The findings were interesting, they noticed after sectioning that the capacitors displayed high levels of porosity."

Now how did they do this porosity check?? Porosity during cross sectioning is intriguing.

"The confusing thing about all this is that the failures are not seen at test."

Test only finds NOW failures, so it isn't surprising for the kind of defect you are seeing to come later as a field defect.

If you have to go back to your old component to save money and effort the best that the buyer can do is to explain that YOU as a Company just Don't KNOW. "Hey it's those crazy engineers" could be his/her excuse.

If the problem is bad I would go back quick to the old component and then analyze at a little more leasure. Years ago I had a similar one that we never figured out and just went forward. The cost of analysis was too high.

YiEng, MA/NY DDave

This message was posted Add this forum to your site! Click to learn more. the Electronics Forum @

reply »

Takfire

#24200

Cracked ceramic chip capacitors. | 18 April, 2003

DST,

What was the crack signature? Was the crack parallel with the internal electrodes or 45� with the terminal electrode? It is also important to determine the orientation of the failed caps on the PCB (in regards to depaneling). Perhaps the caps are cracking due to some type of mechanical stress that was tolerated by the previous supplier's caps.

reply »

#24202

Cracked ceramic chip capacitors. | 18 April, 2003

Please elaborate on "De-paneling method" vs. boards being "snapped out of thier pannel". Read the last post before mine again.

reply »

N.Berkhoudt

#29190

Cracked ceramic chip capacitors. | 21 June, 2004

May be a little late for you, but we experienced the some problems with decouple 100 nF C's. Siemens has reported problem in production in november 1998.

reply »

Heller Industries Mark 5获得了 2015年中国SMT远见奖 - 回焊炉领域

SMT in-printer dispensing