Selective_solder_design_guidelines

Introduction

As a general statement, design issues that are pertinent to the wave solder process apply to selective soldering. It is assumed the reader has access to the Celestica Design Guide in order to reference the wave solder section of said document. For this reason, the reader will not find each and every fact related to the wave solder process contained in this particular document.

True selective soldering, when compared to the use of a masking pallet and wave soldering equipment (i.e. the most common selective soldering method), offers benefit in the areas of manufacturing yield, process waste and (card) design restriction. The process employs the same (3) basic steps as those used in wave soldering. That is, the card assembly is fluxed, heated and soldered. The major difference is the manner in which the flux and solder are applied. As the “selective” portion of the name implies, the equipment applies flux and solder to (only) specified areas of the assembly. This is in direct contrast to the wave soldering process in which the 2 mediums are applied to the entire card.

Celestica Toronto employs the ERSA Versaflow for selective soldering applications. The machine is available in both “single pot” and “dual pot” configurations. The word “pot”, as one might expect given the nature of the topic here, refers to the solder pot. Design guidelines are identical for the 2 models. That said, the advantages of having 2 solder pots within a machine include cycle time reduction and the ability to select the most appropriate pair of nozzles (as opposed to one) for the application.

The figures and accompanying description found below are merely a guide. They are subject to modification depending on the actual card design. Contact Celestica Process Engineering for detailed information and direction.

Equipment details and limitations

Generally speaking, the maximum (overall) height of a card assembly that may be processed through the selective soldering equipment is 3.93” (100 mm). This figure is provided with reference to the bottom side of the card. Maximum bottom side component depth is restricted to 1.18” (30 mm). The card bottom side, once again, is the reference point for this dimension. Refer to Figure 1 for further detail. Card width is limited to 19.68” (500 mm). However, when a custom carrier or pallet is required, maximum card width is restricted to approximately 17.0" (432 mm) such that hold down fingers, transport rails, fiber optic cable trays, etc. may be incorporated into the carrier. Card length is limited to 23.62” (600 mm). In similar fashion to that outlined above, this figure must be reduced when carriers are involved. Card length is limited to approximately 21” (533 mm) in this case. Note that there are also figures associated with minimum card width and length. This is said, of course, given the assumption that a carrier will not be used. A card assembly must exhibit a minimum width and length of 3.5” (89 mm) and 4.5” (114 mm) respectively. Barrel fill difficulties aside, card thickness is not an issue.

The minimum requirement for lead protrusion, like that for the wave solder process, is .020" (.5 mm). Note that protrusion in the range of .040” to .060” (1 mm to 1.5 mm) is preferred. The inspection process, whether manual or automated, is somewhat hindered by lead protrusion of less than .020" (.5 mm) as the component lead is not visible within the solder fillet. This limit is applicable to any (raw) card thickness given that the component lead diameter may be considered somewhat “standard”. That is, .040” (1 mm) or less.

The card transport system within the machine employs both pin chain and edge roller arrangements. For cards and panels that are to be processed without a carrier (i.e. the preferred method), it is necessary to design the card assembly such that bottom side components (PTH and/or SMT versions) are a minimum of .125” (3mm) from the card edge. Top side components must be a minimum of .160” (3.8 mm) from the card edge. Where custom pallets are employed, edge clearance requirements may be reduced to .080" (2 mm) as the pallet will be designed to satisfy conveyor edge clearance requirements.

Continuing on the topic of clearance, it is worthwhile to note that the card is secured within the fluxing and solder modules via vertically mounted, pneumatically operated clamps. Each clamp is comprised of 3 sections as identified by the characters “A”, “B” and “C” in Figure 1. The clamps impose varying height restrictions with respect to component distance from the card edge. Figure 1 outlines the “step function” limitation within the machine. In similar fashion to aiding edge clearance requirements, a pallet can (typically) eliminate restrictions such that the height of edge mounted components is not an issue.

Figure 1

If (only) 2 opposite edges meet the .125” (3 mm) spacing requirement, it simply makes it necessary to convey the card on these (2) edges. This may or may not be an issue depending on overall card size and aspect ratio. Furthermore, it should be noted that components (typically connectors) that overhang the card edge are not desirable. In many cases, the use of such a component results in the need for a card carrier. Temperature related details

The card top side temperature, during the preheat stage of the process, will peak at 200 to 230 deg. F (93 to 110 deg. C). While undergoing the actual soldering portion of the process, the card top side (in the area above the nozzle) may approach 325 F (163 C). The bottom side of the card and the components thereon will experience slightly higher temperatures. The term “slightly” implies, of course, that the component does not come into direct contact with the flowing solder. Note that the temperature of the solder bath may range from 482 F to 527 F (250 C to 275 C).

