In a message dated 2/10/99 4:59:12 AM Pacific Standard Time, [log in to unmask] writes: << Fellow Technetters: In your experiences, at what point will one start to see differences in solubility in wave soldering. I monitor our solder composition and am starting to see my Tin deplete as time goes by. I'm not seeing any immediate defects, but am wondering what is the alarm point. None of the other contaminates are out of spec. I just notice my Tin composition depleting and the balance is the Pb. Any thoughts? I can give technical expectations from looking at the Eutectic, but am looking for the experience point of view. Jason Smith Lexmark Electronics Process Materials Engineer >> Hi Jason! I asked the same question a couple of years ago when I was at another company when I was doing the same thing you're doing, and ironically, about the same time of year...ooooh, this is getting spooky! (GRIN) I got a couple of good responses, one from Aric Parr, and the other from Dave Hillman. As far as what level contaminates will start affecting the solder joint quality, there is a good article that was written by a Dennis Bernier of Kester Solder that I'll paste at the end of this email. I got it at Metcal's WEB page since Kester is re-doing their page and it ain't there no more. It answers a lot questions about possible contributors to different materials as well as their effects. -Steve Gregory- Subject: ASSY: Low Tin Level in Wave Solder Author: [log in to unmask] at internet Date: 1/3/97 3:41 PM Good Day Everyone! We just got our solder analysis back (it's from a lab that everyone uses) for our wave solder pots, and we have one wave solder pot that's low in Tin... 61.4% to be exact. The recommendation is to add 4.3 lbs. of pure Tin for each 100 lbs. of solder to return it to the nominal 63%. My question is; just how in tarnation did we get one wave solder pot low on Tin? It don't make sense...we use the same bar solder in all our machines, we wave solder basically the same kinds of products in all of our machines. By that I mean we don't have one line waving only gold plated boards all the time, or only OSP coated PCB's all the time, etc.. so what gives? This is the very first time that I've ever had a solder analysis come back and tell me to put pure Tin into the pot...this is weird. I've had nickel get a little high before, but that's about it. We've got three machines and all of the analysis looks like this: Pot-1 Pot-2 Pot-3 Tin 62.4% 61.4% 62.7% Lead Balance Balance Balance Antimony <0.005% <0.005% <0.005% Copper 0.039% 0.048% 0.033% Gold 0.017% <0.003% <0.003% Silver <0.001% <0.001% 0.002% Aluminum <0.001% <0.001% <0.001% Arsenic <0.010% <0.010% <0.010% Bismuth 0.001% 0.001% <0.001% Cadmium <0.001% <0.001% <0.001% Indium <0.005% <0.005% <0.005% Iron <0.003% <0.003% <0.003% Nickel 0.006% 0.005% 0.004% Zinc <0.001% <0.001% <0.001% The analysis was conducted according to specifications in IPC-S-815B. As you can see, Pot-2 is the "slacker" tin-wise. The only noticable difference that I can see between the other two is that the copper content is just a tad higher...does that have anything to do with anything? Looking at the tin content on the other two pots they're just barely above the limit which is 62.4%-63.6%...is my bar solder vendor starting to get tight with the Tin? Any light that anybody can shed on this would be 'preciated... __\/__ . / ^ _ \ . |\| (o)(o) |/| #------.OOOo----oo----oOOO.--------# # Steve Gregory # # SMT Process Engineer # # The SMT Centre Incorporated # # [log in to unmask] # #________________Oooo._______# .oooO ( ) ( ) ) / \ ( (_/ \_) Date: 03 Jan 97 16:32:11 -0500 From: "Aric Parr" <[log in to unmask]> To: "[log in to unmask]" <[log in to unmask]> (Return requested) Subject: Re: ASSY: Low Tin Level in Wave Solder Tin oxidizes faster than lead. This means that dross has a higher tin content than the solder in the pot. In a previous job, with a very high dross machine, we regularly added tin to the bath. Have the dross production and dross composition from that machine checked and compare it with that of the other machines. The tin is probably coming from the dross. High antimony alloys were used to reduce this differential. I notice that you are using an antimony free alloy. Date: Fri, 03 Jan 97 14:56:51 cst Hi Steve - One possible explanation for your one solder pot to be low in tin is the amount of drossing