I failed a bunch;so here is my pile of stupid (because that’s how I felt with each failure)
The reason most of these failed was because I refused to check my own damned work. The first set was due to a lack of vigilance. Some of the smaller connections crossed and caused an unusable board. (ref pic) Not that that’s the biggest problem, if there is a crossed line the connection can be forced apart using a sharp object (a knife of significant sharpness is recommended) or just separating the routes. The problem here was that the crossed lines where connected both on long stretches and multiple points. At this point I also discovered that the resistor footprint was way off.
Very annoyed, I attempted the second group. I created the custom footprint available on our downloads page. This time around, things went better; and after transferring the images (more times than I would like to recall), everything looked peachy and ready for… Er…Well, as you could guess, no luck on that either. The image was inverted. The problem was that due to printing space on paper, I had accounted for the 90 degree shift to turn the image but had not for the position of the board itself, which caused the image to be inverted. Logically, you may be asking “why you can’t just rotate the chip?” Unfortunately, that would leave it upside down (not something you can readily do with a SMD component) .It was right around this point that any sane person would realize that doing both the SMD and through-hole printing at the same time is insane, so I scaled back to doing one at a time.
Third attempt. It would seem that I forgot to put in the ground/negative connection on the 555 timer. Not ideal since it was already etched. *Insert several curse words here.* Went back re-routed, and back to printing and etching.
This time around, it turns out I left out a grounding connection on the 4017’s 15 pin. But a bit of MacGyvering later and it worked! It turns out that a small piece of insulated wire can be passed underneath the chip without looking suspicious.
That was the lot of it, a couple of attempts but a usable result and I am quite proud of this little achievement.
Last but not least, the SMD schematic (also available for download here). This one is more for the seasoned solderer. Usually SMD requires some specialized equipment, including but not limited to: a vacuum holder tool, tweezers , a hot air station with hot air soldering iron, solder wick and solder paste. Did I have any of this? Some, yes. To be honest I had a soldering iron, some solder wick and a lot of patience. SMD parts are really small and hard to maneuver. For that, I did however have a pair of small tweezers which helped a great deal. They filled in where the vacuum tool would be used, and they did their job well enough. In the end, working as steadily as I could, I still managed to displace two yellow LEDs, thus securing my favorite piece of advice: always have a couple of extras just in case.
The first step was to drill holes for the coin cell holder and considered my situation. This was hard mode: the first board went a good deal smoother on account of the fact that your components would sit in place.This was not the case with SMD components, which inexplicably had the ability to bounce randomly off even the most ridged surface. So, keeping one eye on the board and the other on the components, I started. First, things went smoothly: the solder will easily transfer to the copper if the solder iron is hot enough, making it relatively easy to “tin” all the connecting pads.
Putting a bit of solder on the tip of the iron and then running it over the pads in one sweeping motion works surprisingly well.The catch, as you would have guessed, is that some will make a solder bridge, for which I used solder wick (yes, not the most efficient use of materials, but it worked) to mop up any excess solder on pads. When pre-tinning pads on something like an IC 555 timer or the 4017 decade counter, I found that having as little solder on the pads as possible helps a great deal later on, ( any elevation discrepancy will be problematic when soldering). So those are the pads done. I then went ahead with placing and soldering the 555 and 4017. Or not...So, unlike their larger counterparts, surface mounted components don’t have a marking explicitly stating which way their polarity is pointing. For that, you are going to have to use some common sense (which I have remarkably little of, because I had to Google this). All components will have identifying text on them, which means that legible text will run from left to right, and in this situation the leftmost bottom pin will always be pin 1. So, yay for simple answers. After a brief consultation of the data sheet (pin 1 will connect to the 6th output on the 4017), I found that the chip needed to be rotated 180 degrees. The 555 timer had the dot, so it was easy to figure out its orientation. OK, all done with the logic parts.Skip ahead beyond the resistors and capacitors and that leaves the LEDs. It’s a personal preference of mine to do the LEDs last, because it allows one to test the circuit before committing all the components and then realizing something went irrevocably wrong.
So first LED on (green, if you need to know) and… Viola! A total fail. The LED turned on, meaning that there is power, but it doesn’t turn off, which means that there is no cycling outputs from the 4017. This has happened before, and by now I understand that the problem can be one of two things: power going in the wrong direction through the 4017, or the 555 timer isn’t giving a pulsed output. I suspected the latter, which was confirmed by a second LED. If two LEDs are connected and both turn on, then it’s most likely a polarity problem. If only one turns on, then it’s the 555 timer (or if you’re really unlucky, it can still be the 4017 but the LED is blown). Confirming that the LED is OK means a problem with the 555. This took me a while to figure out, eventually resulting in the removal of both the 4017 and 555 chips to be replaced. After this, everything worked perfectly, almost. There was one dead LED, which had no reason to be dead. After checking the pin for connection, there was no problem to be found. Replacing the LED also had no effect whatsoever. Eventually, using a magnifying glass I found a hairline break in the copper, which could easily be bridged by solder. After 3 hours of searching, de-soldering, re-soldering and some more searching, the problem was solved in less than 30 seconds.
The end result is very pretty though, and I still consider it well worth the effort.