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

Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.


Insufficient Heel Fillets on plastic 240 pin QFPs

We are building a monster ... Dimensions are 10" x 7" ... 12... - Mar 21, 2001 by Dennis S  

Dennis S

#5687

Insufficient Heel Fillets on plastic 240 pin QFPs | 21 March, 2001

We are building a monster ... Dimensions are 10" x 7" ... 12 layers ... A Logic Assembly with a Power Supply all rolled into one ... The Power Supply is in one section of the board, the logic in the other section ... The Power Supply obviously has a lot of copper in that area of the board. Convection Reflow process.

We have THREE 240 pin QFPs ... two are plastic, one is ceramic ... And to top it off, we are building to IPC-A-610 Class 3! ... We are having major problems getting heel fillets on the plastic QFPs. (The ceramic QFP has fantastic heel fillets ... we even reflowed the ceramic in the plastic location, and it still was good) ... The solder is wicking right up the pins. We have enlarged the aperatures on the solder paste stencil ... minor improvement ... we have tried to clean the pads with MEK prior to pasting ... minor improvement ... considering Vapor Phase (we have one, management doesn't necessarily want us to use it)

Any other ideas, other than a sharp knife??

Thanks!

reply »


CAL

#5691

Insufficient Heel Fillets on plastic 240 pin QFPs | 22 March, 2001

Selective Laser Soldering? Beamworks and Speedline Electrovert Both demo'd units at APEX. I am partial to the Beamwork machine cause it was able to do BGAs.

Are the QFPs acceptable? You mention cleaning the pads, Stencil aperture enlargement could the QFP have solderability issues?

Cal

reply »

Dennis S

#5692

Insufficient Heel Fillets on plastic 240 pin QFPs | 22 March, 2001

I'm not familiar with Laser Soldering.

I don't believe it to be a problem with the QFP themselves, since the two plastic parts are from different vendors in the same package. We also are having similar problems with other plastic parts on other boards ...

The Class 3 is biting us in the butt. We are getting heel fillets, but not the height of the solder plus one lead width.

reply »

Michael Parker

#5695

Insufficient Heel Fillets on plastic 240 pin QFPs | 22 March, 2001

What's the reality of the situation? Are you creating a specmanship issue here just to satisfy Class 3? Can you demonstrate that the paste volume applied to the plastic QFP's give sufficient overall fillets for mechanical and electrical properties? Would this be enough to get a deviation from spec.?

Secondly, my free wheeling thoughts are clued by the ceramic vs. plastic. You say you get good heel fillets on the ceramic parts but the plastic ones are wicking too much up the lead. You obviously can't change the reflow profiles lower to accomodate the plastic parts. I believe they are getting hotter with respect to the ceramics, therefore the wicking. Have you considered appling a heat sink to the plastic part so that its process thermal properties mimic the ceramic? Try laying a flat piece of metal on top of the plastic QFP during reflow and notice the difference in wicking. It would be a judgement call on your part on what metal thickness and square size would work best. I'll bet you got a pocket full of change, try stacking nickels or dimes until you get what you want. It's worth the experimental try.

reply »

#5697

Insufficient Heel Fillets on plastic 240 pin QFPs | 22 March, 2001

I have a board with QFP240 plastic package exhibiting the same characteristics. Minimum heel fillet that doesn't change with stencil mods, paste change, profile change etc ( I am working to class 2, though, so my pass spec, although they are not pretty )

Although the heatsink idea sounds good,I shudder at the thought of trying to balance a stack of dimes on each QFP.

What type of plating are you using on your boards? Mine are gold plated and my next step is to see if i can wrangle any info out of my board shop

reply »

#5698

Insufficient Heel Fillets on plastic 240 pin QFPs | 22 March, 2001

I have a board with QFP240 plastic package exhibiting the same characteristics. Minimum heel fillet that doesn't change with stencil mods, paste change, profile change etc ( I am working to class 2, though, so mine pass spec, although they are not pretty )

Although the heatsink idea sounds good,I shudder at the thought of trying to balance a stack of dimes on each QFP.

What type of plating are you using on your boards? Mine are gold plated and my next step is to see if i can wrangle any info out of my board shop

reply »

Dennis S

#5699

Insufficient Heel Fillets on plastic 240 pin QFPs | 22 March, 2001

In the end, we may have to go to our customer to see if we can get Class 3 relief ...

We actually did an experiment by adding a ceramic cap to the top of the part, using a thermal paste to improve the thermal connection. It didn't do much for us ...

reply »

Dennis S

#5700

Insufficient Heel Fillets on plastic 240 pin QFPs | 22 March, 2001

Our next plan of attack is to dog-bone the solder aperature. By reducing the width of the solder paste at the heel, we hope to slow the flow up the leg of the pin while the temperature of the pad catches up.

