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

Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.


Solder not reflowing properly

Views: 3974

I am seeing poor quality solder joints, mostly on 0402 compo... - Aug 25, 2008 by dyoungquist  

#56148

Solder not reflowing properly | 25 August, 2008

I am seeing poor quality solder joints, mostly on 0402 components. The problem is that while the pad on the pcb is getting fully wetted, the solder is not flowing up the sides of the 0402 component, i.e. the 0402 lead is somewhat de-wetting. The result is a ball of solder on the pcb pad and very little solder flowing up the side of the 0402 component. I have attached a few pictures to show this.

We are using lead free SAC305 paste with no clean flux. The oven profile is a ramp-soak-spike.

My question is what can be causing this and what is the solutuion to it? I have listed several posibilites below and would like to get your opinions as to which one may be the cause.

1) The 0402 component is not getting hot enough for the solder to flow properly on it while the circuit board pad is getting hot enough for proper flow. It is difficult to attach a thermocouple probe to the pad (lead) of an 0402 component so I do not have actual temperature data of the 0402 component itself.

2) Flux activation time: Excessive time will use up the flux prematurely while too little time would not allow the flux to work properly. Could it be excessive or too little activation time?

3) Am I soaking the boards at too high of a temperature? This would probably lead to activation of the flux too soon, right? Soak temperature is about 180C.

4) Peak temperature of the pcb is in the 235C-240 range with time above liquidus (217C) about 37 seconds. Are these good numbers or do I need to increase them?

Thanks for your advice!

Attachments:

reply »

#56149

Solder not reflowing properly | 25 August, 2008

every time that i have seen that kind of problem where I work it always seems to be a problem with the parts having comtamination on the leads is it just one part # giving you problems our all 0402 parts on that assy having problems if its just one p/n its the part if all 0402 random p/n then it might be your process

reply »

#56151

Solder not reflowing properly | 25 August, 2008

Hello,

Maybe you can get closer to a solution by posting a profile. Your problem points are right where your photos show them. So, do a profile at the same points.

Also, contact your flux provider and request a recommended profile using their flux. Compare your profiles.

reply »

#56153

Solder not reflowing properly | 25 August, 2008

The problem is occuring on more than one part # of our 0402 components. That is why I am looking at the reflow oven process very closely.

reply »

#56154

Solder not reflowing properly | 25 August, 2008

Here is a post of the current profile.

I have contacted our solder paste provider. They give a general lead free reflow profile guide. All that shows is genreral recomendations for temperature ranges and time ranges in each section of a typical reflow profile. No specifics for this particualr type of solder paste.

Attachments:

reply »

#56206

Solder not reflowing properly | 27 August, 2008

Well, it's difficult to play with the profile without knowing the root cause for the problem.

A simple cross-section can tell you whether the component is at fault or not. If it doesn't point out to the root cause then and SEM/EDS analysis on one of the failed components should be able to figure it out.

Regards,

Vlad

www.sentec.ca

reply »

#56209

Solder not reflowing properly | 27 August, 2008

Have you given this information to your paste manufacturer???(Profile, Pics of defects?)

The profile seems to look ok(without knowing paste info...) I think the ramprate looks a little aggressive but it could be the scaling of the profile and the info at the bottom seems to fall in spec. Your soak time and time above all look to be ok too. Is it possible your soak is higher than what is recommended? I think it is also odd because it isn't on a specific reel of components and it is not happening on all different types of components/pads. Is it a multi-layer board? Is there any ground/powerplanes running under your problem areas? Besides that, the only other thing I can think of personally is pcb land/pad contamination or component lead contamination? I am no help but would definitley like to know what you find out. All the best! (maybe try a ramp to peak profile???)

reply »

#56210

Solder not reflowing properly | 27 August, 2008

Does your paste provider recommend a TAL?

For the leaded and lead-free pastes that we use, the manufacturer recommends 60 seconds above liquidous. I note that your profile shows the TAL as 30 seconds, which may not be sufficient.

