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

Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.


Pinhole Solder Joint Reliability

iman

#25182

Pinhole Solder Joint Reliability | 15 July, 2003

We have an internal engineering study on blowhole solder joints and had helped reduce the problem to pinholes symtoms. The pinhole occurs in solder joint of a Land-Grid-Array (LGA) side-wall (external perimeter of package) fillets.

Our customer is asking if its possible to reduce the pinhole/blowhole issue to ZERO levels? We are looking into it now.

To better understand the reliability issue on hand, we were wondering : 1) if anyone has the experience (eg. DOE) to reduce LGA pinhole solder? or,

2) is there any written document on the process/assembly subject of blowhole/pinhole? or,

3) is there any common industrial stand point that pinhole is detrimental to solder joint reliability? (two books we read thru, written back since the mid-80's, say pinhole is cosmetic level?)

appreciate some expert advise or guide towards the correct direction. thanks in advance!

reply »

sam

#25194

Pinhole Solder Joint Reliability | 16 July, 2003

I have done a DOE on the SMT components, not the BGA, for reducing void within the solder.

The result did not help in the void reduction, but confirmed that the speed of the IR oven, solder paste thickness may be the important factors for the void.

I belief that the brand of solder paste and the stirring of it should be another factors to be considered for the DOE.

Just hope it can give some ideas. If you proceed with the DOE experiment, can you share the result with us too?

Sam

reply »

Don Bell

#25237

Pinhole Solder Joint Reliability | 21 July, 2003

Blow holes/ pin holes in solder joints are the result of multiple root causes. Suggested analysis might look like this;

1) What method of solder preservation is used on the bare board. (hasl,enig, osp, etc.) HASL can leave unwanted flux residues, gold over nickel should have less issues with residues. 2) Determine bare board cleanliness by ion chromatography method. Expensive but revealing. You may find that the boards aren't as clean as you thought. If contamination levels are unacceptable, work with your pwb supplier. 3) Some voiding can be reduced/eliminated by increasing the amount of time the assembled board spends in the "soak" zone of your reflow oven. (Slow your oven's conveyor speed but remember to adjust the thermal set points if necessary.) Blow holes indicate that you're still getting significant volitilization as the solder joint is solidifying. 4) Moisture adsorption into the solder paste is another possibility, but is usually driven out during the soak stage, and leads to lots of solder balls being thrown around. If your factory climate happens to get near or at dewpoint, moisture can form on your stencil and get into the paste that way. Check your environmental controls. 5) Components can contribute to blow holes if they are contaminated. Send some samples to a lab for analysis.

There are more sources but try looking at these. Hope this helps.

reply »

iman

#25246

Pinhole Solder Joint Reliability | 21 July, 2003

Don,

thanks for the thoughtful insight, can I run some thinking points for your further assistance?

1) ours is a Au over Ni PCB bareboard. Do you think excessive Au thickness is any way a influence for poor solder formation? I did notice the PCB bareboard has tendency to have slightly misalignment green solder mask layer, I was wondering if chemicals could be trapped under the green solder mask "recess" and be a negative influence during the solder reflow process?

2) point #2 is gravely noted, unfortunately we have large stock of such PCB. dunno how my side is going to work around it IF the lab tests prove PCB is "unclean". Thanks.

3)could you help elaborate further on "volitilization" during solder solidification, pls? Does it mean the temperature(degC) is too hot or reflow(sec) duration is too long?

We based the original profile on the paste vendors' "recommended" profile specs but it didn't help at all. We now have a newer current profile that is slightly below the vendors' "recommended" profile specs, and is giving less% of pinholes. But we are looking to eliminate 100% pinhole in solder joint when the PCBA is visual scope-inspected at 20x.

4) we checked our environment controls at the start. 40~60RH% and 22~26degC seems about ok?

5) good pointer! thanks for the tip!

reply »

#25247

Pinhole Solder Joint Reliability | 21 July, 2003

Theory #6 [maybe 3a?]: If you change your paste to a 'non-voiding' paste, your problem will go away. It's the flux in your paste that's killing you. The flux gets trapped in the solder and then releases ejacta that form the volcanic openings in your solder.

reply »

iman

#25248

Pinhole Solder Joint Reliability | 22 July, 2003

Dave F,

thats' the path of thought we are now bumbling along, but no way is the paste vendor is going to admit straight out to any such affirmation.

Any literature online on this possible root cause? as it helps to gain substain-able theory in hand for either customer/vendor discussions. Thanks!

reply »

#25250

Pinhole Solder Joint Reliability | 22 July, 2003

Iman: Don't be too quick to judge. Most paste suppliers offer a 'non voiding' paste, based on their customers' experiences with BGA voiding.

Our speculation is the mechanism for what you are seeing in LGA, QFN, whatever is similar to BGA voiding; except that the entrapped flux / flux residues do not remain in the solder as a void in BGA solder balls. Instead, it becomes ejacta through the side of the solder connection. [As an aside, probably there is substantial voiding in these solder connection, in addition to the more obvious pin holing that we're talking about.] Returning to your pin holing, it probably ejects near the center of the long edge of the solder connection more often than other areas for two reasons - lower surface tension and largest area.

BGA voiding is fairly well documented: * Google => bga void doe * Dr. Lee at Indium Corp. has conducted a number of investigations concerning solder joint voiding and BGA components. Check Indium's site. [As an aside, Lee's book "Reflow Soldering Processing and Troubleshooting SMT, BGA, CSP and Flip Chip Technologies" (Dr. Ning-Cheng Lee; Butterworth - Heinemann; 2002; ISBN: 0750672188) is worth owning.] * Further, Dr. Lee hosted an "On-Board Forum" here on SMTnet a while ago. There may be some discussion there.

reply »

iman

#25255

Pinhole Solder Joint Reliability | 23 July, 2003

Yo Thanks!

right on the money pad, the pinhole occurs mainly in the (fillet's) top-surface central of the whole solder (fillet surface) area. Like you said, large surface area, lowest surface tension.

that's a possible cause, that the potential voiding-crap stuck inside the solder erupts out of the solder fillet and marr the solder surface as a sneaky pinhole or flamboyant blowhole.

Do you think it is an excessive dwell degC/sec? during either preheat or reflow zone, thats causing this ejacta behaviour?

reply »

#25256

Pinhole Solder Joint Reliability | 23 July, 2003

There is much written on the effect of reflow recipe on voiding behavior. For instance: 2 (EP&P 10/98) 2a Number of small and large voids correlate. 2b Reflow time (temperature) is (are) very significant in void formation. Soak time is not a factor. 2b1Increasing reflow temperature increases voiding. Temperature influences void formation 8.4X times greater than reflow time 2b2 Increasing reflow time decreases voiding 2c Small voids near the base (top) of the ball, increase reflow time (60 -> 100 sec) and decrease reflow temp to 205C.

reply »

PCB contract manufacturer