Joule Thief - still not working....

[default]

On Sun, 9 Aug 2009 20:17:32 -0700 (PDT), fungus
<openglMYSOCKS@artlum.com> wrote:

On Aug 9, 10:24 pm, default <defa...@defaulter.net> wrote:

That's the beauty of them.  Only need a solderless breadboard and
maybe a 9 pin RS232 connector plus 2-3 resistors.


It might be worth getting a proper PCB and a ZIF socket
if you're doing lots of them...

Yeah a zif socket would be good if I were to use more of them - or
that way insanity lies - make it too complicated, starting out, by
trying to build in too many "features" and I'd never get around to
building it, and it would cost more.

From what I read on the forum some folks do seem to be in that rut -
always trying to come up with the best, universal, do it all,
breadboard.

I've got two breadboards dedicated to the axes. Both mounted on some
wood to hold the battery holders, with the RS232 connectors and
programming resistors mounted. A three position toggle switch selects
the voltage and turns it off - which is handy to reset the 'axe. I
wire the connections to the program socket and battery connections to
some short lengths of paper-clip wire, and shrink tubing, to make it
rugged and survive being moved around on the BB.

The program resistors aren't necessary on the finished axe project,
but if it is easy to move to the computer (unlike my range controls)
then I just put the resistors in the project and add a three pin
header to reprogram it with.
--

 

[Jon Kirwan]

On Mon, 10 Aug 2009 00:22:41 -0700 (PDT), fungus
<openglMYSOCKS@artlum.com> wrote:

On Aug 10, 7:00 am, Jon Kirwan <j...@infinitefactors.org> wrote:

Hehe.  You aren't too far off, really.  Imagine that the two diodes
are forward biased (conducting normally.)  They exhibit a voltage drop
that doesn't change a lot with changes in current.

Let's see if I can reduce it to terms my monkey brain can understand.

Why not use two 1N4148's, instead?  You could.  The base would be a
little higher up, you'd need a little more battery voltage for
everything to work according to easier design rules, that's all. There
are other differences which relate to how the 1N4148, BJT, and
schottky respond to temperature variations, though.

The two diodes and R1 are chosen so that the voltage across
each diode is just above its turn-on voltage and the voltage
at the bases of the BJTs is enough to allow 20mA through the
LEDs.

Kind of, except that R1 is also sourcing current to the chain of BJTs,
as well. For the most part, though, it's just sourcing into the
diodes. It's selected so that it provides a torrent of current that
way exceeds what the BJTs need but not so much that it needlessly
wastes current/power via the diodes.

His 20-40mA target with the 100 ohm resistor might be considered on
the high side. Or not. But it's definitely not 'just above' their
turn on voltages. It's way, way above that. Just imagine a 4" water
pipe of water running through the 100 ohm resistor and diode stack,
with a very thin garden hose tapped off of it going to each BJT base.
That 4" pipe could just as well have been a 3" pipe or even a 2" pipe
and the hose would have worked just as well. Paul, I guess, didn't
want to fool around and used a 4" pipe. The voltages on the diodes is
determined by the flow, so if there is a LOT of flow that is fixed in
rate and the BJTs only tap off a tiny part of it then the voltage will
be fairly stable. If he'd used a lot less flow, the BJTs might tap
off a significant part and then the voltage would vary a little
depending on how many LEDs there were hooked up.

As the battery voltage drops... the resistance of the diodes
starts to rises exponentially, thus raising the voltage at
the base of the BJT and allowing more current through the LED.

Hmm. I would have used inversely proportionately, than exponentially.
Roughly speaking, if you ignore that diversion I wrote, the voltage
across the diodes remains fixed regardless of the current. And the
current varies with the battery voltage via the 100 ohm resistor. So
the 'resistance' of the diodes is determined by the 100 ohm resistor
and is approximately 100 ohms times 1V/(Vbattery-1V). At Vbattery of
5V, that's about 25 ohms for example. Not exactly exponential.