That said, though, typically we would expect the 0402 components to heat faster, and reflow faster than other thermally dense components on the board. We've had these issues in the past with 0402's as well, and while it's easy to blame contamination, I'm with you on the process review. After looking at the reflow issues, we'd also suggest reviewing the solder paste and placement parts of the process. We've found issues at both in the past that have affected the reflow, especially with lead-free solder. Non-centered parts, insuffient/excessive paste, and insufficient flux (aged paste) have all been addressed in our process...with varying levels of success. Some boards, it works better on, and some boards we have been unable to improve the process yields.

cheers ..rob

reply »

#56213

Solder not reflowing properly | 27 August, 2008

Again, our paste manufacturer specifies TAL to be 30-60 seconds. Out of curiousity, how would part placement cause de-wetting? I am fairly new to SMT so elaboration would be helpful...

Thanks

reply »

#56224

Solder not reflowing properly | 27 August, 2008

Here's a good thread on lead free reflow recipes http://www.smtnet.com/forums/Index.cfm?CFApp=1&Message_ID=47575

reply »

#56233

Solder not reflowing properly | 28 August, 2008

You wouldn't think that placement would affect dewetting or solderability....I wouldn't have either, until it was the only thing left in our process, and when we changed it, things got better.

I couldn't begin to explain why, either. The only conclusion we were able to reach was that a more centered part allowed relatively equal amounts of surface tension on either side of the part. In our case, though, we were only experiencing the poor wetting on one side of the component. Centering the component better seemed to correct the problem for us. Our guess was that the off-center part resulted in additional surface tension on one end-cap...essentially, the beginnings of a tomb-stone.

cheers ..rob

reply »

#56240

Solder not reflowing properly | 28 August, 2008

Have you tried to run the same process but with diffrent solder paste manufacturer.? This give you information whether paste is ok.

Try also print paste on the bare board and run through reflow oven. Check 2-3 boards. Is it the same places where is poor wetting.?

Try wetting test by immersion 0402 components in solder.

I as would recommend setting reflow profile to RSS (ramp to spike) if you can.

janz

reply »

#56256

Solder not reflowing properly | 28 August, 2008

You're using a SAC305 chemistry? Double check your liquidus temp! Your profile shows a liquidus temp of 217 degrees C. SAC305 is not eutectic. (It doesn't transition from paste to liquid instantly). It goes thru a "pasty" stage. We use a temp of 221 degrees C for our SAC 305 liquius temp. You're already on the low side using 217. (less then 40 seconds). I'd recalculate using 221 degrees C. I think you'll find that you need to stay above liquidus (221C) for a few (10-15) seconds longer.

reply »

#56258

Solder not reflowing properly | 28 August, 2008

Couple of points: * We'll tell ya, although it's tough to tell from these pix, but it does not look like dewetting. It looks like nonwetting. * We don't think you're getting this product hot enough. We don't know where you measured the profile you posted, but it barely made it to 240*C. It should be liquidus+20*C for 5 plus seconds. Make it an additional +15*C higher if the board is ENIG. TAL doesn't mean dip. Don't listen to the 'TAL people'. * When you sectioned some of these parts, do you have a good metalurgical bond? Can you post the pix?

reply »

#56259

Solder not reflowing properly | 28 August, 2008

Well, as I've said before, a simple cross-section will shed a lot of light on the root cause :-)

Vlad

www.sentec.ca

reply »

#56263

Solder not reflowing properly | 29 August, 2008

Slow your conveyor down and see what happens. Your profile looks like you just barely get into the reflow zones you need to be.

reply »

#56270

Solder not reflowing properly | 29 August, 2008

Very interesting problem because it shows how systemic this kind of problem can be. So many variables, right?

Firstly, discussions here are similar to everyone's postings, but there are some different observations.

The manufacturer��s profile, we think, is not a good profile; at least it is not as accurate as it should be. Because the profile is relatively low quality, you are basing your profile on poor information, so your results are not good.

Also, the ��T of 9.7 would be unacceptable for us. A ��T of 0.76~1.5 ��C would be normal, and yield an accurate profile.

The profile is not a standard profile, at least in our experience. In Japanese circles, it is a one-mountain type. A two-mountain type is more standard.

In our opinion, although you can get excellent opinions here, our approach would be to contact the solder paste manufacturer for their direct help. They should provide better information and maybe a better profile; if they want to sell you more solder paste. Your reflow manufacturer might also help with this. The reason is because reflow manufacturers know what profiles should be like for different solder pastes, boards, N2 or air, conveyor speeds, etc. It is their business.

