Prototyping With Very, Very Small ICs

Gone are the days when all the cool chips are able to be thrown into a breadboard very easily. [starlino] was working with a circuit that uses an accelerometer, but unfortunately these chips come in hard to solder LGA-16 packages. [starlino] figured out a way to prototype with these packages that doesn’t require a custom breakout board or spending any time watching a reflow oven.

[starlino]’s LGA-16 adapter board began with a piece of perf board drilled out to form a space that perfectly fits his accelerometer. A piece of tape is placed over the pads of the chip and perf board, and the gap between the chip and board is filled in with a two-part plumbers putty.

Once the putty has cured, the leads on the acclerometer are connected to the pads on the board with a silver conductive pen. After putting a few header pins in the corners of the board, [starlino] soldered the pads to the pins and had a permanent breakout board for a very small accelerometer.

It’s not by any means a pretty build, but after [starlino] sealed the entire build in liquid electrical tape and installed it in a DIP socket, he had a completely functional accelerometer in an easy to prototype package. Not bad for a breakout board that can be built from stuff just lying around a workbench.

20 thoughts on “Prototyping With Very, Very Small ICs

  1. it’s quite an old news…. where did I saw that a few month/year ago? was it on HaD or by ladyada, seedstudio or an other site like that? I can’t remember…

    nice trick through…

    1. Yeah I thought that was clever. Maybe as a minor adjustment, etch a pcb with traces that lead right up the edge of the hole. Then you only have to solder-bridge to the pins themselves. I guess that depends on getting the hole *just* right, though.

  2. Neat technique. I don’t recommend leaving pins (like reset/sleep) unconnected though, unless the datasheet explicitly says it’s ok; or shows an internal weak pull up/down resistor connected to those pins. Stuff like that will bite you when you least expect it, and can be incredibly hard to troubleshoot. Plus the encapsulation will make it difficult to later go back and fix.

  3. This is interesting.

    An alternative [and reusable] technique is to etch a simple breakout board, or buy one. Then use conductive adhesive tape to affix the component. No soldering, except to attach breakout pins. The component can be re-used, and so can the breakout board. You can usually find it in small quantities on ebay… Using it requires no special tools [hot air, etc].

    The conductive tape can get a bit pricey, but it’s still cheaper to buy the tape, and be able to put the prototyped component into your final circuit, instead of being left with a component forever potted into the breakout. Just pull it off the tape, and it’s good as new…unless you’ve released the magic smoke :D

    1. Hmm, this talk about removable SMDs gives me an idea.

      There’s a product called Stabilant 22, used for enhancing electrical contacts. It claims to be a liquid polymer semiconductor, that turns on only in the presence of an electric field. The turn-on characteristic is a function of voltage/distance, so it’s supposedly excellent at bridging tiny gaps in an intended but imperfect connection, yet doesn’t cause any unintended connections (like for adjacent pins) because the distance is too great.

      So I wonder if you could spread a bit of this on PCB pads, drop an SMD on it, and get a reliable temporary connection. Would be great for programming and testing SMD ICs.

      I guess I’ll add this to my ever-lengthening “to do” list.

      1. If I had to guess, it probably won’t work as you’re hoping…Let’s say you have two pads in close proximity – GND and VCC. Let’s then say you spread the polymer across the pads. When you power-up, an electrical field will most definitely be present, and will pretty likely create a short. But it’s worth trying if you’ve got some…

        The conductive adhesive only conducts on the Z-Axis. It’s widely used to make pressure-fit contacts for FPC-to-PCB connections [think display cables].

        Aside from prototyping uses, I also use it with success for programming SMD chips prior to soldering in the final circuit. Makes it really easy.

      2. Did not realize the adhesive conducts only in the Z axis. That makes it far more intriguing!

        There’s no chance the Stabilant will bridge adjacent pins/traces, even if you slather it on finely spaced ones. The distance over which it can form a conductive channel is *very* small, probably only a few tens of microns. I do have some, and my previous description was qualified with “claims” and “supposedly” because of some mixed results. Some times it works like a miracle, others it makes the contact worse. I haven’t been able to figure out why, so I don’t know if it really works as claimed.

        I will try it, but not right now. I don’t have any empty SMD breakouts on hand, and my workbench is too cluttered to even think about making one at the moment.

  4. Interesting concept but it seems like an awful lot of work compared to etching a breakout PCB or ordering one.

    A few days ago I had to solder on a few 3mm QFN-16’s. I used the conical tip that came on the iron and .015″ solder.

    I don’t do this type of work often and I found it
    was a little tricky getting the part lined up on the pad. However, once I had one of the pins soldered it was a quick process of dragging the iron and solder along each side of the part.

    What I am saying that in the end it’s probably worth spending the time learning how to solder small SMT instead. My soldering skills are pretty basic so if I can do it, I am pretty sure anyone can.

  5. It would be a good idea to make a pcb with springcontacts, so you can easely insert it and dont have to mess with solder every time your testing.
    Ada fruit should have a look at such a converter..

  6. Might try this with my HV8823s.

    Also a handy use for the Z axis tape sold on Ebay is to repair broken LCD panels where one or more chips have come adrift from the glass.
    Usually if cleaned up with a gentle solvent the ITO is intact and if the chip is reattached that “duff” panel can be revived.

    @Sjors yes this would be a good idea.
    An even better idea would be one where there are little springs between the contacts and thumbscrews a la Jessops battery charger to set the width.
    That way a single adaptor can do lots of different SMD/QFP/LGA/etc parts without messing around too much.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.