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

Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.


Solder Joint

Views: 3221

#47020

Solder Joint | 26 January, 2007

Taking a pole for s&g's. Just got out of a meeting and the customer said the solder joint of the parts only occurs at the fillet. "Without the triangular part of the solder there would be no solder joint." These where actual words spoke. I know differently and was forced to bite my lip, but just wanted to make sure I'm not nuts.

So what do you think?

reply »

#47021

Solder Joint | 26 January, 2007

Chunks! Good to see you here. From what I've been taught, and from what I've read, the strength of the solder joint is in the intermetallic bond which occurs at the atomic level. A solder FILLET is an indicator that wetting took place.

reply »

Bob R.

#47032

Solder Joint | 26 January, 2007

Whether there's a fillet or not depends on the component type. Most have a fillet but some, such as QFN's, don't necessarily have one. Have a look at IPC-610D Workmanship Standards to understand when you should expect one and when you shouldn't.

Typically, fillets do add considerably to the reliability of the solder joint. Take, for example, a resistor being thermal cycled and failing through solder fatigue. The crack will start from underneath the part and propogate outward towards the fillet. The extra solder of the fillet is just that much farther the crack needs to go before an electrical open. The required size of the fillet is application dependent and is spelled out in IPC-610D (class I, II, or III).

reply »

#47055

Solder Joint | 29 January, 2007

I think your customer needs to look at this:

http://www.necel.com/pb_free/en/crosssec.html

reply »

#47065

Solder Joint | 29 January, 2007

Maybe your customer ment to say something like: reliability in actual use does to a large extent depend on the heel fillet and the 'bottom flat' wetting. If the foot length is <3W, then also on the presence of a toe fillet, because short feet 'rock' during thermal cycling. Side fillets are less important, unless you have wetting problems.

Or maybe your customer said something that she knew was a bit unaccurate just to test you. She's probably talking smack on you right now.

reply »

#47071

Solder Joint | 29 January, 2007

Well, they're all "HE"'s, which may be part of the problem (good one Dave) The customer "thinks" there may be a problem but has no actual proof. The part in question is a lug terminal. It's 0.145" by 0.175" and 0.020" thick. They couldn't "see" a fillet with the naked eye so assumed it would fail. IPC J-STD-001C covers it in section 9.2.6.12 but since th pad we have to solder to is the slightly samller than the part (as designed by the customer), you won't get a fillet.

reply »

#47072

Solder Joint | 29 January, 2007

Thanks Steve but I feel the customer may get scared by such technology as cross sections. Remember Cave Man Lawyer on Saturday Night Live? I have Cave Man Customer.

reply »

#47074

Solder Joint | 29 January, 2007

Bring out your Mr. Peepers outfit (I suggest adding a top to reduce distractions) and a couple of apples and maybe you'll find a happy place in the middle.

reply »

#47082

Solder Joint | 29 January, 2007

I'm confused, How can your customer possibly EXPECT a fillet if the pad is smaller than the part? I thought everyone was over the whole toe fillet thing many years ago. Wasn't there some papers from NASA nearly ten years ago that advocated the "less is best" approach? This included PTH where a top side fillet was deemed unnecessary and even detrimental?

reply »

#47083

Solder Joint | 29 January, 2007

Continuing with Darby's point, the surface area of the pin that is in contact with solder is HUGE. You could pound that pin with a 24oz ball peen hammer and not get it to release from the copper in that through hole.

reply »

CK Flip

#47092

Solder Joint | 30 January, 2007

I think most people don't "get" soldering, and advocate the "more-is-better" approach as if solder were duct tape - you know...the more duct tape you put on something the stronger the bond. Such is NOT the case with soldering. People don't "get" that solder is the joining of metals (hence the name solder joint), and that the strength of the solder joint is a function of the unit area of intermetallic you have, which in turn is a function of your part and pad areas.

In Chunk's case, a lug terminal is sitting on top of pad, there's paste on pad, so now, you have 0.145" by 0.175" square area of intermetallic, which if you do the math, will be more intermetallic than whatever toe or heel fillet that you WOULD get with this type of part.

reply »

#47098

Solder Joint | 30 January, 2007

Yup, you guys get it. My customer feels that the fillet is the ONLY joint made (seriously). They fail to realize the faying surfaces bond as well. I believe it's because their theory is "If we can't see it, it doesn't exist!" mentality.

reply »

#47104

Solder Joint | 30 January, 2007

Your customer needs to be beaten sensible with a copy of IPC-610. Strapping it to a 2x4 may increase efficiency.

reply »

#47110

Solder Joint | 30 January, 2007

Sad part is THEY specified IPC J-STD-001C!!!

reply »

Plasma prior Conformal Coating

Reflow Ovens thermal process improvement