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

Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.


Micro Solder Balls in No Clean Selective Solder Process

Views: 4883

#53108

Micro Solder Balls in No Clean Selective Solder Process | 8 January, 2008

We are building class 3 products, using no clean on wave and selective solder machines. Both processes are producing micro solder balls between the leads of through hole components. I have searched the site and haven't found any recent threads on how people have been able to eliminate the issue. We have tried going to a matte solder mask finish and still see some. Our customer has a zero tolerance for the solder balls and we have not been successfull at 100% removal running the boards through a wash process. Varying the process parameters doesn't seem to have much effect on the number of solder balls that are present. Is there a magic solution that anyone has been able to implement? Is there a specific flux that has been found to bring them down to a minimum?

reply »

#53119

Micro Solder Balls in No Clean Selective Solder Process | 9 January, 2008

Matte solder mask greatly reduces the potential for solder balls, and since you altered the machine parameters with no improvement it might be flux related. Do you use a VOC free (water based) or alcohol based no-clean flux? Do you use machine settings following the flux manufacturer�s recommendations?

reply »

#53120

Micro Solder Balls in No Clean Selective Solder Process | 9 January, 2008

Mask finish is the #1 variable for micro-balling with wave and selective processes.

The #2 variable is the Flux. There are some flux formulations out there who're known to have additives that virtually eliminate micro-balling. However, keep in mind, this could be at the expense of activity.

One that I'll mention is...Alpha's Lonco SLS65. Great for solder balls, horrible for soldering - particularly with forced convection preheat.

reply »

#53122

Micro Solder Balls in No Clean Selective Solder Process | 9 January, 2008

Besides changing fluxes, do you have good knowledge of actual flux volumes? Do you have a micrograms per square value? Pre-heats can affect flux by burning off too soon and also not enough - therefore moisture hits the wave- how's your humidity control- do you pre-bake? I have very low humidity 3/4's of the year, rainy season causes fits.

I eliminated a host of variables by going to a more aggressive flux, solder balls and micro webs disappeared.

Would rather not be brand specific, every process is different, best to find "best of practice" and characterize from there. Since your Class spec.'s are so tight you may not be able to trade off from having a "process indicator"(poor top solder for example) yet micro balls are eliminated.

reply »

#53169

Micro Solder Balls in No Clean Selective Solder Process | 14 January, 2008

Hi its normally down to the state of cure of the resist type not the actual type be it matt or gloss. If its not cured correctly it will produce solder balls. Pending where you are you could try our 35-41 or new 35-41-30 flux this will reduce the problem drastically by 'plugging' up the undercured resist and not allowing tacky plasticizers onto the surface of the resist that stops the solder dewetting off the resist.

Hope it helps Cheers Greg York BLT Circuit Services Ltd

reply »

#53193

Micro Solder Balls in No Clean Selective Solder Process | 15 January, 2008

Based on past, actual, hands-on experience, I have found that the PCB mask finish makes a HUGE difference in solder-balling. Glossy has different surface tension and peelback characteristics than matte finish (ie glossy sucks).

http://circuitsassembly.com/cms/content/view/3138/95/

reply »

#53195

Micro Solder Balls in No Clean Selective Solder Process | 15 January, 2008

We did numerous testing some years ago on matte, gloss and semigloss and found after chemical extraction of the PCB and IR analysis the only difference was the actual cure of the resist. The more plasticizers released the worse the balling If you use 60x scope or more you will see the ball is surrounded by a waxy greasy residue and if you roll the ball off this residue it is completely flat on the side sticking to the resist. worth a look Cheers Greg BLT

reply »

#53198

Micro Solder Balls in No Clean Selective Solder Process | 15 January, 2008

SORRY I did not offer a solution so here goes - The best way to get rid of this issue if you cant change flux is to reduce your flux consumption to a minimal with a spray flux and keep the heat REALLY high to boil off the alcohol before it can be absorbed into the porous solder resist. If you cant do this due to wave fluxing or VOC Free flux then you are pretty well stuck really. If you can stop the alcohol absorbing into the resist you will stop the plasticizer being boiled out over the wave. The reason it wont remove in the wash is that you cannot easily remove the plasticizer from around the ball itself, best bet is a fairly strong Alkali solution hoping to dissolve the Photoimageable residues Cheers Greg BLT

reply »

#53200

Micro Solder Balls in No Clean Selective Solder Process | 15 January, 2008

Yes, the previous "non-balling" flux was alcohol based, but very weak activity.

We found it was worth the switch to water-based, to get prettier solder joints - but...at the expense of having to "brush off" solder balls on a couple of assemblies with close-pitch through-hole connectors.

Also, as I mentioned before, alcohol based fluxes perform poorly with convection preheat.

reply »

#53201

Micro Solder Balls in No Clean Selective Solder Process | 15 January, 2008

Yes, the mask finish is prolly the #1 contributor to solder balling. Seen it many times. It's not that it's cured or not, went down that path about 3 or 4 times as well. Different fluxes can help too. But also check your preheat. If your are getting solder balling on two seperate processes, I would think it's your board and/or your flux.

reply »

#53202

Micro Solder Balls in No Clean Selective Solder Process | 15 January, 2008

Hi �Pb-Head�

I agree with everything you say, except for your statement that alcohol based fluxes perform poorly with convection heat. The fluxes you tried might have performed poorly, but that doesn�t mean that all alcohol based fluxes perform poorly.

Here�s a link to a flux that performs excellent in all aspects when the machine parameters are set correctly (and by the way forced convection pre-heat boost its performance).

http://www.interfluxusa.com/Technical/No-Residue_implementation.htm

Just wanted to put the records straight.

Patrick

reply »

#53204

Micro Solder Balls in No Clean Selective Solder Process | 15 January, 2008

Patrick, you're right. I've only dealt with 2 different flavors of Alpha alcohol-based flux. I've always thought that the convection causes the activator to evaporate more rapidly, but this is probably just the case with older formulations.

Again, I'm basing this off hands-on experience, and not from a magazine article. It may be me, being a dinosaur in this industry (stuck in the 90's). I searched the fine archives and found this actually There's a blurb in there that alcohol-based fluxes are "prone" to being dried out by convection preheats: http://www.smtnet.com//forums/index.cfm?fuseaction=view_thread&CFApp=1&Thread_ID=3968&#Message15733

..but again, obviously not the case with these modern fluxes.

reply »

#53207

Micro Solder Balls in No Clean Selective Solder Process | 15 January, 2008

I found some machine manufacturers put convection nozzles in line parallel with each other instead of staggering them so you get a nice parallel line up the board where there is no flux be it no residue alcohol or VOC Free. The best I could find was low solids rosin as rosin was not volatile. Then again full Convection in wave soldering is a waste of space. If you put loads of VOC Free flux on you need massive amounts of heat to dry it off so dont over flux. Solder Ball we tested identical board from six different PCB manufacturers all produced on same machine parameters, flux and solder resist, Balling difference was staggering, few great some really bad. Cheers Greg BLT

reply »

PCB machines

Electronics Equipment Consignment