'LED Flasher Competition' from 1998

[I posted this last night, 2003 Aug 16 11:00 PM, but I didn't see it show up on the server. I don't know why.]

I was googling groups and I came across this post from 1998. It also talked about the Forever Flasher, AKA Infini-Flasher that we talked about here a week or so ago.

He claimed that the Infini-Flasher was efficient and the only circuit that would run for 20 hours off .3 farad.

I built the circuit at the following URL and talked about it in the SED, etc. newsgroups. It says runs 6 months off a 150 mAh battery.

The first one I built used about 100 microamps average at 3VDC and would run for 90 or so seconds off a 6800 uF cap. I built another one, and changed some parts values, and lowered the 4700 ohm resistor to 2200, which gave it a shorter pulse width, and that reduced the average current. Now it draws an average of 44 microamps and runs for over 5 minutes off the 6800 uF cap. It's getting really boring timing this with a stopwatch.

This morning I made further mods and I think the current is down to 12 to 14 microamps, but since the DMM draws a couple uA, I have to kind of finagle that into the total.

Here are the changes I made: Was: Changed to:

4.7M -> 10M and 3.3M in series

0.68 uF -> 0.1 uF ceramic disk

2907 and 2222 -> 2N4403 and 2N4401

680 ohm -> 1.5k

4.7k -> 2.2k

47 ohm -> 220 ohm

With these values and an average current of less than 14 uA, the pulses are 1 every .9 sec or about 1.1 Hz. It will run off the 6800 uF cap for over 11 minutes.

I figure that it would take 130 of the 6800 uF caps in parallel to run this circuit for 24 hours, or a total of about .9 farads. That's about 3 times as much capacitance as the Infini-Flasher. Which seems to indicate that the Infini-Flasher is drawing only 6 or 7 microamps.

Slowing the flash rate down to much less than once a second will also reduce the average current. That's how he made the Infini-Flasher last a lot longer on the .3F caps. I can incease the 13.3M resistor to 22 Megs, which will probably give it about 2 sec between flashes. If I get it to run off the 6800 uF cap for 20 minutss, then the 24 hour cap would need to be less than half a farad.

I didn't want to have to use the inductor. I presume this is used to boost the supply V up to the LED's V. When my circuit gets down to less than 2V, the LED starts to dim and the circuit soon quits. If I were to use a V boost circuit, it would probably operate down to a much lower voltage, like the Infini Flasher.

Here's a part of the original post from 1998.

----------------

This is called the "Forever Flasher" because, given an adequite supply of sunlight once a day it will flash forever (*even* during night-time and low light conditions).

As you said it would be pretty trivial to make an LED flasher that ran on solar cells. This one is special because it will run for about 24 hours or so in complete darkness with only .3F worth of caps. I did indeed try an LM3909 for this and it only lasted about 20 minutes on the same caps!

This is the only circuit idea I know of that will run for about

24 hours in complete darkness flashing an LED off of .3F worth of caps. Can someone come up with another circuit that's more effecient? "The sci.electronics.design LED Flasher Competition!" :-)

------------------

--
@@F@r@o@m@@O@r@a@n@g@e@@C@o@u@n@t@y@,@@C@a@l@,@@w@h@e@r@e@@
###Got a Question about ELECTRONICS?   Check HERE First:###
http://users.pandora.be/educypedia/electronics/databank.htm
My email address is whitelisted.  *All* email sent to it 
goes directly to the trash unless you add NOSPAM in the 
Subject: line with other stuff.  alondra101  hotmail.com
Don't be ripped off by the big book dealers.  Go to the URL
that will give you a choice and save you money(up to half).
http://www.everybookstore.com  You'll be glad you did!
Just when you thought you had all this figured out, the gov't
changed it: http://physics.nist.gov/cuu/Units/binary.html@@t@h@e@@a@f@f@l@u@e@n@t@@m@e@e@t@@t@h@e@@E@f@f@l@u@e@n@t@@

That's not a bad design.

Here's some suggestions for making a more efficient flasher:

#1: Use a FET in a blocking-oscillator setup. The FET will have much lower voltage drop than the transistor in the original design.

#2: Use a small toroidal transformer to couple the FET to the LED, about a

1.5:1 stepdown ratio. Remove the 47 ohm current-limiting resistor.

