The BAT pads are on the back of the XIAO ESP32-C3 which makes surface mount soldering to a mother board problematic. There is also a thermal pad, which I presume is the heat sink for the battery charger IC which may benefit from being soldered to the ground plane of the motherboard. Can anyone please suggest how these pads might be soldered to a mother board by hand? My thought was to add vias on the mother board so solder can be applied to the back of the mother board and wick through to the BAT pads and thermal pad.
I used the via method as you suspect… works perfectly.
Next round I may up the size a tad, works best if you pre tin the top side of the board and the batt pads on the Xiao.
HTH
GL:-)
Relevant to OP - I wonder, since those pads are so close to pins D2 and D3… Obviously this will fry the analog pins, but I wonder if there’s a way to just short the pads to those pins and use the pins superior mechanical strength and ease of soldering and rework as the battery terminals. Would that fry the adc capability of pins D0 and D1 also? What about the whole micro?
The pads are very frustrating to use. I’ve had three boards that were well soldered come loose after the whole project was assembled, they are just very tiny and weak. Very difficult to rework!
OK going to reply to my own post here because it occurred to me that one could (duh) cut the traces of those pins and just use the headers without fear!
Hi I would like to ask a follow up question on this topic. Anybody knows it the process of surface soldering the XIAO battery pads to a motherboard can be done automatically in a PCBA order?
Thanks
Angelo
1 Like
Singing to the Quire here, Talk about a PITA,
I ovaled the pads on my latest PCB and pre tined all four pads (2-chip) and the PCB two.
With nothing soldered I clamp(gently with long spring clamp. Flip it over and solder the holes in the pads (SLOWLY) a tad more heat exposure but it works well. Then I flip it and solder 4-5 top pins, not soldering un-used pins. Took a couple tries but that’s the tech I’m using to hand attach each to my PCB’s 5 so far NO issues.
this is the first one,
HTH
GL
NOTE: on the PCB order those holes are plated thru also.(called-out)
HI, can you share the EDA files you used to do this? I’m trying to do the same thing in KiCad (i can convert them in snapeda if you used something else).
Hi there,
I used the wiki eagle files uploaded them to EasyEDA. I don’t have anything else.
I uploaded the 3d models to Thingeverse.
HTH
GL PJ
I’ve been struggling with this very problem, and trying to follow the guidance on this thread.
Here’s a photo of the relevant part of my PCB:
This is the through hole that I’m using for the battery pads:
The center oval is 1.75mm x 0.75mm. The area right around it is plated, as is the hole.
The first thing I tried was using soldering paste and a hot plate. That sure got the ESP32 attached, but I think that just fried it - the CH LED didn’t light up when I plugged in a USB cable, and it didn’t talk to Arduino either.
When I tried tinning the area around the hole, the hole itself got filled with solder, and although I could get some of it to melt, it seemed to not melt all the way down to the ESP32’s pad which was tinned. (So, in practice, the ESP32 fell off when I removed the clamp.)
I also tried just putting a big blob of solder onto the ESP32’s pad and sending heat through top of the hole with the side of my chisel tip, but that didn’t seem to be enough to melt the solder on the pad - I set my Hakko to 770 degrees, and held the tip against the oval hole for 15 seconds, but no luck.
I just ordered an H18-C05 tip, since it looks like it would be small enough to actually fit through the hole and reach the solder below, but I’m pretty concerned about the tip of the tip getting hot enough to be useful here.
Anyone have other suggestions? Should I make my through holes bigger? Tagging @PJ_Glasso since you seem to have had success doing this.
Very grateful for any advice.
Hi there,
So it took me a while to get it correct, mainly Do not make the hole too big the pad can be oval but the hole can’t be too big or the solder will flow around the pad and not onto it, Yours LQQK a little big ,if that makes sense, I use flux only initially just to hold stuff in place. Generally I don’t tin anything but will put some solder paste on the BAT PADS very neatly. The holes are a “PLATED through Hole” variety, BTW the plating helps transfer the heat from the bottom to the top with the tip of the soldering iron IN the hole against the side wall and the solder being fed in from the side. I fill the hole up slowly and if it’s flowing down into the hole I stop let it receded , you can tell as soon as you remove the clamp if your good or not. it’s on there!
then I flip it over solder only the pins I need to use on the top. Plug in the battery and plug in the usb if it’s Green I’m Clean. 
HTH
GL PJ
I had pads arranged like so as to pick up Bat+ and EN pads on an ESP32S3 Sense;
The pad holes were drills of 1.2mm for BAT+ and 1.4mm for EN.
I used a real fine SMT tip with 0.25mm cored solder and it worked just fine;
Thinking of adding pads to pick up the MTCK, MTDO, MTDI and MTMS pins as well for extra IO.
Hi there,
Yes it is possible, I have one where i used them as I/O pins. Check the demo on here of it.
Works Great!
HTH
GL PJ
Which Demo is that ?
This one here,
HTH
GL PJ
Thanks for the link …
Thanks so much for the thoughts! The idea of using solder paste is intriguing, since it would take less heat to get it to actually melt, though I feel some concern about my own ability to line up the pads and the holes without smearing the solder paste in places that it doesn’t belong.
I had another idea for this, inspired by the castellated vias on the side of the ESP32 breakout board. I’d make one larger unplated oval hole that exposes the pads, then add a couple plated half-holes on the top and bottom that would sit on top of the battery pads. That should give me more room to maneuver the soldering iron without needing to line things up quite so carefully. It would look something like this:
The light yellow ovals represent where the BAT pads are; the bigger orange oval is the unplated opening; and the fat yellow half-circles are the places where I’d actually make the connection.
Any thoughts on whether this is a good or bad idea? Would this violate any common design principles? KiCad won’t let me actually set a 0.00mm diameter for the plating of the big oval, and when I set it to 0.1mm it says “Warning: pad hole will leave no copper”, but that’s what I actually want.
