Close

#9 - Setbacks, or how Pb-Free solder paste ruined my protos

A project log for MAD - Modular Audio Devices

A set of Eurorack modules to control synths, make custom MIDI-compatible interfaces and USB audio, powered by Teensy 3.6

michele-perlaMichele Perla 09/29/2017 at 15:460 Comments

Quick update: I ruined my first batch of protos.

I received the PCBs, real beauties; I received all the components, and I received TWO kinds of solder pastes: a ChipQuik low melting point one, and a CIF lead-free because enviromental concerns yaddah yaddah.

I made the first CORE module and a set of IN MINI connectors with the ChipQuik paste just fine, here they are:


The paste is ok, but it does not stick very well to the pads and it keeps oozing out of the syringe even after you stop pushing the plunger. 

The first CORE proto is fine (except for a spurious copper polygon on the bottom layer that accidentally shorted the power pins ... OOPS :D luckily I only had to scrape some copper away and that's it), the jacks are pretty straightforward.

THEN: I decided to use the Pb-free paste for the rest of the protos, that is another CORE module plus the PSU, IN and OUT modules. The paste is a beauty to dispens, it sticks perfectly on the pads and does not ooze, allowing a precision job to be done really easily.

The downside? IT NEVER MELTS. Take a look:

(oh, in the last picture you can see that the package for the TPSA6120A2 is smaller than it should be but luckily the 10 useful pins could match up with the fooprint)

I literally put the board through 4 cycles of reflowing in the oven.

I first used a standard profile, then a longer and higher temperature one twice, then a custom made with even longer and higher temperatures... nothing. The paste is half-melted half-not, the board is literally baked. In an attempt to recover this situtation I turned on the manual setting on my oven to try to keep the temperature even higher for a longer period... the only result I got is that a capacitor exploded, the other 3 remaining electrolytics are swollen and ready to explode at any time, and most probably all the components on the board are useless by now. 

The thing is that I could not even get this solder paste to melt with my iron nor with my hot air gun. I seriously put the tip of the iron to more than 300 celsius but the paste JUST. WON'T. MELT.

The only time I got one component properly soldered was when I literally put the hot air gun at 320 celsius with the nozzle at 2 mm distance from the component. I know I can't try this technique on all the components because if they're not broken by now (I highly doubt it) they'll surely be after this treatment.

So... yeah, a lot of cash just went down the drain. This sets me back a lot in terms of money and time, because I should have spent this week testing the protos but I ended up sad with no money to invest right now and with no protos to test :( 

I'm going to order the components to make another panel in the next days, in the meanwhile I'll solve some software issues that does not let me use the CORE module properly (there's something in the AudioOutputI2S object of the Teensy Audio Library that crashes I2C communication with the AK4558).

More to come.

Michele

Discussions