Not all (component) materials need be rated to withstand the temperatures mentioned above. Card component menu, component standoff height, base material characteristics and card thickness are but a few items that may or may not result in a component experiencing extreme (high) temperatures.

General design guidelines

Components that employ an interstitial pin design are prone to (solder) bridging and, as such, are not recommended. Components that utilize a lead spacing of less than .075" are, in similar fashion, not recommended. With respect to spacing between PTH pads (i.e. solderable surfaces), .025” (.64 mm) is necessary in order to ensure bridge free results. Note that this figure is consistent with that used for the wave soldering process. Refer to Figure 2 for clarification of this requirement.

PTH or SMT components that are deemed “fine pitch” and are to be selectively soldered benefit from the incorporation of solder "thieves" (or "dummy" pads) in the board design. Thieves should be provided at both ends of the component. This allows Process Engineering the luxury of determining the optimum direction of nozzle travel via experimentation. Refer to the wave solder portion of the Celestica Design Guide for further information regarding solder thief design.

SMT components must not be placed in the (bottom side) area where leads of the PTH component(s) reside. That is, SMT components are not to be located between component pins. If this guideline is not followed, manually intensive measures must be taken to shield these components from the molten solder. These measures typically yield inferior (soldering) results when compared to those obtained with a properly designed card assembly. A much preferred alternative, of course, is to place these parts on the top side of the assembly.

A total of 7 different nozzle sizes are available for use with the selective soldering system. Sizes range from a low of 4 mm to a high of 14 mm. Note that these figures pertain to the outside diameter of the nozzle. A complete list of available nozzle sizes may be found in Table 1. In an ideal situation, the card design would be such that the largest (i.e. 14 mm o.d.) nozzle could be used to solder all components. Heat transfer is increased (when compared to a smaller nozzle) which permits reduced dwell times and, hence, improved cycle times. Furthermore, multi-pin soldering (at one instant) is possible which, again, results in cycle time reduction.

The following diagrams serve to outline the basic requirements with respect to spacing between components that are to be soldered and those that are not. For the most part, the figures listed are the minimum requirements. The designer is urged to allow greater spacing wherever possible. Doing so will result in optimum yields and increased throughput.

Figure 2 outlines the basic (minimum) requirement that .100” (2.5 mm) is necessary between the outside edges of the PTH pad and the neighboring SMT pad(s). The scenario assumes that there are no components whatsoever to the right of the PTH pads. This being the case, use of a large nozzle is not an issue. It is also advisable to leave a minimum component free distance of .200” (5 mm) between the outside of the PTH barrel of the first and last pin within the row and the nearest neighboring SMT component. That is, the component or pad that would be considered above or below (as opposed to beside) the (PTH) component pins. This will permit “over travel” on the part of the nozzle and minimizes the occurrence of (solder) bridging. Figure 2 details this requirement with respect to the top most PTH pad in the diagram. Where PGAs are concerned, the .200” (5 mm) space should be granted at the end of each pin row and column. It should be noted that orientation of SMT components (or alternate devices that are not to come into contact with flowing solder) does not affect spacing requirements. The simple premise is that the prescribed distances between solderable surfaces must be maintained in order to realize optimum soldering results.

Figure 2

Figure 3 assumes the PTH pads (i.e. the component leads) are centered between the rows of SMT components. Table 1lists the spacing requirements with respect to the distance between SMT pads (dimension “A”) on a per nozzle basis. Quite simply, the spacing required is (nozzle diameter + .200”) or (nozzle diameter + 5 mm). This statement also applies to situations where 2, 3, 4 or more columns of pins (a connector for example) are situated between the SMT components. The key is to provide .100” (2.5 mm) clearance between the outermost PTH pads and the neighboring SMT pads on each of the left and right sides AND adhere to the minimum “A” dimension requirements…regardless of the number of component pin columns.

Figure 3

Figure 4 assumes that the PTH pads (i.e. the component leads) are not centered between the rows of SMT components. This being the case, the .100” (2.5 mm) requirement must be met on the left (in similar fashion to Figure 2) while allowing sufficient clearance for the selected nozzle to travel between the 2 rows of SMT components. Dimension “B”, again on a per nozzle basis, translates to (nozzle diameter + .100”) or (nozzle diameter + 2.5 mm). Figures associated with dimension “B” are found in Table 1.

Figure 4

The following table is that which is referred to numerous times in the above section.

Table 1