The side note I wrote was more about the __change__ in the 1V estimate
used above, as current varies through them. In other words, the 1V
used in the large scale overview (zeroth order, assuming constant
behavior) is adjusted slightly by a smaller scale event that makes
that equation more complicated, when you look a little deeper (beyond
the zeroth order.) If the immediately above paragraph were completely
true, then the LED currents wouldn't change at all from varying the
battery voltage. But they do, because of a smaller effect that messes
with the voltage a little, which I tried to explain. Although it's at
orders below zeroth order, it does matter enough to care about at
times. The voltage across the diodes can easily vary by 10% or more
because of it. Which is why I bothered to write about it, at all.

Jon

 

[bw]

"fungus" <openglMYSOCKS@artlum.com> wrote in message
news:1fa52b8a-5e8b-44e1-931e-cd4ca416185e@k26g2000vbp.googlegroups.com...
On Aug 10, 9:24 am, ehsjr <eh...@NOSPAMverizon.net> wrote:

fungus wrote:

snip

As an aside, would a transformer not work for a joule
thief?

While it may have been called an "inductor" in the thread,
that's just imprecise wording. Throughout the thread, it
has always been a transformer in spite of however it has been
worded, as long as it referred to the component in the joule
thief circuit.

I guess what I meant was, would a 'transformer'
which is designed for efficient conversion of
voltages be suitable for making efficient joule
thieves. Maybe the characteristics are completely
different.

All the cheap film cameras with flash have them. A local Goodwill store will
have a box full of cheap chinese cameras. Take one apart. The flash unit
just takes the 3 volts (from 2 AA cells) and fills a 100 uF capacitor to 330
volts. The switching xformer is about a 1/4 inch cube. There will also be a
"pulse" transformer that is not in the power supply, it only flashes the
xenon tube.

 

[David Eather]

bw wrote:

"fungus" <openglMYSOCKS@artlum.com> wrote in message
news:1fa52b8a-5e8b-44e1-931e-cd4ca416185e@k26g2000vbp.googlegroups.com...
On Aug 10, 9:24 am, ehsjr <eh...@NOSPAMverizon.net> wrote:
fungus wrote:

snip

As an aside, would a transformer not work for a joule
thief?
While it may have been called an "inductor" in the thread,
that's just imprecise wording. Throughout the thread, it
has always been a transformer in spite of however it has been
worded, as long as it referred to the component in the joule
thief circuit.


I guess what I meant was, would a 'transformer'
which is designed for efficient conversion of
voltages be suitable for making efficient joule
thieves. Maybe the characteristics are completely
different.

All the cheap film cameras with flash have them. A local Goodwill store will
have a box full of cheap chinese cameras. Take one apart. The flash unit
just takes the 3 volts (from 2 AA cells) and fills a 100 uF capacitor to 330
volts. The switching xformer is about a 1/4 inch cube. There will also be a
"pulse" transformer that is not in the power supply, it only flashes the
xenon tube.

Don't be tempted to power the camera in the dismantled state. The
capacitor can store power for a little while and will give you a serious
bite. (take a 5 or 10 minute delay after powering the circuit before you
start pulling it apart)

 

[ehsjr]

David Eather wrote:

bw wrote:

"fungus" <openglMYSOCKS@artlum.com> wrote in message
news:1fa52b8a-5e8b-44e1-931e-cd4ca416185e@k26g2000vbp.googlegroups.com...
On Aug 10, 9:24 am, ehsjr <eh...@NOSPAMverizon.net> wrote:

fungus wrote:

snip

As an aside, would a transformer not work for a joule
thief?

While it may have been called an "inductor" in the thread,
that's just imprecise wording. Throughout the thread, it
has always been a transformer in spite of however it has been
worded, as long as it referred to the component in the joule
thief circuit.


I guess what I meant was, would a 'transformer'
which is designed for efficient conversion of
voltages be suitable for making efficient joule
thieves. Maybe the characteristics are completely
different.

All the cheap film cameras with flash have them. A local Goodwill
store will have a box full of cheap chinese cameras. Take one apart.
The flash unit just takes the 3 volts (from 2 AA cells) and fills a
100 uF capacitor to 330 volts. The switching xformer is about a 1/4
inch cube. There will also be a "pulse" transformer that is not in the
power supply, it only flashes the xenon tube.



Don't be tempted to power the camera in the dismantled state. The
capacitor can store power for a little while and will give you a serious
bite. (take a 5 or 10 minute delay after powering the circuit before you
start pulling it apart)

And short the capacitor with a screwdriver blade as soon as you
get it far enough apart to do so. Otherwise, you might be in
for an "exhilarating" experience.

