Google+ Facebook Twitter Pin It Reddit

Jameco Challenges

SMT Soldering Nightmare

Submitted by an anonymous Jameco Customer

Hi Everyone,

smt I just thought I should share my SMT soldering experience. A particular chip I wanted to use, specifically a USB-enabled microcontroller, came only in an SMT package (44 pins at 0.8mm spacing). I wasn't confident, but I figured I could do this. My iron was old and clumsy, but with careful application of delicate quantities of solder, lots of flux, and a reel of wick I thought I could make do. I've done some passive component work in the past and they came out fine.

The soldering went okay. Messy, with all the flux and solder bridges between pins, but with a little more heat, the wick sucked the excess away. Careful metering showed all the connections were good. I realized that I needed sharp meter probes for this. The blunt ones are safer, but I needed a sharp one to be sure I'm on the pad at one end and on the pin at the other. While I won't try this at home, I've seen a professional use scalpel blades welded to his meter probes just to get them sharp enough.

Testing was the key, and I tested for both shorts between pins (most common) and open connections (they happen too). To fix a bridge, a quick dab of some more flux, and a bit of wick was an easy fix. I heated it up and watched the excess solder slurp away. I was careful not to over do it because it's easy enough to get an open connection instead. With every repair came new testing to make sure I wasn't going backward. Fighting back the nerves after testing, fixing and retesting, I plugged it in and turned it on! Smoke came out of the chip. Not good, but I guess it makes for a more interesting story.

This required more inspection of the board layout and more testing. I found a few minor concerns that were easy enough to fix. Sadly, however, my patient died on the table. The chip was no longer working and was pronounced dead due to a hard short across the power rails.

To remove the corpse I used a hot air gun, usually used for shrinking heat-shrink tubing. It runs much hotter than your typical hair-dryer, so don't use it for that, but for heating things, it works quite well. I clamped up the board and began directing heat over the chip. Sadly, my strategy was a bit south of perfection, for my plastic parts (sockets, etc) began to melt.

I stopped and regrouped, electing to make myself a heat shield out of aluminum foil. The foil wraps around the board and covers all the components, except where I've razored out a little square to clear the now-dead chip. This worked although the nearby plastic parts still melt a little, but they were still functional and now had a bit of extra personality.

While blowing heat, I lightly picked at the chip with a long sharp awl, and eventually (this took some time) it popped loose and fell off the board, off of the table and onto my lap. I was wearing shorts. It was hot and it hurt! Fortunately, the chip wasn't very large and only left a small souvenir of the experience.

The circuit board clamp, however, offered more resistance, and I managed to singe a couple of fingertips along the way. Now I cleaned off the pads with more flux and wick, I addressed the board layout concerns, and it was time to stick down a fresh new chip. It went well with a similar procedure. All the connections, including the edited ones on the PCB layout, tested out fine. I cross-checked it with the specification sheet to confirm I had done it correctly.

I pulled out my handy little microscope and also did a visual inspection this time of the solder joints. I was ready to try again. Power connected. Checked for polarity. All was well, until I turned it on.

Fizzle. Smoke. Pop.

Another board was inspected, edited, and provided, in case something was wrong with the first one. Another chip was installed, with an absolute minimum of supporting components.

Fizzle. Smoke. Pop.

Fortunately I had purchased quite a few of these chips. Much more checking ensued. On-line designs for similar boards were cross-checked with the chip spec sheet. The spec sheet was crossed with the schematic. The entire schematic was carefully metered out with a continuity checker, including possible shorts. All appeared well.

But the chips kept going up in smoke. A few more iterations of aluminum foil and heat-gun work, plus several more hours sticking down and testing more chips, resulted only in more burnt chips.

At this point you must be wondering if this story is going to have a happy ending. Time passed as did the waves of frustration before I was finally able to establish what may have been the problem. It wasn't a soldering problem, a circuit board layout issue, personal ESD discharge, the solder or the flux I used. Then I started to think that maybe it was the soldering iron itself that was the problem.

Never missing an opportunity to buy some new gear from my friends at Jameco, I bought a new iron with a properly grounded tip and a much finer tip at that. I stuck down yet another chip, and it worked without drawing excessive current and/or boiling itself. Yes... It seems at the moment it was my soldering iron that was destroying these things. Further experiments are still in progress.

A brief note to all my fellow hobbyists who might want to try SMT work. You should get an ESD-specified soldering iron for anything more complicated than surface-mount passive components!

Click here for related articles

Click here to order your "Getting Started SMD Experimenter Kit"

If you have an electronics story to share, send it to