2012-03-11 Quantsuff’s Simple Solar Flasher

QS has added two nice little flashers to his web page.  The first, a Simple JT Flasher, uses a regular 1.5V Alkaline cell.  The second uses a rechargeable cell, and recharges it with a solar cell.

I think I would not consider using the solar cell and rechargeable battery on a flasher.   Why?

Well, my thoughts are that the average current drain of the flasher is so ridiculously low – just microamps – that for all practical purposes, the solar cell is just replenishing the rechargeable battery’s normal loss of capacity, which is about 1 percent per day.  My thought is to not bother to use a rechargeable battery on a flasher when a fresh AA cell will last for 2 years when it’s flashing 24/7, and probably double that long when it’s turned off during daylight hours.

I think it’s a waste of a solar cell.  And I’ve found that when I put a solar cell on a project, all of a sudden it’s now dependent on being located in a location where there is bright sunlight during the day.  Or else I have to put a long pair of wires on the cell so that it can be remotely located at the bright light location, and the rest of the circuit located somewhere more convenient, or maybe more appropriate, in a place less inconvenient.  The whole point of a flasher is to attract attention and that doesn’t happen when the ambient light is so high that the flashes can’t be seen.

The other Simple JT Flasher, it would be better I think to turn it off during the daylight hours with a LED and transistor.  My Blue Blinky (its schematic is here but it doesn’t show the LED and transistor that shut it off) has this on the PC boards and it works okay.

Here’s how it would connect in QS’s schematic.  The emitter would be connected to negative.  The LED cathode would be connected to negative and the anode to the base.  The collector would be connected to the junction of the R1 and C1, which takes only a few microamps collector current to pull it to negative.  When the light hits the LED, now used as a sensor, it turns on the transistor and the blinking stops dead.

My Blue Blinky has something like fifty to a hundred milliamps peak battery current draw when it blinks, but the pulse is only on for milliseconds and then off for the rest of the second.  I don’t know what the average current of the Simple JT Flasher is, but it may be less than my Blue Blinky.  It may run for a year or more with a fresh AA cell.  Since the circuit is so much cheaper, it seems logical to seal it including the AA cell in a waterproof container and use it and several others outdoors for holiday decorations, then let them stay until they run down to nothing, then dispose of them.

Update Mar 15, 2012 – Referring to the schematic of QS’s Simple JT Flasher, I built one very similar to this schematic.  I used a coil made from a Fair-Rite 2673002402 toroid with 11 turns bifilar wound, one wire being a 24 AWG solid insulated telephone wire, the other being a 30 AWG enameled wire.  I used a BC338 transistor and white LED.  The resistor between the coil’s feedback winding and transistor’s base was 3.3k just like the schematic.  The resistor between the +1.5V and the coil’s feedback winding was 470k, with a 4.7 uF capacitor in parallel (see schematic).

Okay, my gut feeling was that I’ve built some similar circuits before, and they didn’t flash.  Now, when I first applied power, The LED briefly lit up brightly, then went very dim, and stayed that way.  My gut feeling was confirmed. This circuit is very similar to QS’s, but it doesn’t blink by itself.

I noticed that when I moved the circuit around the very dim LED made the strobe effect where the LED was not a blur, it was a series of flashes of low frequency.  I thought that was a good reason that I should measure the frequency.   However when I clipped the DMM leads onto the emitter and collector leads, the LED started blinking.   Fast, but brightly, though.

I took the leads off and it stopped blinking.  Hmmm.  I took a 330 pF ceramic capacitor and clipped one lead to negative, and touched the other lead to the collector, and it started to blink, but quickly.  I changed to a 2.2 nF (2200 pF) capacitor, and it blinked slowly.  I tried others but I finally settled on 560 pF, which gave a blink of about 1 sec.  Without the capacitor across the LED and transistor, the blinking stops and the LED just glows dimly.

Recapping what I have done, I replicated QS’s circuit, with one added component, a 560 pF capacitor in parallel with the LED and transistor to maintain the flashing.

Back to experimenting…