Double Sided Reflow Soldering for PCBA: Methods and Process Flow
To fit more circuitry into less space, modern products mount components on both faces of the printed circuit board. A board populated on two sides uses its area twice and allows a smaller final product, so double sided PCBA has become the normal configuration in consumer electronics, industrial controls, automotive modules and communications equipment. Soldering such boards reliably needs more than a single-pass process.
When both sides hold SMT components, the whole board cannot be soldered in one run. The usual sequence is to print solder paste, place the components and reflow the first side, normally called side A, then turn the board over and repeat the operation for side B. The complication is that side A has already been soldered, and it faces the heat of the oven again while side B is being processed: its joints soften and can melt once more, so components with large mass or tall bodies may drop off. Two established methods answer this challenge, and both are explained below.
The first method combines red glue, a heat-cured SMT adhesive, with solder paste; the second prints solder paste on both sides. Neither approach is automatically better. Red glue solves the problems of component height and weight, but it can disturb fine-pitch leads, while the double solder paste method gives the cleanest joints on dense IC boards but demands careful control of the second pass. The best choice depends on the component population, package types, weight distribution and the downstream steps planned after reflow.
The Challenge of Soldering Both Faces
In double sided reflow soldering, the board is heated once for every face that carries solder paste. Side A enters the oven first; the board is then turned over, side B is printed, placed and reflowed, and side A passes through the whole thermal cycle a second time. The joints made on side A are reheated close to the melting temperature of the solder, and the main risk is re-melting: when the joints soften or turn liquid, tall and heavy components lose their support and can slide off or fall while the board is moving through the oven.
The second pass also changes the direction of gravity. During the first pass the components sit on top of the board and gravity presses them into the paste; during the second pass the components from side A face downwards, held only by their joints and by the surface tension of the molten solder. A large connector, transformer or tall capacitor can overcome those forces when the temperature overshoots, and parts can be lost within seconds. Every reliable plan for a double sided PCB assembly controls two things: the method used on each face and the temperature of the second pass.
Method One: Red Glue and Solder Paste Combined
The first method is often called the red glue method, because one face of the board is assembled with red glue, a heat-cured SMT adhesive, instead of solder paste. The other face is printed with solder paste and reflowed first, and the adhesive face is finished afterwards. The glue is screen printed or dispensed onto the board before the second group of components is placed, then cured after placement, and the cured adhesive grips the body of each component mechanically. It fits boards where one side is densely populated with parts of very different heights and sizes, especially when large, heavy components make drop-off the main risk.
The SMT red glue process handles mixed component heights well, because the adhesive pattern is designed for the position and body shape of each part. Glue is applied under the middle of the component body and never touches the solder pads: adhesive on a pad contaminates the surface, so the terminations cannot be soldered later, leaving a component that is firmly stuck but electrically open. The printing pattern and the dispensing program must therefore be checked whenever the component population changes.
Red Glue Process Flow: Order and Temperatures
The process order of the red glue method is fixed by temperature. Side A is soldered first and the adhesive side is processed afterwards, because typical red glue cures at about 180 °C while the solder paste face must reach well above 200 °C to form sound joints. If the glue face were cured first and the board later ran through a solder paste reflow, the cured adhesive would face temperatures it was never designed for: it turns brittle, loses its grip, and many components fall off. The rule is simple: run the hot process first and keep the glued face away from further solder-melting heat.
A complete double sided reflow soldering plan with the red glue method runs as follows: incoming inspection of the bare boards; solder paste printing on side A; placement; QC or AOI inspection; reflow soldering of side A; board turnover; red glue printing or dispensing on side B; placement on side B; visual or AOI inspection; adhesive curing; the final soldering step for the glued face when required, typically wave soldering on mixed boards; and finally cleaning, testing and repair. Each inspection gate costs little compared with finding the same defect after final testing.
Method Two: Solder Paste on Both Sides
The second method prints solder paste on both faces and reflows each side in turn. It is the preferred solution when the board carries many components on both sides, and it is almost unavoidable when both faces hold dense fine-pitch ICs or BGA packages, the population typical of SMT PCB assembly orders. For such parts the red glue method is risky: adhesive under a fine-pitch body can lift or shift the component until its pins no longer align with the pads. Solder paste behaves differently: components settle onto the pads during reflow, and the surface tension of the molten alloy pulls fine-pitch leads and BGA balls into alignment as the joints form.
