Light guide clock Part 12, originally uploaded by Black Heart Industries.
Another bad photo. The LEDs are glowing white. I reworked the circuit to reduce the current to about 65 mA. Basically the same as the schematic (courtesy of Peter Wendt). The only difference is I used a LM317LZ (transistor package) and instead of a resistor and pot, I just used two 10 Ohm resistors. If I use just one 10 Ohm resistor, the LEDs are even whiter and the current is 125 mA. That’s probably what I’ll use. There will be six LEDs, four for the numbers and two for the colon.
Update: From Peter Wendt: “If you use 4 diodes constantly powered the 125mA distribute to >30mA per diode, which also exceeds the maximum allowed forward current of 20mA. The 65mA given on your blog is quite nice, could be expanded to 80 in best case. Even if you use 2 additional LED (as blinking colons / separators) I would stay below or close to 80mA in either case, because these are *not* constantly on. When they are off the remaining diodes need to handle the current as well. If you now get 65mA you have 1.25V (the regulator reference voltage) / 0.065A = 19.23 Ohms in series with the regulator towards load. The minor difference to 20 Ohm is caused by the bypass current through the regulator. If you want 80mA you divide the 1.25V / 0.08A and get 15.63 Ohm. Using a 10 Ohm and 2 x 10 Ohm parallel will bring you about there.”
He’s right again. I just pulled out an electronics text book. I want to learn the math so it’s second nature. After all, I’m good at the math, use it every day.