Ed

 

[bw]

"ehsjr" <ehsjr@NOSPAMverizon.net> wrote in message
news:2Ifgm.1688$Jg.556@nwrddc01.gnilink.net...

David Eather wrote:
bw wrote:


All the cheap film cameras with flash have them. A local Goodwill store
will have a box full of cheap chinese cameras. Take one apart. The flash
unit just takes the 3 volts (from 2 AA cells) and fills a 100 uF
capacitor to 330 volts. The switching xformer is about a 1/4 inch cube.
There will also be a "pulse" transformer that is not in the power
supply, it only flashes the xenon tube.


Don't be tempted to power the camera in the dismantled state. The
capacitor can store power for a little while and will give you a serious
bite. (take a 5 or 10 minute delay after powering the circuit before you
start pulling it apart)


And short the capacitor with a screwdriver blade as soon as you
get it far enough apart to do so. Otherwise, you might be in
for an "exhilarating" experience.

Ed

Good advice. Newer caps seem to hold charge for days or longer. Older caps,
such as in Polaroids, leak faster. You can see the technology advance in
that the same voltage & capacity caps are now much smaller volume, and leak
less.
Charge drops slowly to about 50 volts, then holds there for much longer,
even weeks.
Also, I've found a couple flash units that use a single AA cell to charge.
Most use 2 AA cells, older and larger flash units used 6 volts. Older styles
are also easier to reverse engineer and had larger transformers, making them
easier to remove and re-use.

 

[Paul E. Schoen]

"fungus" <openglMYSOCKS@artlum.com> wrote in message
news:a1ffc1c2-83e0-4bd1-808a-3c59f6bc7e20@j9g2000vbp.googlegroups.com...
On Aug 9, 10:24 pm, default <defa...@defaulter.net> wrote:

That's the beauty of them. Only need a solderless breadboard and
maybe a 9 pin RS232 connector plus 2-3 resistors.

It might be worth getting a proper PCB and a ZIF
socket if you're doing lots of them...

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

I have always used Microchip tools and their PICkits are good and
inexpensive. Their development tools (MPLab) are free and excellent. Now
that I mostly use SMT components, I design a 5 pin SIP header on the PCB
and I can either stick a connector in the holes temporarily or install a
header and use a female connector to do the programming. The PICkit will
provide the 5 pin connection with everything needed for ICSP, and it only
costs about $25.

Paul

 

[Paul E. Schoen]

"Jon Kirwan" <jonk@infinitefactors.org> wrote in message
news:bgk0851212gk2vk6s973umplt268f7ee21@4ax.com...

On Mon, 10 Aug 2009 00:22:41 -0700 (PDT), fungus
openglMYSOCKS@artlum.com> wrote:


Let's see if I can reduce it to terms my monkey brain can understand.

The two diodes and R1 are chosen so that the voltage across
each diode is just above its turn-on voltage and the voltage
at the bases of the BJTs is enough to allow 20mA through the
LEDs.

Kind of, except that R1 is also sourcing current to the chain of BJTs,
as well. For the most part, though, it's just sourcing into the
diodes. It's selected so that it provides a torrent of current that
way exceeds what the BJTs need but not so much that it needlessly
wastes current/power via the diodes.

I tried to make the circuit as simple as possible while allowing it to work
at the minimum possible battery voltage. The 100 ohm resistor allowed it to
work at lower voltage but at the expense of excess current at the high end.
A better design would be a simple current regulator into the two diodes, or
even a regulated current into a resistor which would provide a stable
reference voltage.

Better regulation can be achieved with a higher reference voltage, at the
expense of a higher minimum operating voltage. Two 1N4148s would mean a 1.2
to 1.4V reference, and about 0.6 volts dropped in the emitter resistors.
The 1N5818 allows this to be as low as 0.2 volts, but regulation and tempco
suffer.

A precision 0.9 volt reference is available and it draws only a few mA from
the battery. It's in a tiny SOT-23 package and costs less than a dollar.
But then the circuit's efficiency is still limited to the resistive drop
from the battery voltage to the forward voltage of the LEDs. And you need
enough supply voltage to drive the highest voltage LED, which could be
about 3 volts for white or blue.

