I'm going to prototype using Wookie's driver to see if it works.

It's nothing special, really. But it should work.

It should have a 1ma current out the collector as set by the emitter resistor.

Depends upon the differential between Vcc/Vdd and GND. It's a simple "pick your poison" type thing.

Assuming Q1 really sucked and had a beta of 10 it would still only draw .1ma from the base, and 1ma should turn Q2 on pretty well.

If you have a current draw problem with the circuit I drew, the problem is with the components, not the circuit.

My current plan is to build the first 2 designs using TLC555s, then use my stock of 7555 for the rest of the experiments.

Are you running out of breadboards? This design isn't special enough to solder it down, unless it's going to be implemented somewhere.

I bought a 15 pack of PN2222 and another of 2N3904 (I believe that was the P/N) to stick with the Radio Shack parts line.

Why didn't you pick up a couple of Radio Shack PNP/NPN assortments, just to be sure you were working with "off the shelf" RS stuff?

The 1µF 250V cap was huge, so I'll try the tandilidium cap for the prototype too.

What's a "tandilidium" ?

 

It's nothing special, really. But it should work.

Thing I like about it is it has a lot of the advantages of a Darlington and doesn't have some of the disadvantages, like the large Vbe drop and Vce saturation drop. One of those tricks I'll try to hang on to.

What's a "tandilidium" ?

A common electrolytic cap with a southern pronuciation by a bad speeler.

Why didn't you pick up a couple of Radio Shack PNP/NPN assortments, just to be sure you were working with "off the shelf" RS stuff?

I don't have to guess what they are, my eyes aren't what they used to be unfortunately. I literally have thousands of transistors, thanks to Dan's Small Part and Kits assortment, but I like having real parts (from RS) so I can draw it to scale. I don't freehand it, I measure this stuff and put it into my PaintCAD. One of the downsides to doing something for a book I suppose.

Are you running out of breadboards? This design isn't special enough to solder it down, unless it's going to be implemented somewhere.

One of the experiments is duration. I plan on building it and then letting it run in the living room, this will give me a battery life measurement. Besides, I like blinky lights.

If you have a current draw problem with the circuit I drew, the problem is with the components, not the circuit.

Actually that was my point, even with substandard parts I think the design would work.

I've been doing a lot of tired designing, which explains the numbers of "DOH!"s lately. Heavy overtime, lousy sleep (3rd shift sleeping during the day doesn't agree with me), and some health issues all contribute. Truth to tell, I've been obsessing on this design, the kids think its funny when they hear me muttering "five time ten to the negative fourth times...". I've been using this site as a way to keep sane, if you can call it that. The good news is I seem to be able to draw in my sleep (really).

 

The first protoboard layout...

 

OK, the red LED flasher has been working for over 5 days, it is a little dimmer, but the batteries were a bit used and abused, so it is good.

It occurred to me that the CMOS 555 chippies will limit their own current, so my next project is something like this. While it isn't shown, a 100µF capacitor across the batteries is recommended, it will brighten up the LED when the batteries get a bit low.

I'll let ya know how it goes.

 

The datasheet for the ICM7555 Cmos timer shows that its typical output current into a 2V red LED with a 3V supply is only 4mA. When the supply drops to 2.45 then the current is only 2mA. Kinda dim.

Why not use the Cmos 555 to drive a transistor that will make the LED look bright?

 

Why not use the Cmos 555 to drive a transistor that will make the LED look bright?

Like I said, that circuit has been working for 5 (now 6) days.

Dim is relative. I'll measure the real current and post it, since the the LED drops 2.1 V it might even be less that 3ma. Too dim and I'll drop it.

I used to have a pretty good stash of LM3909. I gave them to a coworker many years ago so he could make a bunch of flashers that would mark a dark path for an astronomy club. Something like this would work for this too.

With such a low current draw something like this could last a very long time.

 

Well, that is surprising. I tried a 7555, and got a shade over 9ma. I switched over to a TLC555, and had a shade under 24ma.

Guess I'll publish it, with the proviso that the current goes all over the place, and how to measure it.