For a 2N4401 at a few dozen milliamps, the V drop will be less than a tenth of a volt. That's no worse than a FET. For a BC337, it's even less. Besides, you need a few volts to turn the Enh mode FET on fully, and there may not be a few volts available.

Too expensive and too much hassle. The toroid has to be hand wound, and that's a big pain, too. I bought some Amidon toroids, a buck apiece! That's more expensive than all the rest of the parts combined.

--
----------------(from OED Mini-Dictionary)-----------------
PUNCTUATION - Apostrophe
Incorrect uses: (i) the apostrophe must not be used with a plural
where there is no possessive sense, as in ~tea's are served here~;
(ii) there is no such word as ~her's, our's, their's, your's~.

Confusions: it's = it is or it has (not 'belonging to it'); correct
uses are ~it's here~ (= it is here); ~it's gone~ (= it has gone);
but ~the dog wagged its tail~ (no apostrophe).
----------------(For the Apostrophe challenged)----------------
 From a fully deputized officer of the Apostrophe Police!



I bought some batteries, but they weren't included,
so I had to buy them again.
-- Steven Wright

FOR SALE: Nice parachute: never opened - used once.

F
o
d
d
e
r

f
o
r

s
t
u
p
i
d

n
o
t

e
n
o
u
g
h

i
n
c
l
d
u
d
e
d

t
e
x
t

e
r
r
o
r

m
s
g
.

lower

I respectfully disagree. FET's can have much less than an ohm of on resistance.

For a BC337, it's even

That's the beauty of a blocking oscillator-- all you need is a bit of forward bias and the oscillator coil does the rest. No high voltage needed!

about a

I thought we were optimizing for best efficiency, which is often not the same as least cost!

In article , snipped-for-privacy@umn.edu mentioned...

A BJT can have less than an ohm, too. But then you don't need to use a 2N3055 for a few dozen milliamps. A 2N7000 is a typical enhancement mode FET that could be used for a single LED. It has 5 ohms resistance.

That's the problem, too. It requires an oscillator coil.

--
@@F@r@o@m@@O@r@a@n@g@e@@C@o@u@n@t@y@,@@C@a@l@,@@w@h@e@r@e@@
###Got a Question about ELECTRONICS?   Check HERE First:###
http://users.pandora.be/educypedia/electronics/databank.htm
My email address is whitelisted.  *All* email sent to it 
goes directly to the trash unless you add NOSPAM in the 
Subject: line with other stuff.  alondra101  hotmail.com
Don't be ripped off by the big book dealers.  Go to the URL
that will give you a choice and save you money(up to half).
http://www.everybookstore.com  You'll be glad you did!
Just when you thought you had all this figured out, the gov't
changed it: http://physics.nist.gov/cuu/Units/binary.html@@t@h@e@@a@f@f@l@u@e@n@t@@m@e@e@t@@t@h@e@@E@f@f@l@u@e@n@t@@

In article , snipped-for-privacy@umn.edu mentioned...

A BJT can have less than an ohm, too. But then you don't need to use a 2N3055 for a few dozen milliamps. A 2N7000 is a typical enhancement mode FET that could be used for a single LED. It has 5 ohms resistance.

That's the problem, too. It requires an oscillator coil.

I thought we were making a flasher, not a piece of test equipment. :-) But I don't see anything wrong with using a toroid, or an inductor. It's just that a toroid isn't as easy to wind as a ferrite bobbin. I'm thinking it should be on the level of the average person's skill. And an average person doesn't have toroid building skills.

I also wouldn't expect that a 5 or 10 amp power FET would be used in this circuit, which needs only a hundred or so mA peak to drive the LED. A 2N7000 costs a dime, a high power FET can cost several dollars. A high current, low Vce(sat) transistor such as the 2SD965 in the TO-92 package might cost only a quarter.

I've recently built some blocking oscillators using a toroid, fine wire, and a single transistor. This is for a V boost circuit to convert a single 1.5V cell into 3.6V for the white LED.

The trend is for these circuits to use a Maxim or similar chip, surface mount it with other parts such a schottky diode and transistor to make a nickel-sized circuit board. SSome chips have a shutdown pin, which could be gated to give the LED flash capability.

Some of these circuits have an efficiency of close to 90 percent. But because the chip runs at 2 MHz and has high currents, it's like designing an RF power amp, so you might as well buy the evaluation PC board. And the board has to act as a heat sink for those tiny parts.

I'm just looking for a circuit that is optimized for inexpensive and readily available parts, simple enough for the average hobbyist to build. I'm not against change. A year or two ago I saw a number of projects that used the BS170 or 2N7000 enh mode FETs for low power switching, and I thought that these were too unusual and hard to find for the average hobbyist. But I bought a bag of the 2N7000s, and I've now seen how they can be used for projects to take advantage of the FET's unique characteristics. But I still can't see using a power FET with .05 ohms and costing two dollars or more to drive a single LED.

Oh, one other thing about a blocking osc. I've worked on tube equipment that used the transformers for blocking oscillators. But these worked at hundreds of kHz, not 1 flash a second. I would think that to get one to oscillate at 1 Hz, the transformer would have to be very big.

I'd like to see a circuit that uses both V boost and a flasher. That way, a blue or white LED could flash off a 3V supply.

--
@@F@r@o@m@@O@r@a@n@g@e@@C@o@u@n@t@y@,@@C@a@l@,@@w@h@e@r@e@@
###Got a Question about ELECTRONICS?   Check HERE First:###
http://users.pandora.be/educypedia/electronics/databank.htm
My email address is whitelisted.  *All* email sent to it 
goes directly to the trash unless you add NOSPAM in the 
Subject: line with other stuff.  alondra101  hotmail.com
Don't be ripped off by the big book dealers.  Go to the URL
that will give you a choice and save you money(up to half).
http://www.everybookstore.com  You'll be glad you did!
Just when you thought you had all this figured out, the gov't
changed it: http://physics.nist.gov/cuu/Units/binary.html@@t@h@e@@a@f@f@l@u@e@n@t@@m@e@e@t@@t@h@e@@E@f@f@l@u@e@n@t@@

I've done this, though admittedly using 5 transistors. Two two transistor astables, one power transistor.

The fast astable has the base resistors commoned and run from the output of the slow one. Starts up at ~0.498V to run a 3.6V LED.

In this case, the variable output power is a feature to indicate battery state.

Efficiancy is quite good, as the fast astable (with the low value resistors) has only leakage current flowing when the power is off, and the slow astable has very high value resistors.

However, I've been disssapointed in my (very brief) search for high value SMD resistors, ideally in the smaller packages, the ranges seem to stop at 1M.

25M would be nice, 100M would be lovely.

--
http://inquisitor.i.am/    |  mailto:inquisitor@i.am |             Ian Stirling.
---------------------------+-------------------------+--------------------------
"The theory of everything falls out trivially." -- Etherman, sci.physics kook.

In article , snipped-for-privacy@mauve.demon.co.uk mentioned...

Astable meaning astable multivibrators? The ratio of a mvb assuming equal parts is 50%, but for a flasher, it might be better at 10% on and 90% off. Easy to do with the right choice of caps and res. But it mioght be easier to use a flasher circuit. I've built some flasher circuits that draw only a few mA, used as yours did to drive the base resistor of a CW RF beacon.

If you ever want to see some odd astables, visit Torrens' 4QD website and check those out.

--
@@F@r@o@m@@O@r@a@n@g@e@@C@o@u@n@t@y@,@@C@a@l@,@@w@h@e@r@e@@
###Got a Question about ELECTRONICS?   Check HERE First:###
http://users.pandora.be/educypedia/electronics/databank.htm
My email address is whitelisted.  *All* email sent to it 
goes directly to the trash unless you add NOSPAM in the 
Subject: line with other stuff.  alondra101  hotmail.com
Don't be ripped off by the big book dealers.  Go to the URL
that will give you a choice and save you money(up to half).
http://www.everybookstore.com  You'll be glad you did!
Just when you thought you had all this figured out, the gov't
changed it: http://physics.nist.gov/cuu/Units/binary.html@@t@h@e@@a@f@f@l@u@e@n@t@@m@e@e@t@@t@h@e@@E@f@f@l@u@e@n@t@@

Current was a couple of uA at 1.2V, using rather high value resistors, and very unequal capacitors. (1n and 680n)

It doesn't quite work like a 'proper' astable, the 680n capacitor doesn't charge anywhere near all the way. IIRC, it was around 60:1.

Will have a look.

--
http://inquisitor.i.am/    |  mailto:inquisitor@i.am |             Ian Stirling.
---------------------------+-------------------------+--------------------------
  Pieces of eight!
  Pieces of eight!
  Pieces of eight!
  Pieces of eight!
  Pieces of eight!
  Pieces of nine!