Electrical – Hot air soldering without destroying ICs

failurehot-airsolderingthermalzener

I have designed a circuit board that has about 20 parts on it, including a surface mount LMV431 Adjustable Zener Shunt Regulator and MMBT2907 PNP transistor.

It appears that every time I solder up a board, the Zener and/or PNP transistor get cooked and do not function properly afterwards. Specifically, the Zener never goes to a high-Z state as it is supposed to (keeps conducting regardless of conditions) and the PNP transistor doesn't shut off (keeps allowing current flow even with zero base current).

Here is my soldering process:

  • Heat up bottom of PCB with hot air rework station set to 150C on maximum air setting for about 2 minutes holding hot air tip at 30° relative to PCB. In last 20 seconds ramp temp up to 280°C
  • Reduce air flow to ~4/8 setting, keeping temp set at 280°C, and immediately begin heating up entire top surface of the PCB, moving the tip constantly at a speed of ~ 8cm/s and tip distance of approx. 1.25”. Do this for 20-30 seconds
  • Reduce tip distance from board to ~0.5” and begin focusing heat in 2 cm diameter swirls on the section of the board nearest the USB connector landing area. Focus heat on capacitors and resistors in this area, and leads of FET. Make sure to revert back to the 8cm/s area heating from the last step on the part of the board that is not being reflowed in order to make sure its temp stays up. Probably reverting back to this every 10 seconds for 1 or 2 seconds is advisable.
  • Blow on 45° angle all around the FET to get the solder underneath to reflow
  • Move reflow target area to the Zener/PNP side of PCB, taking care to keep tip moving fairly fast. Try to avoid aiming tip directly at Zener or PNP, instead trying to blow just to the side of them to get their leads to reflow

According to the Zener datasheet, the maximum temperature during soldering is 260 C:

enter image description here

At this point I'm worried because I can't seem to assemble a working prototype in order to validate the design (need to be 100% sure that the inconsistent go/nogo between the 5 boards that I've assembled is indeed due to thermal stress during reflow and not an inherent flaw in the circuit).

Should I give up trying to do this with a hot air rework station and buy a reflow oven?

Best Answer

Using hand soldering worked for me. I'm guessing that I need to play around with my hot air soldering station a bit more in order to get a feel for what the actual temperature it outputs is compared to what the digital display shows. I'm guessing that the output temp is higher than what the meter says