The double solder paste flow is: incoming inspection; solder paste printing on side A; placement; QC or AOI inspection; reflow soldering of side A; board turnover; solder paste printing on side B; placement; QC or AOI inspection; the second reflow; then cleaning, inspection and repair. Because both faces are really soldered, the board undergoes two complete reflows, and the final quality depends on the second profile as much as on the first, which is why this method needs the temperature control described next.
Controlling the Second Reflow Pass
The second pass must achieve two opposing goals. Side B faces upwards and must get hot enough for its paste to melt, wet and form reliable joints, while side A faces downwards and should stay below the point where its joints melt again. If side A is overheated, the first-pass joints turn liquid and heavy components fall; if side B is under-heated, the new joints are dull and unreliable. The answer is to set the second-pass peak slightly below the first-pass peak, commonly by roughly ten to twenty degrees on lead-free lines, so the upper face reaches its full wetting window while the lower face stays below the melting point of the alloy.
Before production, the reflow soldering temperature profile should be measured with thermocouples attached to real boards of the product, in dense areas and close to the heaviest components, because thermal mass delays heating locally. The aim is a profile in which side B reaches its full soldering window while the measured temperature on side A stays below the re-melting point of its solder. The same logic protects red glue boards: their adhesive side is cured below 200 °C after side A is reflowed, so the first-pass joints are never reheated to melting.
Choosing the Right Method for Your Board
The choice between the two methods should be made from the bill of materials before production starts. Count the fine-pitch ICs, BGA packages and connectors, and list the heavy and tall components with their positions. If both faces are dense with IC packages, solder paste on both sides is the safer route; if one face carries fewer parts or a mix of tall and heavy components of different heights, the red glue method protects that face. Boards that also carry through-hole parts are often planned as mixed technology PCB assembly, where the SMT face is reflowed, the second face is glued, and a single wave soldering step completes the remaining joints.
Layout decisions made early make both methods easier to run. Placing the heaviest components on the face that is soldered first keeps them on top during their own reflow and held by solid joints when the board is turned over, and keeping tall parts away from the panel edge reduces stress during transport and depaneling. A disciplined PCB design layout also leaves fine-pitch devices clear of high components, with room for stencil apertures, placement nozzles and AOI cameras. Board thickness, panel size and batch volume matter too: thin panels warp more easily on the second pass, and very large panels make the temperature difference between center and edge harder to control.
Inspection, Cleaning and Common Defects
Both methods depend on inspection at the right moments. After placement on each face, a QC check or AOI run catches missing parts, shifted chips and misplaced glue before the board is heated; after reflow, inspection catches bridging, insufficient joints, solder balls and open joints. Hidden joints under BGA packages on the lower face need X-ray inspection, and the finished assembly must be verified electrically. A complete PCBA testing plan therefore covers both faces and is agreed before the first production batch runs, not improvised after a failure appears.
Two-sided boards should also be cleaned on both faces, because flux residues from two reflows can remain under components and beside fine-pitch leads, attracting moisture and eventually causing leakage or corrosion. The defects seen most often in double-sided production are dropped components from the first-soldered face, brittle or missing glue, open joints caused by adhesive on pads, and misaligned fine-pitch parts caused by uneven glue thickness. Each of them traces back to the process controls described above, which is why profiling and inspection gates are never optional on a double-sided line.
How gopcb Runs Double-Sided Reflow Lines
gopcb applies the same logic to every double-sided order. During the quotation review, the engineering team studies the component list, package types and weight distribution of the board, and the chosen method is written into the manufacturing plan before production starts. Stencils, adhesive patterns and placement programs are prepared for both faces together, and the reflow profile is validated on instrumented boards at the start of every run. Each face is inspected at the gates described above, and the results are kept with the batch records.
For customers who want the entire chain handled by one supplier, this planning can be combined with turnkey PCB assembly, so the design files, components and production records stay under one roof. Send gopcb your Gerber files, bill of materials and expected volumes for a free DFM and manufacturability review, and the team will recommend the best method and reflow strategy for your product, with a quote that reflects the real cost of running both faces reliably.
