3
\$\begingroup\$

I was visually inspecting a board that had functional problems, and I have noticed a strange pattern in the pads surrounding the IC.

solder failure

Upon electrical inspection, there are two pads that are shorted out: the second and third one on the horizontal part right to left.

This is a prototype board and there was lots of debugging and probing the board to solve some problems.

Does it seem to be consequence of mechanical damage, such as external probing with pointy tip, or a defect from the reflow process?

Could just a touch up with the iron reshape the soldering joint? I am afraid the IC pins are bent with the surrounding solder, which would make the repair difficult.

\$\endgroup\$
4
  • 1
    \$\begingroup\$ Certainly looks plausible. Flux and a drag outwards with a large iron tip or a hot air pencil should fix it - doesn't really look like pins are bent, just the very soft solder alloy has flowed away from probe pressure. \$\endgroup\$ Jun 18, 2018 at 23:20
  • 1
    \$\begingroup\$ Bent pins? from the picture above I would have guessed that the chip is actually a QFN. \$\endgroup\$
    – MAB
    Jun 18, 2018 at 23:38
  • \$\begingroup\$ Indeed @MAB is right. That's not a QFP with fine leads, but rather a QFN package that does not actually have "pins" just areas of metallization on the side and bottom of the package. That said, the problem can happen with a QFP too, and unless someone is really heavy handed the issue is generally limited to the solder, not the pins. Actual bent QFP pins typically result from trying to remove chips without using hot air or IR. \$\endgroup\$ Jun 18, 2018 at 23:48
  • \$\begingroup\$ Your photo is not clear enough to tell it for sure. To me it seems just bad solder joints, maybe due to someone soldering test wires directly on the pins (classic 30 AWG "bodge wire" perhaps, I guess) and then removing them without reworking the joints accurately. \$\endgroup\$ Jun 19, 2018 at 4:06

3 Answers 3

2
\$\begingroup\$

I suspect your idea is correct.

First, it shouldn't be too hard to fix the damage - take an Xacto knife and bend the solder "lips" upward, then touch up with a very fine iron.

Second, invest in some needle-tip probes. These will make good contact with less force, and will also leave smaller divots.

Finally, whoever did the debugging should be shown your pictures, or even better, shown the board. If you don't know which person did it, show them all.

\$\endgroup\$
2
  • \$\begingroup\$ A typical #11 xacto blade is just about exactly as thick as the center-to-center pin pitch of a QFN, so even the bevel may not make it fit in between - picks that actually do are quite tiny. But if the blade is on hand doesn't hurt to try. And I would very much agree that this is readily repairable. \$\endgroup\$ Jun 18, 2018 at 23:32
  • \$\begingroup\$ @ChrisStratton - Yeah, the idea is to use the bevel of the blade, putting pressure and rotating from side to side by about 20 or 30 degrees. With luck the solder lips will be pushed aside. With no upward pressure being applied, the pad probably won't lift. Takes a delicate touch, though. \$\endgroup\$ Jun 19, 2018 at 0:21
1
\$\begingroup\$

You need two things:

Thing #1) The first one is flux
Thing #2) The second one is a soldering iron. (it need not be a fine tip, but it does need to make contact with the pads and clear the package)

You put the flux on (not too much but enough to cover the pads)
Then you clean the iron on a brass or water sponge and clear the solder off the iron. Wipe the iron tip by the pads , and they will magically fix themselves.

Oh, actually you need three things, then you need to remove the flux with flux remover.

I only use solder braid to remove solder if there is a inordinate amount. For small bridges only flux and an iron is needed. It really is amazing what flux will do.

Here is a video for illustration.

\$\endgroup\$
0
\$\begingroup\$

I would use copper litze. First add flux then put the thinnest lize you can find on it. Short press with solder iron.

Sometimes you have to repeat it twice,

I find that solves the problem 99% of the time.

I would not use knifes until the solution above fails.

\$\endgroup\$
1
  • 2
    \$\begingroup\$ Yes, though given that this is an issue of displaced solder rather than extra solder, the copper braid may not actually be needed. And to those who have not tried it, the appropriate size would be extremely tiny. Use braid of too large a size, and it's likely to become frozen to 4-5 pins at once and require getting a lot of heat in there to remove, possibly bringing enough solder with it that if the board is even slightly imperfect the connections left behind are broken and need to have solder added back in. \$\endgroup\$ Jun 18, 2018 at 23:44

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

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