Additionally, if you do not already have profiling software, you can ask one to demonstrate how their software would help you solve this problem. This is something you can ask once, but not two times. Some companies�� software will automatically refine your profile for your reflow oven, and solder paste.

Please do post your final results.

reply »

#56274

Solder not reflowing properly | 29 August, 2008

and put a few bucks in your pocket for free.

reply »

#56303

Solder not reflowing properly | 2 September, 2008

small pads like 0402 tend to get hot quite easily. looks like excessive heat is your problem here. pads are wetted, component leads are not. this means solder paste flux on top layer was exhausted before entering in reflow. your profile is the 'in-between' type, not a real linear (no soak) profile, neither a real flat soak profile. go for either one or the other. if you require a soak zone ( to minimize delta T or voiding for example) put one at 180�C flat as possible for 20 seconds. linear ramp to the soak 1,5�C/sec rise. if you use no soak go linear to 210�C at 1�C/sec, then peak. another thing: are the hot air blowers of your oven running at the correct speed? too much air flow on the small pads will exhaust the flux more rapidly.

reply »

#56334

Solder not reflowing properly | 3 September, 2008

Nope. I'd be happy to do the analysis for free just to prove the point.

Regards,

Vlad

reply »


Ron

#56725

Solder not reflowing properly | 28 September, 2008

> small pads like 0402 tend to get hot quite > easily. looks like excessive heat is your problem > here. pads are wetted, component leads are not. > this means solder paste flux on top layer was > exhausted before entering in reflow. your profile > is the 'in-between' type, not a real linear (no > soak) profile, neither a real flat soak profile. > go for either one or the other. if you require a > soak zone ( to minimize delta T or voiding for > example) put one at 180�C flat as possible for 20 > seconds. linear ramp to the soak 1,5�C/sec rise. > if you use no soak go linear to 210�C at 1�C/sec, > then peak. another thing: are the hot air > blowers of your oven running at the correct > speed? too much air flow on the small pads will > exhaust the flux more rapidly.

> small pads like 0402 tend to get hot quite > easily. looks like excessive heat is your problem > here. pads are wetted, component leads are not. > this means solder paste flux on top layer was > exhausted before entering in reflow. your profile > is the 'in-between' type, not a real linear (no > soak) profile, neither a real flat soak profile. > go for either one or the other. if you require a > soak zone ( to minimize delta T or voiding for > example) put one at 180�C flat as possible for 20 > seconds. linear ramp to the soak 1,5�C/sec rise. > if you use no soak go linear to 210�C at 1�C/sec, > then peak. another thing: are the hot air > blowers of your oven running at the correct > speed? too much air flow on the small pads will > exhaust the flux more rapidly.

> small pads like 0402 tend to get hot quite > easily. looks like excessive heat is your problem > here. pads are wetted, component leads are not. > this means solder paste flux on top layer was > exhausted before entering in reflow. your profile > is the 'in-between' type, not a real linear (no > soak) profile, neither a real flat soak profile. > go for either one or the other. if you require a > soak zone ( to minimize delta T or voiding for > example) put one at 180�C flat as possible for 20 > seconds. linear ramp to the soak 1,5�C/sec rise. > if you use no soak go linear to 210�C at 1�C/sec, > then peak. another thing: are the hot air > blowers of your oven running at the correct > speed? too much air flow on the small pads will > exhaust the flux more rapidly.

did you ever get this resolved? I too have seen the same problem, only 0402 show cold solder everything else looks great.

reply »

#57605

Solder not reflowing properly | 15 December, 2008

Sorry it has taken me so long to respond. I got pulled off onto other projects and was not able to work on the solution to the problen until recently.

The solution we have come up with is to change from no-clean flux based solder paste to water soluble flux based solder paste. Using the same oven profile and the water soluble flux solder paste, the solder joints look very good. The solder flows properly on the pcb pad and up onto the 0402 component, i.e. there is proper wetting.

Using the water soluble flux does require cleaning the boards in an ultrasonic cleaner. However, the time it takes to clean the boards is about 1/10 as long as it was taking me to rework the poor solder joints I was seeing with the no-clean flux solder paste.

reply »

Reflow Oven

reflow oven profiler