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

Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.


QFP soldering issue

dwzeek

#26795

QFP soldering issue | 8 January, 2004

We have an ongoing problem with our QFP144 packages soldering. It has not received much attention until late due to some customer failures. Most of the failures would show up in ESS testing and be repaired. The failure rate is about 2 in 1000 parts falling out for one or more leads not soldered. This is a 25-mil pitch part and we use 14x70 mil pads. Apertures are reduced one mil per side with a 5-mil stencil. We use metal squeegee blades and alpha ws-609 (63/37) solder paste (water-soluble). Prints look good about 6 mils high on HASL, same problem on some copper finish PCB's. Profile is solid, meets alpha's requirements.

A couple independent labs came back with what we consider inconclusive reports; one was "possible" lead solderability issue, one was "possible" lead co planarity issues. We investigated the co planarity issue in depth with our machines measurement capabilities. The leads that failed all met the part specifications of 0.100mm.

We have a hard time accepting either as "the" cause and may do an extensive DOE to define optimum process settings for this issue.

Before we get started I was looking for any input/experience others may have had related to this type of issue. I think the high surface area of 144 leads has something to do with it, as in the part tends to float above the solder when it goes liquidous. If it does not wet for one reason or another it will exhibit this solderablity issue.

Inputs?

reply »

Dean

#26796

QFP soldering issue | 8 January, 2004

My first concern would be the choice of solder paste you have. There are better chemistries out there that far exceed the capabilities of the 609

2. Yes, the part will partially float on the solder (assuming plastic qfp). Cross sectioning would show a gap between the base metal and the lead form. The difference is made up of intermetalic layers and tin/lead solder.

3. 2 of 1000 errors. Thats 2000 defects per million, 4.4 sigma process or a yield of 99.8 % If this is the only issue you are having you are doing quite well. This is a typical failure on QFP devices. Lead planarity is and always will be an issue. If the design could be migrated to BGA your yields would be even better.

4. If it was a solderability issue a more aggressive flux might be able to power its way through the contaminent. My experience with ws609 was very poor. Low stencil life, clogged aperatures....and so on. Evaluated many different solder paste...ended up recycling 12,000 dollars in (un-opened)solder paste inventory because the cost of defects due to solder paste issues was many times greater than the cost of the solder paste. This was 4 or 5 years ago...maybe its better now?

5. More aggressvie flux will not solve your lead planarity issue. Test / inspection can "cirlce the wagons" around the problem in the short term. If this is high reliability applications I would push back on the IC house. You may end up with "special" restrictions on lead form and trim requirements...undoubtedly this will cost you with increased component pricing. Most typial responses are test and inspection.

6. You did not mention the lead finish. You may want to look into this. How old are the parts? Grey market parts or new? Makes a diference...especiall, with the choice of solder paste.

Hope this helps. Keep us posted on your findings.

reply »

Dean

#26798

QFP soldering issue | 8 January, 2004

Oh I forgot...is this product wave soldered? Opens can occur if your wave preheat and dwell are not controlled.

I ran into this years ago....wild situation.

Good luck

reply »

Paul F

#26826

QFP soldering issue | 12 January, 2004

I may have a similar problem at the moment with a QFP-68. Do the unsoldered leads leave a perfectly formed footprint in the reflowed solder? Do you usually find this issue occurring towards the corners of the QFP rather than in the middle of a row?

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

reply »

mantis

#26839

QFP soldering issue | 13 January, 2004

Hi All, I have seen this issue on a QFP160.Firts of all as previously suggested if the board is wave solderded make sure that you are not seeing topside reflow as the board goes over the wave.Secondly you should verify your reflow profile.I seen this issue come on one reflow oven and not another even though they were the same oven and running the same profile.IT was caused by a variance in the calibration of the oven leading to one oven running colder than the other. So,I'd starte by Verifying your profile Hope this helps.

reply »

#26882

QFP soldering issue | 13 January, 2004

We have seen this in reflowed parts, where there is flux res collected in the dimple in the solder, where the lead 'presses' in the solder.

Generally, we attack this as a thermal issue: * Thermal pads sucking the heat away from the leads. * Leads not getting hot enough due to poor temperature control.

Try profiling on the lead to eliminate this thermal issue.

reply »

Patrick

#26894

QFP soldering issue | 14 January, 2004

More on QFP soldering issue as per Paul f original e-mail. 1. Problems occured on different PCB's, on different lines using different equipment 2. Flux residue is common in the dimple on the solder. 3. The fault occurred using both "Kester 265" & "Alpha LP 735" 4. Extensive profiles where carried out compairing delta T across the PCB, QFP's and the difference between various leads on the same component 5. Various profiles, with different soak timrs % reflow temp.were use ( as per paste manufactures quideline )

6. No significant issues were noted with respect to lead shape, plating, solderability / wettability, internal package / stress or lead co-planarity. 7. All PCB's have a HASL finish 8. Leads of the QFP's are of "ALLOY 42"

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

reply »

#26895

QFP soldering issue | 14 January, 2004

Good Day,

I have had a similar situation with a QFP. It turns out that the PCB HASL finish was too thin causing oxidation on the copper. The problem was only evident on the QFP as this part had the smallest deposition of paste (and therefore, flux) per lead. We found that fluxing the leads after reflow with an agresive flux and then hitting them with a hot air pencil produced satifactory results. We then purged stock of he old boards, got some new ones and the problem dissapeared.

Hope this helps.

Good Luck

Chris

reply »

Patrick

#26902

QFP soldering issue | 15 January, 2004

Thanks for the reply,

I do know that the HASL finish is "domed " on opposite sides of a QFP site & "smooth" on the other two sides. (ie. Sides 1 & 3 are domed, & sides 2 & 4 are smooth ) depending on the direction the PCB was extracted from solder dipping.

HASL finish domed side 16.8�m average HASL finish smooth side 3.1�m average

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

reply »

#26909

QFP soldering issue | 15 January, 2004

Are your qfp's being stored properly. I have seen this in the past and it turned out to be a moisture issue. We saw most of the problems on the corners. The component would swell during the reflow process thus lifting the corners. After baking the remainder of the parts the issue went away.

good luck!!!

reply »

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

ii-feed SMD Intelligent Feeder