PCB007 Magazine

PCB-Aug2015

Issue link: https://iconnect007.uberflip.com/i/551288

Contents of this Issue

Navigation

Page 69 of 83

70 The PCB Magazine • August 2015 immediately and helps avoid etching of the cop- per. Remember, you have a sulfuric acid solution which will etch copper if no current is flowing. Normally with half-ounce copper this is not a problem, but with today's thinner base copper down to 1-micron more typically being used to produce fine lines, any accidental etching before plating can become a big problem. The amount of tank air you run is also impor - tant in a few ways. I'm not a fan of bottom air- induced tank solution movement. To get the cor- rect replacement copper ion solution flow across the panel from air bubble movement from the bottom to the top of the tank, you need enough air to lift the top of the surface one to two inches with bubbles; any less and you will get panel cor- ner and high current-area burning. The bubbles should not be left on when the tank is not used, as it does speed up consumption of your bright- eners and levelers. Also, the bubble action as they burst in the air puts a lot of copper fumes into the air, which we now need to remove via big fans or your employees and equipment suffer. I do try to suggest solution impingement tanks to my clients; the improvement in plating quality and the reduction in fumes alone in the plating area is worth the conversion cost. Copper ductility is determined by a test done at an outside service, and is affected by extra metals and contaminants in the plating solution, and if the current density is too high it can cause large grain deposits. Ductility testing should be done on a regular basis. Pre-Clean Line Considerations First test to see if you are properly removing the last little bit of dry film smut out of the devel - oper—you may be surprised. A potassium sulfide 1:5 concentration can be used to color copper black but will not color any area that still has dry film smut on it. Apply after developing or after your pre-clean line and examine with a micro- scope for areas that are not fully black; these are caused by dry film residue that did not develop off and will not plate. Complete developing is important to plat- ing success. The first tank in the pre-clean line is designed to help remove any dry film residues (but it is wise to minimize the work it needs to do by improving the developing process). The next tanks are twin or triple cascading rinses. Using counterflow rinses reduce the volume of water needed as well as reducing the amount of chemical carry-over between active tanks. Many companies have a dead tank at the end of double cascading tanks to further reduce water usage; however, stringent bath change times must be adhered to. The next tank is the copper micro-etch tank. Quite a few versions are available, including hy - drogen peroxide-based or sodium persulfate with sulfuric acid. The tank's function is to slightly etch the surface of the copper and remove a few millionths of the copper surface. Copper is very reactive and oxidizes before you can blink, and this oxide can prevent copper plating from prop- erly sticking to the surface, resulting in peeling copper. It is very important to do micro-etch rate coupons in this tank as it quickly varies in etch rate as it absorbs copper. It is the best way to de- termine the etch rate and adjust the process time a few times during the day. (We used to have a chalkboard at the micro-etch tank with the pres- ent dwell time in the tank, and it was tested every four hours and marked down. Operators were required to follow the time accurately—it's that important.) The best etch rate is around 15–20 millionths; below that etch amount you will get lower copper plating adhesion numbers, and above that is not really better as it consumes Figure 2: example of insufficient anode area as well as insufficient air bubbling. FeAture PLATING AND QUALITY ARE CLoSE PARTNERS continues

Articles in this issue

Archives of this issue

view archives of PCB007 Magazine - PCB-Aug2015