Printed Circuit Board Assembly & PCB Design Forum

SMT electronics assembly manufacturing forum.

SMTnet
»
Electronics Forum
»
SOLDER BALLS

SOLDER BALLS

Subscribe to Thread

Hi, we are facing a problem with a cobar solder paste and i ... - Apr 28, 2009 by Nicoleta

This is a snip from an ealier post here on SMTnet. Gener... - Apr 28, 2009 by davef

Well the only spec is 10.88% flux, 89.12% metal. the probl... - Apr 28, 2009 by Nicoleta

Confirm the source of the issue. Run a coalescence test on b... - Apr 28, 2009 by davef

already did this test. No solder balls. i have printed the p... - Apr 29, 2009 by Nicoleta

Commonly when we see solder balling, we think "too much past... - Apr 29, 2009 by davef

was the first thing we did. No improvements. In the meanwhil... - Apr 29, 2009 by Nicoleta

you have to lots of paste. one makes balls, one doesnt. Th... - Apr 29, 2009 by WaveMasterLarry

sorry. i forgot to mention. we have now two cases! one with ... - Apr 30, 2009 by Nicoleta

Have you contacted Cobar? They should have a sample of your... - May 04, 2009 by Loco

Check out the attached and compare your issue w/ the issue i... - May 05, 2009 by Timothy O'Neill

Nicoleta

#58664

SOLDER BALLS | 28 April, 2009

Hi, we are facing a problem with a cobar solder paste and i need some support here. We have 2 lots of solder paste: one makes solder balls, one no-solder balls. Can you suggest me any analyses that we can performed in order to see the differences between these 2 lots? I will have suopprt from local universities. Thank you in advance. Nicoleta

reply »

davef

#58667

SOLDER BALLS | 28 April, 2009

This is a snip from an ealier post here on SMTnet.

Generally, solder pastes are comprised of: 90% solder powder, 5% flux, 4% solvents, and 1% activators [by weight]. Where: * Solder powder, when reflowed, is a metal bonding agent that mechanically and electrically connects components to the board. * Flux is a chemically-active compound which, when heated, removes minor surface oxidation, minimizes oxidation of the basis metal, and promotes the formation of an intermetallic layer between solder and basis metal. * Solvents dissolve the flux and promote the cleaning action of the flux. * Activators get jiggy wit flux. [Activators are compounds that decompose at soldering temperatures yielding ammonia or hydrochloric acid in the process.]

Having said that, every solder paste is different. Get more specific information from the solder paste MSDS at the manufacturers' [eg, kester, alpha, etc] sites.

There is a fairly extensive "terms and definitions" in the SMTnet Library.

ANSI/J-STD-005 "Requirements for Soldering Pastes" is the industry definition for these materials. You can find it at http://ipc.org

reply »

Nicoleta

#58669

SOLDER BALLS | 28 April, 2009

Well the only spec is 10.88% flux, 89.12% metal. the problem is only at one specific capacitor we have this issue. Sn over Ni finish. Sn-Ag solder paste. I have read all the arhives from the site, but nothing explains me what is happening.everything is checked: stencil-Pcb design, humidity in air 40%, etc. The paste is resin based: cobar Sn-Ag XM3S. This issue is recently discovered. In the past, no probl with this component.

reply »

davef

#58671

SOLDER BALLS | 28 April, 2009

Confirm the source of the issue. Run a coalescence test on both lots of paste.

A simple coalescence test can quickly determine the condition of solder paste after prolonged use. Simply print a small disk [around 4-5mm diameter and 0.2mm thick - a business card with a hole from a standard paper hole punch makes a good stencil] of paste onto a non-wettable substrate [eg, glass microscope slide, solder mask on board, ceramic tile], and re-flow as normal. * A single solder ball in a clear pool of residue indicates good coalescing ability. * Numerous solder balls remaining in the flux residue pool indicates poor coalescing ability and the paste may have become damaged.

Note this is not an absolute test of the condition of the paste. It is only a first-line check and should only be used to confirm paste failures. [based on Multicore Solder Paste Handling Guidelines, September 2004]

reply »

Nicoleta

#58673

SOLDER BALLS | 29 April, 2009

already did this test. No solder balls. i have printed the pads without the component. No solder balls. Only with the component. The problem is, it splashing the around components. solder balls wew found on the body of a connector and on top of it. This is already strange. With the rest of the SMT components there is no problem. Only with this capacitor.

reply »

davef

#58675

SOLDER BALLS | 29 April, 2009

Commonly when we see solder balling, we think "too much paste." As a result, we pinch the apertures or change the shape to a "home plate." We don't think that's what's going on here.

But what if this component has a thicker solder plating on its terminations than we expect? This thicker plating could be combining with the routine amount of solder paste to result in too much solder that causes the solder balling.

Can you reduce the amount of paste that you apply at the pads for this component and then check for solder balling after reflow? For purposes of this experiment, just put tape across part of the apertures this component and print it.

reply »

Nicoleta

#58677

SOLDER BALLS | 29 April, 2009

was the first thing we did. No improvements. In the meanwhile we have change the lot of the components. and it seems that there is humidity absorbed by the components. we are cooking now the lot and waiting for improvemnets. thk u for the support

reply »

WaveMasterLarry

#58678

SOLDER BALLS | 29 April, 2009

you have to lots of paste. one makes balls, one doesnt. Then why bake parts to over come bad paste? Sounds like the paste maker mixed the formula wrong on the one lot of paste. I used to work at a paste place and a lot of hand mixing is done at these places - you'd be surprised. Go visit one before you call me stupid too.

reply »

Nicoleta

#58686

SOLDER BALLS | 30 April, 2009

sorry. i forgot to mention. we have now two cases! one with two different lots of solder paste and one with a bad lot of components. May be u are right. Somebody did something. By the solder ball test on glass no solder balls were found. Only in the process is happening. We were modifing the reflow profile, and nothing. After we have change the sodler paste with a different lot...no problems. SO...we still have the bad solder paste. A lot of it. What happent?

reply »

Loco

#58720

SOLDER BALLS | 4 May, 2009

Have you contacted Cobar? They should have a sample of your batches of pasta they could analyse.

reply »

Timothy O'Neill

#58731

SOLDER BALLS | 5 May, 2009

Check out the attached and compare your issue w/ the issue in the attached. If it's 'beading' versus 'balling' - check thru the punch list and if you've addressed all those potential causes - then it may be paste specific. The Sn/Ag alloy choice is a bit unusual... customer/application specific?

Attachments: