2AA 2Q Step-up Circuit help needed

[Illuminated]

I bought several of these on close-out after Halloween at Meijer for cheap.

I was suprised to find a 2-AA step-up circuit similar to that of the Brinkmanns. Interesting how the LED is ground similar to those of LEDCorp's PR-type as used in the Turtle Light I and others for wider dispersion. Here's a crude circuit sketch...

It is intersting how the LED is turned off if the solar panel is exposed to light (if the switch is on). I'm thinking these might be well suited for power-outage lights and possibly even used for backpacking, if I re-package them.

My question is...could I boost the output enough to drive 2-3 5mm Leds at reasonable brightness? Change the inductor and/or play with the value of R1? Also, what does the diode in series with the LED do for this circuit?

These things are alright as they are given the price I paid, but I'm hoping they have the potential to be better...

Thanks in advance - John

 

[MrAl]

Hello there,

From what i can see in that drawing, the circuit looks a little
strange. Normally someone doesnt put one or two caps in series
with a transistor CE junction and power supply when that
transistor acts as a switch in the circuit. Usually
there is at least one resistor in series with the cap(s).
Also, experienced designers know that putting a diode
in series with something (the LED here) doesnt protect it
from reverse current flow as good as it can if it was
connected in anti-parallel to the device. The capacitance
of the diode doesnt stop reverse current flow for some
small time period.

You might want to double check the circuit layout to make
sure this is the correct connection. Also, if you could
draw the part values and/or part numbers a little larger
i could throw this thing into a circuit simulator and
see what comes out of it.

Offhand, you might try two LED's in series first.
If the current is too low, you may be able to carefully
lower the value of R1 (if it really is a resistor) to
increase the current output somewhat.

I could probably tell you more if you clarify the
circuit and parts in a new drawing so i can do
a simulation.

Take care,
Al

 

[Illuminated]

MrAl,

Thanks for the reply - I edited my original post with a better scan of my sketch. (Sorry, I don't have any software for diagramming on my home computer).I rechecked the circuit and I believe this is correct. Cap values were not visible on the devices, but were screened on the ckt board. Resistors had values on both devices and ckt board. I assume the color coding for the inductor folows that of carbon resistors. Transistor markings were difficult to read even with a loupe, and I seem to remember seeing different markings on at least one of the other units I looked at.

I'm guessing that the 220K resistor and 2700pf cap have something to do with the auto-off feature whenever the solar cells generate energy. LED dims proportionately with amount of light reaching solar panel.

Thanks for taking a look - John

 

[MrAl]

Hello again John,

Sorry but your new schematic isnt showing up in the web
page. Perhaps you could check the url to see if it's
correct or something?

Take care,
Al

 

[MrAl]

Hello again John,

That's a much better schematic :-)

Taking a look at it here's what i found out...

I used test 'batteries' of 2.4v and 3.0v for
high input and low battery input during the
simulations.

1.
30ma out at 3.0v in, 15ma out at 2.4v in,
efficiency was only 43% at 3.0v in.

2.
Connecting the diode in anti-parallel to the LED
instead of in series, the current went up:
40ma out at 3.0v in, 20ma out at 2.4v in, and the
efficiency jumped way up to 62% !!

3. After connecting the diode in anti-parallel AND
increasing the 2400 ohm resistor to 4000 ohms, the
output was back to 30ma out at 3.0v in and 15ma out
at 2.4v in, and the efficiency went up a little more,
to 70% .


The conclusion here is that connecting the diode
in anti-parallel to the LED instead of in series
raises the efficiency quite a bit.

Multiple LEDs on the output:
Dropping the 2400 to 1000 and the 10k to 5k the circuit
looked like it was able to drive 4 LED's on the output
connected in parallel. The efficiency dropped also
though, to about 55% .

Connecting two LED's in series on the output didnt
look like it would work very well because at about
20ma output current the efficiency dropped to around 16% .


Take care,
Al