The most efficient circuit would have several LEDs in series, and step up
the battery voltage enough to drive them. A charge pump and a PWM drive
into a small inductor can approach 90% efficiency, but requires some
relatively complex circuitry using ICs. A variation on the basic Joule
Thief can be constructed with only two BJTs and several resistors,
capacitors, and diodes, with a good coupled inductor (transformer) for a
couple of dollars, and probably give 20% regulation and 70% efficiency
without much difficulty.

For special effects, such as sound modulation, more circuitry must be
added, and the effects need to be defined. You can have several colors
respond to various intensities and tones of sound, and have features like
fast-attack and slow-decay. For such a project, it's probably best to boost
the battery voltage to perhaps 12-20 VDC and then use a PIC to do all the
"dirty work". That way you can just make a simple circuit and then use
programming to achieve all the effects, and make changes as you desire
without even touching the wrong end of a slobbering iron.

Paul

 

[Jon Kirwan]

On Tue, 11 Aug 2009 21:56:11 GMT, "Paul E. Schoen" <paul@peschoen.com>
wrote:

"Jon Kirwan" <jonk@infinitefactors.org> wrote in message
news:bgk0851212gk2vk6s973umplt268f7ee21@4ax.com...
On Mon, 10 Aug 2009 00:22:41 -0700 (PDT), fungus
openglMYSOCKS@artlum.com> wrote:


Let's see if I can reduce it to terms my monkey brain can understand.

The two diodes and R1 are chosen so that the voltage across
each diode is just above its turn-on voltage and the voltage
at the bases of the BJTs is enough to allow 20mA through the
LEDs.

Kind of, except that R1 is also sourcing current to the chain of BJTs,
as well. For the most part, though, it's just sourcing into the
diodes. It's selected so that it provides a torrent of current that
way exceeds what the BJTs need but not so much that it needlessly
wastes current/power via the diodes.

I tried to make the circuit as simple as possible while allowing it to work
at the minimum possible battery voltage. The 100 ohm resistor allowed it to
work at lower voltage but at the expense of excess current at the high end.

Yup. I didn't want to 'go there' in the explanation, since he was
trying to reduce things "to terms my monkey brain can understand."

A better design would be a simple current regulator into the two diodes, or
even a regulated current into a resistor which would provide a stable
reference voltage.

Provided by one BJT, within some reasonable definition of an attempt
at a constant current. Of course... headroom is needed for that, too.

Better regulation can be achieved with a higher reference voltage, at the
expense of a higher minimum operating voltage. Two 1N4148s would mean a 1.2
to 1.4V reference, and about 0.6 volts dropped in the emitter resistors.
The 1N5818 allows this to be as low as 0.2 volts, but regulation and tempco
suffer.

Yeah.

A precision 0.9 volt reference is available and it draws only a few mA from
the battery. It's in a tiny SOT-23 package and costs less than a dollar.
But then the circuit's efficiency is still limited to the resistive drop
from the battery voltage to the forward voltage of the LEDs. And you need
enough supply voltage to drive the highest voltage LED, which could be
about 3 volts for white or blue.

Yes.

The most efficient circuit would have several LEDs in series, and step up
the battery voltage enough to drive them. A charge pump and a PWM drive
into a small inductor can approach 90% efficiency, but requires some
relatively complex circuitry using ICs. A variation on the basic Joule
Thief can be constructed with only two BJTs and several resistors,
capacitors, and diodes, with a good coupled inductor (transformer) for a
couple of dollars, and probably give 20% regulation and 70% efficiency
without much difficulty.

For special effects, such as sound modulation, more circuitry must be
added, and the effects need to be defined. You can have several colors
respond to various intensities and tones of sound, and have features like
fast-attack and slow-decay. For such a project, it's probably best to boost
the battery voltage to perhaps 12-20 VDC and then use a PIC to do all the
"dirty work". That way you can just make a simple circuit and then use
programming to achieve all the effects, and make changes as you desire
without even touching the wrong end of a slobbering iron.

The boost itself could be driven by the PIC, of course. The special
effects would be fun, too.

