4
\$\begingroup\$

I have a QFN16 breadboard adapter and an ASXL335 chip, which I'd like to solder to it:

enter image description here

However, the accelerometer chip has a central pad, while board does not. This causes an annoying problem of soldering paste gathering under the pad and pushing the whole chip up (I have not expected this problem beforehand).

Is there any way to workaround this problem or is it futile? Will, for example, a small hole in the center help?

UPDATE: just to be clear. I do not need the pad as I am not concerned with thermal or physical strength issues (datasheet mentions the pad is there so accelerometer sits firmer to receive all those Gs it is supposed to measure). Also, I am using rework station fir this task.

UPDATE2: I've soldered it to the board after drilling a hole in the center and using less soldering paste. The device seems to have survived all these attempts. This time Z channel is not functioning, but X and Y do. Also, the picture was taken before adding GND-Vcc capacitor, as per datasheet.

enter image description here

\$\endgroup\$
  • 1
    \$\begingroup\$ possible duplicate of Soldering heatsink pad on bottom of IC \$\endgroup\$ – Leon Heller Sep 10 '13 at 18:10
  • 2
    \$\begingroup\$ My question is not a duplicate, as I am not concerned with thermal or physical strength issues, but how to solder the chip at all. Also, I am using hot air and the question you cite seems to be about soldering with iron. \$\endgroup\$ – Roman Susi Sep 10 '13 at 18:48
  • \$\begingroup\$ I don't follow how the solder paste getting onto the pad in the first place. Clean it all up, maybe dab of superglue on the pad to hold the chip in place. Don't use hot air, I don't think. You can solder these packages with liquid flux, a good tip and a microscope to check the joints. \$\endgroup\$ – dext0rb Sep 10 '13 at 19:00
  • \$\begingroup\$ Interesting. I can probably try this after finding a way to remove tin from the pad (solder wick perhaps?). I am skeptical about superglue though. Also, is liquid flux safe enough for this kind of chip when it gets under the chip? \$\endgroup\$ – Roman Susi Sep 10 '13 at 19:10
4
\$\begingroup\$

This should be a non-issue. Since you are not using the center pad, you should not tin it with solder, nor add solder paste below it. Just do the same thing you would do if it didn't exist.

Use a soldering iron and some de-solder wick to clean the center pad as best you can, and reflow it again.

If it really is a problem, you could try drilling out the center of that dip adaptor. This would give any pooled up solder paste somewhere to go instead of making the chip float.

\$\endgroup\$
  • \$\begingroup\$ Thanks for explaining. I can accept this answer even though I have not achieved the result yet (cleaned pad, reflown 2 times). It's double pity, that before using nice protoboard I have successfully soldered the same chip to my home made (combined chemically etched + cut with knife) QFN16 PCB, where the only problem was with Z axis pin. I thought the nice proto-board will give much better result. Accelerometer is dead by now, I suspect. \$\endgroup\$ – Roman Susi Sep 11 '13 at 17:33
  • \$\begingroup\$ Typo: The problematic pin was Y last time, not Z. And the accelerometer is still alive. \$\endgroup\$ – Roman Susi Sep 23 '13 at 17:26

Not the answer you're looking for? Browse other questions tagged or ask your own question.