Getting a discrete semiconductor to stay put on a board sounds like the easiest thing in the world. Stick it in the hole, solder it, done. Except every experienced technician knows the truth — leads pop out, joints crack, parts shift during reflow, and boards come off the line looking like a war zone. The difference between a board that survives shipping and one that fails in the field often comes down to how you fix those leads before the iron ever touches the joint.
Most people skip this step entirely. They bend the leads, drop the part in, and hope for the best. That works fine for a prototype in your garage. On a production floor, it is a disaster waiting to happen.
When you push a component through a wave solder machine or a reflow oven, the board vibrates, the solder melts, and surface tension takes over. If the lead is not mechanically locked to the pad, the component floats, shifts, or falls off entirely. A diode that ends up 45 degrees off its pads still has solder on it — it just does not connect to anything. You will not catch that with a visual inspection. You will catch it during functional test, which is the most expensive place to find a mistake.
Fixing the leads properly before soldering takes ten extra seconds per component. Those ten seconds save you hours of rework later.
For through-hole discrete semiconductors like diodes, transistors, and voltage regulators, the way you bend the leads before insertion determines whether the part stays put or walks away during soldering.
Take a small signal diode and bend both leads outward at a 45-degree angle about 2 millimeters from the body. This creates a mechanical lock — the lead presses against the inside wall of the hole and the pad on the bottom side holds it from below. The part cannot pull out without bending the lead, which means it stays put during wave soldering, handling, and shipping.
Do not bend the leads too close to the body. Less than 1 millimeter and you risk cracking the glass seal on glass-encapsulated diodes. More than 3 millimeters and the mechanical advantage drops — the lead can still flex and pop out under vibration.
Power transistors in TO-220 or TO-247 packages have thick leads that do not flex easily. A simple outward bend is not enough to hold them. Instead, form a tight U-shape on each lead so the tip points back toward the component body. When you insert the part, the U-shape springs open against the hole wall and the lead tip grips the pad from underneath.
This technique works especially well on boards that go through wave soldering. The U-bend creates enough friction to resist the upward force of the solder wave, which tends to lift tall components off the board.
For tall discrete parts like power regulators and large transistors, bending alone is not enough. After inserting the component, flatten the lead tips against the pad with the iron tip or a flat tool. This is called staking. The flattened lead cannot spring back, so the part is locked mechanically even if the solder joint fails.
Staking also improves heat transfer during soldering. The flattened lead has more contact area with the pad, which means the joint heats faster and needs less iron contact time. Less time means less heat soaked into the semiconductor junction. It is a win-win.
SMT discrete components do not have leads you can bend. They have termination pads that sit flat against the board. The fixing challenge here is entirely different — it is about keeping the part from tombstoning, shifting, or floating during reflow.
If the pads on your PCB are too small, the component will shift no matter what you do. Follow the land pattern recommendations for each package type. For SOT-23 transistors, the pads should extend beyond the component body by at least 0.3 millimeters on each side. This gives the solder something to grab onto and prevents the part from sliding during reflow.
For SOD-323 and SOD-523 diodes, keep the pad-to-pad spacing consistent. If one pad is larger than the other, the solder melts at different rates on each side, creating the surface tension imbalance that causes tombstoning.
When hand-soldering SMT discrete parts, never try to solder all pins at once. Place the component with tweezers, tack one corner pad with a tiny blob of solder, then check alignment under magnification. If it is off, reheat the tack and reposition. Once the first corner is locked, solder the remaining pads one at a time.
This tack-first method is the single most effective way to prevent misalignment. It costs you an extra five seconds and saves you from pulling the part off and starting over.
For reflow soldering, the solder paste itself acts as a temporary adhesive. A well-printed paste deposit holds 0402 and 0603 discrete components in place during the entire heating cycle. If your paste is not holding the parts, the problem is usually stencil aperture design or print pressure — not the paste itself.
Thicker stencils put down more paste, which means more holding force. For discrete semiconductors, a stencil thickness of 100 to 125 micrometers works well for most packages. Going thicker increases the risk of bridging on fine-pitch parts, so find the balance that holds the component without creating shorts.
Wave soldering puts discrete components through a lot of mechanical stress. The conveyor vibrates, the flux sprays, the preheat zone bakes, and then the wave hits. If the part is not locked down before it enters the machine, it will not survive.
Yes, adhesive. For through-hole discrete semiconductors on wave solder lines, a small dot of epoxy or UV-curable adhesive on the component body before insertion keeps the part glued to the board during the entire process. The adhesive cures during preheat or under a UV lamp before the board reaches the wave.
This is standard practice on high-reliability boards — automotive, aerospace, medical. The adhesive does not interfere with the solder joint because it sits on the top side of the board, away from the pads. It simply prevents the part from lifting or shifting when the wave hits.
Clip the component leads to 1 to 2 millimeters above the pad before the board goes into the wave. Excess lead acts like a spring — it flexes under the wave force and can lift the component off the pad. Short leads stay rigid, and rigid leads stay soldered.
Cut the leads flush after wave soldering, not before. Trimming before the wave gives you a clean, short lead that resists mechanical stress. Trimming after means you are cutting long leads that have already flexed and possibly damaged the joint.
Bending leads with your fingers instead of pliers. Finger pressure is uneven and can crack the lead frame near the body. Always use needle nose pliers with a smooth jaw surface.
Bending leads in the same direction on both sides. This makes the part sit crooked on the board. Bend one lead clockwise and the other counterclockwise so the part sits flat and both leads enter the holes at the same angle.
Skipping the staking step on tall components. A bent lead on a TO-220 transistor is not enough. The part is too heavy and the leads too stiff. Always flatten at least one lead tip against the pad.
Using too much solder to "help" the part stick. Excess solder creates bridges, cold joints, and inspection failures. A proper mechanical fix eliminates the need for solder to do double duty.
Reheating a joint more than twice. Every reheat cycle degrades the semiconductor junction. If the joint is cold, fix the root cause — iron temperature, pad cleanliness, flux coverage — not the joint itself.
After every soldering operation, check the mechanical stability of the component. Grab the part gently with tweezers and try to lift it. It should not move. If it wiggles, the lead is not properly locked and you need to rework the joint.
For wave soldered boards, inspect the first board of every lot under magnification. Look for lifted leads, misaligned parts, and cold joints. A component that is slightly off-center might still solder, but it will fail under vibration or thermal cycling.
Pull a cross-section on at least one joint per board type. Cut through a diode joint, a transistor joint, and a power regulator joint. Verify that the fillet climbs at least 75 percent of the lead on the component side. If the fillet is thin or does not wet the lead, the lead was not properly fixed before soldering, and the joint will fail under stress.
Check the intermetallic layer thickness under the microscope. A thin, uniform layer means good wetting and proper heat control. A thick, spiky layer means the component absorbed too much heat during soldering, which usually happens when the lead was not acting as a heat sink because it was not properly clamped or staked.
Contact: Joanna
Phone: Info@addcomponents.hk
Tel: 852 5334 3091
Email: info@addcomponents.hk
Add: FLAT/RM C -13/F HARVARD ,COMMERCIAL BUILDING 105-111 THOMSON ROAD,WAN CHAI HK