I suppose someone could come up with a PIC coded up to allow
non-programmers to create special effects with it without having to
program -- or to design a special language that is focused on LED
driving and makes coding a snap to do. Perhaps a windows program
where you click on the pins and set functions for them with
standardized 'blocks' that you drag and drop and get to see simulated
on the screen before you 'download' it into the USB PIC-holder stick.
Pop the PIC out of the USB stick and drop it into your circuit, power
it, and away it goes.

Might be a specialized market.

Jon

 

[Paul E. Schoen]

"Jon Kirwan" <jonk@infinitefactors.org> wrote in message
newsd1485pj4cuuovqcgfgqsimgk04h8d35p6@4ax.com...

On Tue, 11 Aug 2009 21:56:11 GMT, "Paul E. Schoen" <paul@peschoen.com
wrote:


The most efficient circuit would have several LEDs in series, and step up
the battery voltage enough to drive them. A charge pump and a PWM drive
into a small inductor can approach 90% efficiency, but requires some
relatively complex circuitry using ICs. A variation on the basic Joule
Thief can be constructed with only two BJTs and several resistors,
capacitors, and diodes, with a good coupled inductor (transformer) for a
couple of dollars, and probably give 20% regulation and 70% efficiency
without much difficulty.

Here are some figures for the unregulated version:

Vin Iin Vout Iout Eff
1.10V 0.06A 5.91V 5.7mA 51%
2.03V 0.09A 7.08V 16.8mA 65%
2.61V 0.11A 7.75V 24.3mA 65%
3.16V 0.13A 8.40V 32.1mA 66%
3.57V 0.14A 8.85V 37.9mA 67%
4.10V 0.15A 9.43V 45.4mA 69%

And for the regulated version:

Vin Iin Vout Iout Eff
0.96V 0.045A 5.73V 4.3mA 57%
1.08V 0.045A 5.86V 5.4mA 65%
2.14V 0.051A 6.51V 11.1mA 66%
2.58V 0.043A 6.51V 11.1mA 65%
3.15V 0.031A 6.53V 11.3mA 75%
3.71V 0.026A 6.54V 11.4mA 77%
4.24V 0.023A 6.54V 11.4mA 76%
6.13V 0.016A 6.61V 12.1mA 82%
7.55V 0.021A 7.30V 19.2mA 88%

I used a 50 ohm sense resistor and I added a 39 uF capacitor on the output.
A 30 ohm resistor should give similar results with about 20 mA output. This
is still a very simple, inexpensive circuit that I was able to build in
about 20 minutes. I was actually surprised at how well it regulates up to
the point where the input voltage exceeds the LED forward drop.

Following is the LTSpice ASCII for the circuit.