and/or drossing conditions that it has seen since your last analysis. You will lose tin content from your solder pot as part of the de-drossing operations. Since solder is more tin than lead (ie. 63 tin - 37 lead) and tin preferentially oxidizes versus lead, its almost logical that your dross will impact your tin content to some level. Some testing results we gathered on our dross content (for another reason entirely than low tin) confirmed that this could be one contributor. You should probably check out what method of solder alloy analysis was used too - what is the typical measurement error for the analysis method and could it impact the number you are getting reported as a result. Then again if you were cooking potatoes or getting rid of solder paste on your solder pot then lots of strange interactions could happen!!!! Good Luck. Dave Hillman Rockwell Collins [log in to unmask] The Effect Of Metallic Impurities On The Wetting Properties Of Solder by: Dennis Bernier Vice President, Research & Development Kester Solder Company SOURCE OF IMPURITIES The solder used for this investigation was all from one batch with the following analysis: Element Weight % Element Weight % Sn (tin) 60.1 Fe (iron) 0.006 Sb (antimony) 0.02 Bi (bismuth) 0.006 Cu (copper) 0.004 As (arsenic) <0.01 Au (gold) <0.002 In (indium) 0.005 Cd (cadmium) 0.0003 Ni (nickel) <0.001 Zn (zinc) 0.0002 P (phosphorous) <0.001 Al (aluminum) <0.001 S (sulfur) <0.001 Ag (silver) 0.001 Pb (lead) balance Though the purpose of this testing was to determine the effect of impurities dissolved during the soldering process, it is important to note that national specifications for solder are not strict enough to assure obtaining high purity metal. Secondary or refined metal could contain excessive impurities and shorten the usable life of the solder bath. Impurities such as copper, antimony, zinc and aluminum have an effect on soldering quality and should be kept to a minimum. Assuming that high purity solder is being used, the impurities are introduce into the solder from parts being soldered, from holding fixtures and from the solder pot itself. Copper - Nearly everything on a printed circuit assembly is made of or plated with copper which dissolves rather rapidly in solder. The circuit board itself, component leads and jumper wires all introduce copper into the solder in a wave soldering machine. Gold - No longer used as an overall protective plating, gold is used on certain component leads such as nickel-iron alloy used to make transistors, diodes and integrated circuits. Cadmium - Sheet metal chassis frames and other parts might be cadmium plated to prevent rusting and improve appearance and solderability. Zinc - Brass is an alloy of zinc and copper; so brass terminals, lugs and bolts are sources of impurities. Aluminum - Fixturing devices, bolts and fabricated metal parts might be made of aluminum. The tough oxide film on the aluminum will usually prevent solder wetting; but with multiple solder immersions or if abraded, aluminum can dissolve in the solder. It is doubtful that aluminum will remain in the solder under production conditions since it will dross out when combining with copper, gold or antimony. Silver - Many parts are silver plated to preserve solderability. Like the other coinage metals, gold and copper, silver will dissolve in the solder. Iron - Temperatures over 430 degrees C will cause the solder to dissolve iron from the solder pot itself. An improperly alloyed solder using too much heat could contain excessive iron. A new solder pot -- whether cast iron, cold-rolled steel or stainless steel -- will have exposed iron available for dissolution into the solder. Excessive cleaning of the pot walls with a wire brush can also introduce iron into the solder. The problem associated with iron contamination is excessive drossing which usually clears up as the iron compounds are removed with the dross. Sulfur - It is very unlikely that sulfur would contaminate the solder bath during normal production. Sulfur might be present in secondary metals since it is used to remove copper during the refining process. Sulfur should be limited by national solder specifications to avoid its presence in solder. Phosphorous - The main source of phosphorous is copper that has been deoxidized with phosphorous. PROBLEMS ASSOCIATED WITH IMPURITIES Phase 1 of the investigations