Conception of 45° Clamping Technology

 By: Dave Corey

 Back in the late 80s, ACS Technologies, in partnership with Lantronic began investigating the possibilities of clamping Printed Circuit Boards (PCBs) at a 45° angle during the Hot Air Solder Leveling (HASL) cycle. The goals set out by ACS and Lantronic were:

• Improving SMT Quad Pack co-planarity
• Decrease solder variation between the horizontal and vertical pads
• Eliminate solder shorts in SMT Quad Packs below 25 Mil pitch  

Preliminary research and test findings were very encouraging and the pursuit for perfect transportation of a PCB at the 45° angle was realistic. However, I would first like to share a little history on the traditional vertical HASL clamping technique and why we felt the need to improve the existing process by positioning the PCB at the 45° angle during the leveling cycle.     

When vertical HASL first appeared on the market, there was a lot of skepticism regarding the overall ability of the process. Similar to most new process introductions, the concerns came from those who sold horizontal levelers. Proving that the concept of solder a consistent thickness over various pad orientations while being applied in a vertical position proved to be an industry challenge. Then, after initial success in the market, the next problem facing the vertical process was the thickness disparity of solder leveling between horizontal and vertical pads. The horizontal process was having problems in this area as well.

         Our customers and we have discovered over the years that the vertical HASL process does in fact work well. Over the last twenty years significant amounts of data has proven that the vertical process equals or exceeds the performance characteristics of the horizontal process. But the real test lies is how well the vertical process stacks up against current user requirements for horizontal to vertical pad solder thickness variance. But first we need to take a look at the normal expectations one should have when it comes to vertical leveling.

        As mentioned above, the data accumulated over the past several years on vertical leveling performance is substantial enough to draw some meaningful conclusions. One of the first such reviews we have done examining the data so we can chart normal expectations of results using the vertical process. In doing so, we choose what we believe are normal operating parameters for our machines. After reviewing our history of leveling performed within these parameters, we then developed a data matrix that compares the horizontal pad thickness verses vertical pad thickness across the PCB consisting of pad widths ranging from 8 Mils to 24 Mils. What we considered normal operating parameters are shown below and a graph showing our results can be found at the top of the next page.

             Solder Pot Temperature:               485° F Lead Solder - 495° F Lead Free Solder

            Clamp Ascent Speed:                    2 Seconds

            Dwell Time:                                        4 to 5 Second

 Air Knife:                                           Temperature 550° F - Pressure 75 to 85 PSI

            From the above graph, you can see the clear relationship that exists between narrow pads and thick coatings and wide pads and thin coatings. This is a universal relationship among all liquids. The smaller the area the liquid is confined to the more height you will get with each drop, or bead of liquid applied. The key issue here, however, is not the solder level within a given pad width, but rather the delta (which we call variation on the graph) between the vertical and horizontal pads of similar width. You will notice from the above graph that this delta starts out low in the 8 Mil pad widths and rises as you approach the 24 Mil pads. This is also normal in solder leveling and occurs whether you use horizontal or vertical leveling process.

          In today’s market the whole issue of horizontal to vertical variation has been reduced to the user’s need for solder height stability on all sides of a quad pack. Since two of these sides are vertical and two are horizontal, the need for minimal height differences is important.

          In summary, you can see from the graph above that the variation between the horizontal and vertical pads across the PCB clearly fall within specifications. We also sorted our past data in such a manner that we were able to see what happens to the solder thickness using different air knife pressure settings. Using pads in the 16 to 18 Mil width range, we used various air knife pressures. Keeping all normal parameters constant, the results are in the following graph.

As might be expected, the higher the PSI used in the air knives the lower the solder thickness. While the above is a representation of nominal values, we caution against using high PSI to solve thickness problems. There are inconsistencies that occur once you go above 85 PSI and we feel that you will find the thickness above this range too erratic for normal processing purposes.

          Given the preceding and resultant data that supports the conclusion, the 45° angle theory will truly meet your current and future HASL needs. Basically, our premise was that if the 90° angle of horizontal and vertical pads is such a major contributor to solder level differential, then the 45° setting had to be an improvement.

         To test our theory, we took a population of PCBs and divided it into three groups. The PCBs were all identical in terms of circuitry and we focused our attention on the leveling performance of the PCB’s 20 mil pitch Quad Packs. One group was ran 65 PSI, the second at 75 PSI, and the third at 85 PSI. For each of these settings, the nominal solder thickness values for horizontal and vertical pads are charted on the next page. For all of these tests, we have chosen normal parameters, except the PSI settings. In addition, the front to back PSI differential between air knives was kept to a 2 to 3 PSI minimum.

As you can see, clamping the PCB at the 45° angle results in an improved Quad Pack co-planarity. Even though the results at 65 PSI were mediocre at best, the improvement over the normal clamping method at 65 PSI on the previous page is worth noting, as it correlates with the improvements seen at 70 and 75 PSI.  At 75 PSI we are starting to make some serious progress; we were able reduce a 300 thickness variation down to roughly 180 thickness variation.

         At 85 PSI, however, we seem to have hit upon something. Driving solder thickness to a nominal of 287 on the horizontal we hit a 272 nominal on the vertical. The test was repeated four times with the following results:

 Conclusion

The goals set out in an attempt to improve SMT Quad Pack co-planarity, decrease solder variation between horizontal and vertical pads, and eliminate solder shorts in SMT Quad Packs below 25 Mil pitch were successfully accomplished utilizing the 45° technique. During this process, our findings led us in discovering a new and radical way of clamping and supporting PCBs during the HASL process. This new clamping system, “Tru Track,” changed the traditional way of clamping PCBs.   

 The patented 45° Tru Track Clamping System is the heart and soul of the Lantronic Model TT30. When leveling the PCBs at a 45° angle, the PCBs are clamped at the top corner while the bottom of the PCB is supported by the Tru Track System. Traditional clamping technique only clamps the top of the PCB and has no bottom support. When trying to duplicate the Tru Track Clamping System, utilizing a traditional clamp in combination with Side Guides, the 18" x 24" panel very unstable during the HASL process. In addition, the 45° angle is not secured.

          Furthermore, the Tru Track Clamping System guides the PCB instead of pushing the PCB into the solder pot. The Tru Track Clamping Systems bottom bar breaks the surface tension of the solder and eliminates the PCB from bowing during the decent stroke giving an even greater front to back uniformity. As stated before, solder shorts were eliminated; thus, ACS was able to successfully HASL 10 mil pitch Quad Packs without any solder shorts.

Simply stated, processing PCBs at the 45° angle utilizing the patented Tru Track Clamping System is far superior than traditional clamping technology, as well as traditional clamping in combination with Side Guides in an attempted to duplicate the 45° angle.