2
\$\begingroup\$

The solder fillets around the base of the header pins are almost flat - very different from the kind of fillets you'd expect to around a 0.1" header. These kind of fillets are very common on ST's evaluation boards. By the consistency of the soldering, it look like it has been done on a machine. I am interested in knowing the soldering technique involved in this.

enter image description here

Kind of fillet you'd generally expect to see: enter image description here

\$\endgroup\$
3
\$\begingroup\$

I suspect it has to do with the pad geometry. The pad rings look very narrow with a fairly large hole (for the pin size) which prevents the normal solder fillet from forming. Notice there is more of a fillet on the larger square pad for pin 1. There are robots that do the soldering ensuring a measured amount of solder on each joint.

It does not look like paste in pad (using the stencil process to put paste on the through hole pads), as you would end up with paste on the pins if they were inserted after paste was applied.

It was probably not done on a selective solder machine (like a one pin wave soldering fountain) because there is no access on the bottom side and the pins on the top would end up with solder on them if selective soldering was done upside down.

\$\endgroup\$
5
  • \$\begingroup\$ I think that it was done by either a universal soldering robot (like Dean mentions) or some sort of custom soldering jig that solders all the pins at once. When you look at those boards in person, you can see very consistent small puddles of flux that were left by soldering tip. \$\endgroup\$
    – desqa
    Nov 6 '17 at 6:26
  • \$\begingroup\$ A robotic system using solder paste (no clean flux) metered by a dispensing system and a hot air system could also produce consistent results like these. Doing this with a board heater would also make the process easier and faster. \$\endgroup\$ Nov 6 '17 at 9:20
  • \$\begingroup\$ Yes, I think that would be possible but I don't know how practical that would be. You would need to use more expensive headers that can withstand higher temperature (going through reflow oven/hotair). \$\endgroup\$
    – desqa
    Nov 6 '17 at 9:36
  • \$\begingroup\$ If the hot air was applied from the top you could use standard headers. Using a board preheater would reduce the solder cycle time reducing the peak temperature at the plastic spacers. I'm pretty sure the headers on the Nucleo boards are not rated for reflow. \$\endgroup\$ Nov 6 '17 at 9:54
  • \$\begingroup\$ Yes, I think you are right on that. But, as I mentioned earlier, I am not sure how viable would that be for mass production. \$\endgroup\$
    – desqa
    Nov 6 '17 at 10:05
2
\$\begingroup\$

My guess would be that they used paste-in-hole soldering technology (aka intrusive reflow soldering) to solder the headers, but did not do the calculations properly to determine how much solder was needed to provide a fillet (or maybe they intended for there to not be a fillet at all). Here are a couple of links describing this process:

https://blog.samtec.com/post/guidelines-paste-hole-reflow-processing/ http://suddendocs.samtec.com/processing/through-hole-printing.pdf

By using a stencil that is too thin, less solder would be applied to the holes and it would give you the result shown in your image.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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