We are making progress with the QuadroCopter. I had made the PCB yesterday, but tried to reflow the LGA chips (accelerometers, gyros, compass), but it didn’t work well. I tried to tin the LGA pads by doing a drag soldering technique, but it ended up pulling the pads off the board. Seemed a bit messy, and since the reflow seems like a one shot thing, I didn’t want to risk it, so I decided to remake the PCB. This time when remaking it, I made a few improvements to the manufacturing process. I cleaned up the transparency really well so no dust would effect the PCB traces. I shortened the exposure time and developer time slightly, and the board ended up coming out pretty clean cut. Also the etching process seemed to have come out 10 minutes faster than last time. I heated up the Ferric Chloride to a higher temperature, however no higher than 55 degrees Celsius.
Etching is still a test of your patience as it doesn’t happen right away, takes a good solid 15-20 minutes. But thats nothing compared to drilling the PCB holes. This time I bought more drill bits. I bought size #77 for the vias. Last time the vias were drilled too large and then the majority of the pad gets eaten up, then there’s nothing there to solder. You have to be extremely patient when drilling though. There’s somewhere around 377-400 holes to drill on the QuadroCopter mainboard, but each one has to be done slowly and carefully. If you rush it, the drill bit easily snaps. They’re really fragile. I broke about 10 at least. Even handling them breaks them easily. If it drops 1 inch, its guaranteed to snap. One technique I used when drilling it so just scrape the surface initially. Some dust from the fiberglass will come up, blow it away. Pull upwards on the drill bit to get the junk out and blow it away. I setup a fan next to the drill press to help out in the process.
I tried different techniques to do the reflow. First tried the oven in the kitchen. Preheated it to 385 degrees Fahrenheit. Using an infrared thermometer I checked the board temp and tried different techniques adjusting temperature. It doesn’t seem to work well and you can’t ramp up the temperature that quick. Then I tried using a regular toaster that you put two pieces of bread in, that doesn’t work well either. That seems to ramp up the temperature too quick. Even if the timing is right, I’m not sure if the heat distribution is even. You can time the slope of the temperature better by plugging on and unplugging the toaster. Using hot air seems to be ideal and a heater for the underside of the board to preheat it from the bottom.
The method I’m planning on using for the reflow will involve hot air and preheating the underside of the board with a hotplate. This has worked before when me and my friend reflowed an old nVidia card. There was one point at which nVidia switched from led solder to led-free solder and the temperature range was slightly different. So when the GPU heated up during use, after a while the connections on the BGA (ball grid array) became disconnected. A simple reflow often fixes the issue.
I would follow this temperature profile as it works and reduces thermal stress to the component as much as possible. Preheat the board from the bottom at 100 degrees Celsius. At the same time, preheat the top of the board and the components for 1 to 2 minutes. This is the soak time. Keep it at 150 degrees Celsius during the soak time. Then bring up the hot air temperature to 260 degrees Celsius. During this ramp up, the solder should melt at 217 degrees Celsius and flow at the peak 260 degrees Celcius. Once the solder appears to have flown properly and the components self-align, then drop the temperature. I would drop the temperature slowly rather than quickly to avoid thermal shock or stress. The reflow time between 217-260 degrees should last for 30-90 seconds.
Use flux when doing reflow for new PCBs. Flowing solder doesn’t work well without flux. Also there is a thing called solder paste which you can apply at room temperature which contains the solder and flux and flows really well almost like a typical adhesive. Here’s an example video showing reflow with solder paste.
And here’s a guide for solder reflow using hot air.
- Sony Cybershot W120 Teardown and LCD Replacement
- How to Hook up an N-Channel MOSFET
- LCD Screen No Power, No Signal








[...] issue was that the foil was not tight against the board, so I guess hot air from the reflow process got trapped between and made burn marks on the whole left side of the PCB. Adding a special heat [...]