In my previous post, I gave you my thoughts on my desire for a portable power supply. Since then I have drawn up the circuit with Altium, and laid out the board. My normal procedure from there is to leave the project alone for several weeks so that I have forgotten the fine detail. You may think this is strange, however when you have no one else to Check your work for errors then by letting the project lie for a few weeks then when you return to it, you have to re acquaint yourself with the details and I often then notice things I have overlooked initially. Having checked the board I sent the files to China to be manufactured.
Once I received the blank boards from my supplier, I was anxious to assemble it and get going.
A while back I purchased a second hand infrared reflow oven and was looking forward to trying it out.
This was particularly good as a couple of the chips were very small and were of the style that have no legs protruding. Partly because it was a prototype and also a short run I did not order a solder paste shim, so using a syringe with solder paste I put paste on the PCB using the easy method of just a string of paste on each IC. It is quick and easy, but usually results in either shorts or unsoldered pins on some ICs. Having the paste on the board I then used tweesers to place the components. It is actually easier to use a vacuum pickup but I haven’t set that up yet. I wasn’t very careful placing the components as I knew that during the reflow process the surface tension of the solder will pull the components into place unless they are way out of line.
Since I had set up the oven heat profile set up a few days prior, I put the board in the oven and started the cycle. Fifteen minutes later the board was back at ambient temperature. Unfortunately the board wasn’t soldered properly, the board had not been made hot enough to reflow the solder properly. I decided to modify the profile then ran the board through again. This time the solder reflowed properly, but as expected I had about five shorts and a couple of dubious joints. It took a short time to fix the problems. Great now I could apply power, I started by feeding power in through the battery connection. As always when testing a new design, I start with the power supply set to zero then slowly raise the voltage while keeping a close eye on the current. This time I barely raised the voltage when the current was zooming up. The finger test told me the first regulator was dissipating the power, so started checking the circuit for unseen shorts, but there were none. I re checked the circuit for any errors on my part, no none of those. After a few more minutes I realised that I had the power supply connected backwards. Fixing that, and now I could raise the voltage to four volts with only a small expected current flow, and the power LED glowing brightly.
Testing the circuit with a multimeter showed that everything was as expected, without software yet in the microprocessor. The only part untested was the USB power input. I connected power there but alas no joy. It took me a little while checking the circuit against the board and against the data sheet finally I worked out that I had forgotten to put an inductor on the output of the USB power supply regulator. I will put that on the board over the weekend, and complete the DC testing.
Once the DC tests are complete I will start putting the software together in small chunks and test each section individually.
My next post will be after I have done some software testing