We have boards that were HASL ...

reply »

#5705

Insufficient Heel Fillets on plastic 240 pin QFPs | 22 March, 2001

Hold off on the sharp knife for little bit. [Yano, I get "queasy" when I see much of my blood, but other people�s blood doesn�t bother me much. Weird, eh?] [Oh bye-da-bye, I didn�t win the bar bet on "Ground Hog Day". I didn�t loose it either. The answer is thirty four times. { http://www.khoral.com/staff/ele/play/movies/groundhog.html} Can you believe it?]

Yes Dennis, you are correct. Heel fillets can be a pain in the heel [er there abouts]. But, heel fillets are important, especially for hi-rel, type 3 boxes. * Heel fillet is the most critical part of the SMT solder joint and is responsible for about 70-80 percent of the "strength" of the soldered connection. * Continuous side fillets extending from the heel fillet and along both sides of the lead for about 2X the width of the lead are the next most critical fillets. * Toe fillet is the least critical for the longer foot lengths, like yours.

Areas for investigation are:

Board layout [because there can be differences between the shape of leads on ceramic and plastic parts]: * Do you have enough space on the pad behind the lead to form a good heel fillet? * Is there enough solder available to form a heel fillet of the volume you require?

Differences in the leads on ceramic and plastic parts: * What is the solderability protection on ceramic and plastic parts? * Is it possible that if you increase your time over liquidous that you get good heel fillets on the plastics?

reply »

Dennis S

#5712

Insufficient Heel Fillets on plastic 240 pin QFPs | 22 March, 2001

Ah ha! ... You hit upon something that I discovered earlier today ... As I said in an earlier post, I was going to experiment with dog-boning the solder paste ... But upon further review, I discovered that our customer removed 75% of the pad behind the heel in the latest design change ... which means there isn't much solder behind the heel ... which doesn't give me enough room to do my dog-bone experiment ... I am hoping that we can get our customer to agree to add more pad behind the heel ....

I have also considered increasing the time over liquidous ... but haven't done any experiments thus-far

reply »

#5713

Insufficient Heel Fillets on plastic 240 pin QFPs | 22 March, 2001

Don't mess with the doggie stuff until you get longer pads [and then you won't have to mess with it] because you'll just ruin your lead definition with more paste and still not have the heel fillet you desire.

reply »

Dennis S

#5714

Insufficient Heel Fillets on plastic 240 pin QFPs | 22 March, 2001

ah ... but we had insufficient heel fillets in the old design toom when we had more pad area ... just not as bad

reply »

#5725

Insufficient Heel Fillets on plastic 240 pin QFPs | 23 March, 2001

Consider posting actual measurements of your pads, component leads, and apertures.

reply »

#5728

Insufficient Heel Fillets on plastic 240 pin QFPs | 24 March, 2001

Can you increase the placement force on your P&P machine? It should squeeze the solder paste to all sides including the heel.

reply »

Dennis S

#5733

Insufficient Heel Fillets on plastic 240 pin QFPs | 26 March, 2001

The pads are 0.070 x 0.012 ... as spec'ed, and as measured. The pad extends approximately 0.017 behind where the Class 3 fillet is required to be. (It used to be 0.037 on an older design that exhibited similar, but not as bad of heel fillets.)

reply »

#5737

Insufficient Heel Fillets on plastic 240 pin QFPs | 26 March, 2001

Two things:

1 IPC pad calculator is an excellent resource for you and your board layout staff. Consider that working together with the calculator may allow you both an opportunity to balance your interests.

2 It's unlikely that the ceramic parts have 0.04 greater spacing between leads than the plastic parts. So, there must be something in addition to the pad length inside the leads of the plastic parts that is contributing to the amount of heel fillet [or lack thereof].

reply »

Ken S.

#5738

Insufficient Heel Fillets on plastic 240 pin QFPs | 27 March, 2001

Have you a reflow oven that has upper and lower heaters? Have you a reflow oven that uses hot air as well as elements to control the transfer of heat to various areas? If yes to these, I would suggest increased air flow or change in direction of air flow. Try to increase the temp on lower heaters at the pre-reflow stage or solder curing stage. One last question. Do you use nitrogen ovens? If you are using nitrogen feed in your ovens you will have better solderability. I hope these suggestions will help.

reply »

Dennis S

#5741

Insufficient Heel Fillets on plastic 240 pin QFPs | 27 March, 2001

We have tried to adjust the upper and lower heaters. We have a dillema in this area however. The power supply portion of the assembly has thermal vias in the solder pads underneath larger pads. We have a problem with solder wicking through these holes. It gets worse when we heat the board up. We are in the middle of a design change which puts solder mask on the bottom side of the board. Once the solder mask is in play, we can perform further experiments.

We have also tried to convert our ovens to nitrogen, but our plant doesn't have the proper system at this time to provide enough nitrogen flow.

Thanks for your inputs!

reply »

Metcal soldering rework

PCB Soldering Tools