There really isn't a way around it, there ain't no such thing as a free lunch (a favorite Heinlien quote). The good news is you can use D cells and get really good results. You might think about duplicating R3/Q1... to see if the 7555 can drive two of them, for 6 LEDs, that is. Don't use more than 3V, you don't want those LEDs conducting without a limiting resistor on a power supply.

For what it's worth, you are working on something other people will eventually want to use.

 

Bill....To tell you truth I'm surprised that it's 2009 and this has never been done. Seems like the death of the LM3909 ended the whole idea. Who was it that said..."If you build it they will come" OK maybe it's not baseball but it's just as important at least from an engineering point of view. So many great minds out there but it took finding this forum. I've spent the last couple of days delving into LED's. I'll be changing the LEDs from a 60% viewing angle to a 20% viewing angle. Also they have a mcd of 6500 verses 3000 and yet they retain a 2.2vF. This is in hopes of improving the brightness is daylight hours. I'll post again and let you know if I gained anything.

 

I'm back..... I just looked at Bill's post (#60) and that is not the one I built, I built #54. So now the question is which circuit is moe better, and yes I know it should be more not moe. Just a little late night humor. Bill I'll build up #60 if it's a better circuit. I know I was using a LM555 not a 7555 on my original build.

 

No problem. Wookie and I were brainstorming how to conserve battery power. I went back to the original design.

I still think we can improve this one, but I'll wait for another day and post it on this thread.