involved a compilation of analyses performed over the last ten years for the specific purpose of solving soldering problems. The amount of impurities in the solder was related to observed defects or solder conditions. These defects, specifically cause by contaminated solder, are noted below with some discussion about the impurities which caused the problem. The table following this discussion shows the percentage range of impurities which seemingly caused the observed soldering defects. Icicles, Shorts, Bridges Cadmium, zinc and aluminum in trace amounts increase the surface tension of the solder to cause this defect. Copper and gold increase the solder viscosity to cause the same problem. Large Solder Fillets Copper, gold and antimony increase the melting point of the solder and the intermetallic compounds with tin or lead make the solder more sluggish. The result is larger fillets and more solder consumed to create the solder joint. Unfilled Holes The speed of wetting is reduced by the presence of copper, gold, antimony and cadmium. Though no instance occurred with zinc and aluminum, these metals are likely to also affect wetting speed because of their ability to increase the surface tension of the solder. Dull Solder, Gritty Solder Cadmium and zinc in trace amounts make the solder surface dull. Gold also dulls the surface but is quite often indicated by a sparkling, crystalline surface condition. Bismuth or antimony in large amounts above 2.5% also dull the surface. Copper and aluminum contamination result in a gritty-looking solder surface. Both phosphorous and sulfur have caused gritty solder though rarely are these two impurities found in solder samples. Dross Inclusions Dross inclusions in the solder show up as visible particulate grit or hidden inside a bump or pimple in the otherwise shiny solder surface. Quite often the source of this problem is an unusual amount of iron in the solder. Cracked Joints Inclusions in the solder such as intermetallics of tin or lead with copper, gold and antimony can provide the nucleus for crack propagation. Dewetting Zinc, antimony and phosphorous can cause solder to dewet on copper. By looking at the real world of wave soldering and the ten years of analytical records, we can summarize the impurity levels which traditionally caused problems. Impurity % When Problems Occur Cu (copper) 0.250 - 0.500 Au (gold) 0.005 - 0.200 Cd (cadmium) 0.005 - 0.150 Zn (zinc) 0.001 - 0.010 Al (aluminum) 0.001 - 0.006 Fe (iron) 0.010 - 0.100 Sb (antimony) 0.100 - 1.000 Ag (silver) 0.200 - 2.000 Bi (bismuth) 0.250 - 1.000 As (arsenic) 0.030 - 0.100 In (indium) no data Ni (nickel) 0.010 - 0.030 P (phosphorous) 0.010 - 0.100 S (sulfur) 0.002 - 0.030 Immediately obvious by an examination of this list is the fact that the percentages established by experience are not precise numbers. The explanation for this is that the defects cause by the impurities may be acceptable at one company and cause for rejection at another company. Rigid inspection requirements for aerospace or military products might reject solder joints which are acceptable for consumer products. Difference between fluxes, soldering machines, circuit board density, component layout, hole sizes, solderability and amount of heat all contribute to the quality of soldering. REFERENCES 1.Soldering Manual, 1959, New York, American Welding Society 2.C. L. Barber: Solder, 1965, Chicago, Kester Solder Company 3.H. Manko: Solders and Soldering, 1964, New York, McGraw-Hill 4.M. L. Ackroyd et al. : Tin Research Institute Publication No. 493, 1975, The Metals Society. 5.D. Mackay: Proceedings of Institute of Printed Circuits, Meeting, September, 1972, San Francisco ################################################################ TechNet E-Mail Forum provided as a free service by IPC using LISTSERV 1.8c ################################################################ To subscribe/unsubscribe, send a message to [log in to unmask] with following text in the body: To subscribe: SUBSCRIBE TechNet <your full name> To unsubscribe: SIGNOFF TechNet ################################################################ Please visit IPC's web site (http://www.ipc.org) "On-Line Services" section for additional information. For technical support contact Hugo Scaramuzza at [log in to unmask] or 847-509-9700 ext.312 ################################################################