Dragging a 0.5 mm pitch QFP and clearing bridges
At 0.5 mm pitch the gap between leads is smaller than the diameter of most solder wire, so the alloy has to come from the tip's reservoir rather than from a hand-fed wire. Bridging is not a mistake to avoid - it is a stage in the process to be cleared afterwards.
Procedure
- 1
Align all four sides under 20x with grazing side light and tack opposite corners with 0.2-second touches.
- 2
Flood one side with 0.3 mm SAC305 at low feed rate, deliberately bridging all leads on that side.
- 3
Drag a 1.2 mm hoof along the row at 350 degrees C at 3 mm per second, keeping the concave face loaded.
- 4
Lay 0.8 mm braid perpendicular to the leads, add RMA flux, and sweep with the edge of the tip for under two seconds.
- 5
Inspect all four sides at 20x with grazing light; re-flux before any further passes and cap each pass at two seconds.
Workshop notes
Alignment before any heat
At 0.5 mm pitch a 50 micron misalignment is visible at 20x and a 100 micron misalignment will bridge. Align by two opposite corners under the microscope, tack each with a 0.2-second touch, and check all four sides with light coming from a low angle - grazing light shows lead-to-pad offset far better than top-down illumination. Do not commit to the drag until all four sides look correct.
Flood the side, then drag with the reservoir
Flood one side with SAC305 using a 0.3 mm wire at low feed rate so the leads are all bridged together. Then drag a 1.2 mm hoof along the row at 350 degrees C; the concave face holds the excess and the surface tension of the liquid alloy pulls it back into the tip. Move at 3 mm per second - slower than for SOIC, because the leads are finer and cool faster.
Bridge removal without lifting pads
Wick with 0.8 mm braid laid perpendicular to the leads, and use the edge of the tip rather than the flat, so contact area is small and the dwell is short. If a bridge survives two passes, the cause is usually insufficient flux, not insufficient heat. Re-flux and try again rather than pressing harder; 0.5 mm pitch pads lift at around 3 seconds of contact.
Practice drill
Solder one 48-pin QFP and count bridges before wicking and after. A good result is under eight bridges across all four sides before wicking, and zero after two passes. Then re-do the same package with the setpoint 20 degrees lower and compare the bridge count.
Failure modes
| Symptom | Cause | Treatment |
|---|---|---|
| Bridges reappear on the same two leads every pass | Flux burned away at that point, usually where the drag speed slowed | Stop applying pressure, re-flux the full side, and approach that segment last with a fresh tip load |
| Pad lifts during bridge removal | Flat of the tip used instead of the edge, and braid held for more than three seconds | Rotate to the tip edge, change braid every two passes, and cap the pass at two seconds |
| Leads bent sideways during alignment | Tweezers used to nudge the body while the leads were already touching the pads | Straighten the leads with a 0.15 mm blade under the microscope before placement, and move the body only while all four sides are clear of the pads |
Supplies
- 0.3 mm SAC305 solder wire
- ROL1 gel flux in a fine syringe
- 0.8 mm desoldering braid
- Isopropyl alcohol and a soft brush for final cleaning
Tools
- Station at 350 degrees C
- 1.2 mm hoof tip and a spare 0.4 mm conical for touch-ups
- 20x stereo microscope with side illumination
- Fine tweezers with 0.2 mm tips
Safety
0.5 mm pitch work is done under a microscope, which puts your face close to the fume plume. Route the extraction intake to the side of the board, at the same height as the joint, and take a break every 30 minutes of continuous microscope work.
Neighbouring cards
- Drag soldering a SOIC-8 with gel flux 340 °C / 1.5 s
- Two-iron removal of a 0805 chip 330 °C / 1.5 s
- Hand-placing a 0402 passive 320 °C / 1.0 s
- Fixing tombstoned 0603 and 0805 parts 310 °C / 1.0 s