Friday, 12 August 2011

Hot air soldering

Jason brought along his hot-air rework tool to BuildBrighton this evening and, having worked with SMT stuff recently, we couldn't wait to give it a go.

He's the rework tool. It has a digital heat selection and a fan controller. The fan is actually housed inside the handle of the tool and controls the rate of flow of hot air onto your board(s).



Using our flux-and-solder-paste-everywhere approach, we blobbed some around on the pads of a SOIC package chip and let the hot gun do it's work.


The end result was quite impressive. It takes about a minute of heating before the solder paste goes noticeably dull then suddenly liquifies and heads off for the nearest bit of tinned copper trace it can find! The end result is nice shiny pads and no tricky pin bridging. Very impressive (and easy to do).

On these pads, we deliberately went crazy with the solder paste to see how the heat gun would deal with it. The result was the same, although there is noticeably more solder on these pads, they are still well soldered and no sign of bridging anywhere



Ribbon cable wasn't quite as successful.
Predictably the plastic casing melted and deformed. But this also caused the cable to flex and bend, so not all the cores were in contact with the board. Where they were, the wires soldered quite well. But where the ribbon cable lifted, the solder paste stuck to the undersides in blobs and bridged the fine 0.025" pitch traces and generally made a mess.



Further heating with the rework tool, to try to reflow this solder to remove the bridges caused the plastic to blister and burn in a series of small black bubbles. Prolonged use of heat caused the board to scorch (see above) and start to blister in places too.

In short, it looks like we're stuck with the soldering iron for connecting ribbon cables. But the hot air tool looks very promising for sticking down fiddly little SOT-23 sides transistors, and to get a lovely finish on the SMT chips.

We expected the hot air to blow the little tiny components all over the board, but the air flow isn't so strong as to make them move about too much. The only time we noticed any movement was as the solder liquified, the components "floated" on the hot solder. But sometimes this actually helped, as the surface tension automatically pulled the components in to line on the pads.

Now if only there was some easy way of converting a CNC machine into a pick-and-place machine, to simplify the actual board assembly.....