Paul

-------------------------------------------------
Version 4
SHEET 1 880 680
WIRE -64 32 -80 32
WIRE 0 32 -64 32
WIRE 288 32 0 32
WIRE 400 32 368 32
WIRE 448 32 400 32
WIRE 464 32 448 32
WIRE 464 48 464 32
WIRE 0 64 0 32
WIRE 288 64 288 32
WIRE 368 64 368 32
WIRE -80 144 -80 32
WIRE 400 144 400 32
WIRE 464 144 464 112
WIRE 288 160 288 144
WIRE 368 160 368 128
WIRE 368 160 288 160
WIRE 0 208 0 144
WIRE 16 208 0 208
WIRE 112 208 96 208
WIRE 208 208 192 208
WIRE 224 208 208 208
WIRE 0 240 0 208
WIRE 464 240 464 208
WIRE 208 272 208 208
WIRE 256 272 208 272
WIRE 288 288 288 256
WIRE 400 288 400 208
WIRE 400 288 288 288
WIRE -80 352 -80 224
WIRE 0 352 0 304
WIRE 0 352 -80 352
WIRE 160 352 0 352
WIRE 208 352 208 336
WIRE 208 352 160 352
WIRE 288 352 288 288
WIRE 288 352 208 352
WIRE 368 352 288 352
WIRE 464 352 464 304
WIRE 464 352 448 352
WIRE 256 384 256 272
WIRE 368 384 368 352
WIRE 368 432 320 432
WIRE 464 432 464 352
WIRE 464 432 448 432
WIRE 160 480 160 352
WIRE 256 480 160 480
FLAG 368 384 0
FLAG 448 32 out
FLAG -64 32 in
SYMBOL npn 224 160 R0
SYMATTR InstName Q1
SYMATTR Value 2N3904
SYMBOL res -16 48 R0
SYMATTR InstName R1
SYMATTR Value 5k
SYMBOL cap -16 240 R0
SYMATTR InstName C1
SYMATTR Value 10n
SYMBOL ind2 304 160 R180
WINDOW 0 36 80 Left 0
WINDOW 3 36 40 Left 0
SYMATTR InstName L1
SYMATTR Value 100ľ
SYMATTR Type ind
SYMBOL ind2 96 224 R270
WINDOW 0 32 56 VTop 0
WINDOW 3 5 56 VBottom 0
SYMATTR InstName L2
SYMATTR Value 100ľ
SYMATTR Type ind
SYMBOL LED 448 144 R0
SYMATTR InstName D1
SYMATTR Value AOT-2015
SYMATTR Description Diode
SYMATTR Type diode
SYMBOL LED 448 240 R0
SYMATTR InstName D2
SYMATTR Value AOT-2015
SYMATTR Description Diode
SYMATTR Type diode
SYMBOL LED 448 48 R0
WINDOW 0 24 1 Left 0
SYMATTR InstName D3
SYMATTR Value AOT-2015
SYMATTR Description Diode
SYMATTR Type diode
SYMBOL voltage -80 128 R0
WINDOW 123 0 0 Left 0
WINDOW 39 0 0 Left 0
SYMATTR InstName V1
SYMATTR Value 3
SYMBOL res 0 224 R270
WINDOW 0 32 56 VTop 0
WINDOW 3 0 56 VBottom 0
SYMATTR InstName R2
SYMATTR Value 1k
SYMBOL schottky 224 336 R180
WINDOW 0 24 72 Left 0
WINDOW 3 18 5 Left 0
SYMATTR InstName D4
SYMATTR Value 1N5818
SYMATTR Description Diode
SYMATTR Type diode
SYMBOL schottky 384 128 R180
WINDOW 0 32 68 Left 0
WINDOW 3 16 -2 Left 0
SYMATTR InstName D5
SYMATTR Value 1N5818
SYMATTR Description Diode
SYMATTR Type diode
SYMBOL cap 384 144 R0
SYMATTR InstName C2
SYMATTR Value 39ľ
SYMBOL npn 320 384 M0
SYMATTR InstName Q2
SYMATTR Value 2N3904
SYMBOL res 352 368 R270
WINDOW 0 32 56 VTop 0
WINDOW 3 5 63 VBottom 0
SYMATTR InstName R3
SYMATTR Value 50
SYMBOL res 352 448 R270
WINDOW 0 32 56 VTop 0
WINDOW 3 0 56 VBottom 0
SYMATTR InstName R4
SYMATTR Value 500
TEXT 80 96 Left 0 !K1 L1 L2 1
TEXT -112 376 Left 0 !.tran 15m
TEXT -400 112 Left 0 ;4V 13.1mA 57 kHz 92%
TEXT -400 136 Left 0 ;3V 12.8mA 64 kHz 90%
TEXT -400 160 Left 0 ;2V 12.8mA 77 kHz 84%

 

[fungus]

On Aug 11, 10:23 pm, "bw" <bweg...@hotmail.com> wrote:

Also, I've found a couple flash units that use a single AA cell to charge..
Most use 2 AA cells, older and larger flash units used 6 volts. Older styles
are also easier to reverse engineer and had larger transformers, making them
easier to remove and re-use.

I've seen a lot of web sites using the flash
unit from disposable cameras to make JTs.

eg. http://www.overunity.com/index.php?topic=6942.0

They use them to light up fluorescent tubes,
400LEDs with a single AA battery, etc.

 

[Paul E. Schoen]

"fungus" <openglMYSOCKS@artlum.com> wrote in message
news:d45ed82f-6cc1-4ee7-8b91-4eb529bbf97d@c2g2000yqi.googlegroups.com...
On Aug 11, 10:23 pm, "bw" <bweg...@hotmail.com> wrote:

Also, I've found a couple flash units that use a single AA cell to
charge.
Most use 2 AA cells, older and larger flash units used 6 volts. Older
styles
are also easier to reverse engineer and had larger transformers, making
them
easier to remove and re-use.

I've seen a lot of web sites using the flash
unit from disposable cameras to make JTs.

eg. http://www.overunity.com/index.php?topic=6942.0

They use them to light up fluorescent tubes,
400LEDs with a single AA battery, etc.

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

I would be careful about what you may read on that website. Just the name
overunity is a red flag. Usually the claims of getting more power out than
is put in are based on incorrect assumptions and bad measurements that
don't take into account the effects of unusual waveforms and phase shift
and cheap meters. I did not see any figures for efficiency of any of the
circuits shown, but I did see links for nonsense like a fan that charges
its own battery (unless it uses external wind power).

Paul

 

[fungus]

On Aug 12, 9:35 am, "Paul E. Schoen" <p...@peschoen.com> wrote:

I would be careful about what you may read on that website. Just the name
overunity is a red flag. Usually the claims of getting more power out than
is put in are based on incorrect assumptions and bad measurements that
don't take into account the effects of unusual waveforms and phase shift
and cheap meters.

Oh, don't worry ... I'm not going to pay any attention
to all that that nonsense. I believe firmly in the laws
of physics.

 

[default]

On Tue, 11 Aug 2009 21:31:03 GMT, "Paul E. Schoen" <paul@peschoen.com>
wrote:

"fungus" <openglMYSOCKS@artlum.com> wrote in message
news:a1ffc1c2-83e0-4bd1-808a-3c59f6bc7e20@j9g2000vbp.googlegroups.com...
On Aug 9, 10:24 pm, default <defa...@defaulter.net> wrote:

That's the beauty of them. Only need a solderless breadboard and
maybe a 9 pin RS232 connector plus 2-3 resistors.


It might be worth getting a proper PCB and a ZIF
socket if you're doing lots of them...

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

I have always used Microchip tools and their PICkits are good and
inexpensive. Their development tools (MPLab) are free and excellent. Now
that I mostly use SMT components, I design a 5 pin SIP header on the PCB
and I can either stick a connector in the holes temporarily or install a
header and use a female connector to do the programming. The PICkit will
provide the 5 pin connection with everything needed for ICSP, and it only
costs about $25.

Paul

From what I read, it looks like one of their programmers will read and

write to the "protected" areas of memory. It should be possible to
clone the picaxe boot loader with that programmer. I think it was
pickit 2 that does that - been awhile since I looked them up. I think
the clone of the programmer costs ~$25.

It isn't like RevEd is charging too much for the preprogrammed chips,
they do provide support, and are always porting the boot loader to
more of Microchips products, so cloning them isn't justifiable. But
it had crossed my mind . . .


--

 

[fungus]

On Aug 11, 11:56 pm, "Paul E. Schoen" <p...@peschoen.com> wrote:

Kind of, except that R1 is also sourcing current to the chain of BJTs,
as well.

Surely it's a tiny amount, not enough to affect the
voltage level at the transistor base...

I tried to make the circuit as simple as possible while allowing it to work
at the minimum possible battery voltage. The 100 ohm resistor allowed it to
work at lower voltage but at the expense of excess current at the high end.
A better design would be a simple current regulator into the two diodes, or
even a regulated current into a resistor which would provide a stable
reference voltage.

There was one in the middle of all this that seemed
to work best - it had a transistor where the two
diodes are and a direct feedback loop. I'm sure the
extra transistor is worth it.

The main problem I see with with this circuit is that
a 'deadbug' version for six LEDs would be quite messy.

The most efficient circuit would have several LEDs in series, and step up
the battery voltage enough to drive them. A charge pump and a PWM drive
into a small inductor can approach 90% efficiency, but requires some
relatively complex circuitry using ICs.

The PICAXE mentioned earlier can set PWM output on
its pins and might be ideal for this setup - you'd
even have a couple of pins left over for special
effects and an analogue input for sensing.

For special effects, such as sound modulation, more circuitry must be
added, and the effects need to be defined. ... use a PIC to do all the
"dirty work". That way you can just make a simple circuit and then use
programming to achieve all the effects, and make changes as you desire
without even touching the wrong end of a slobbering iron.

I don't think a PICAXE is powerful enough to be able
to do a frequency analysis of an input sound signal...

 

[fungus]

On Aug 12, 2:10 am, Jon Kirwan <j...@infinitefactors.org> wrote:

The boost itself could be driven by the PIC, of course.  The special
effects would be fun, too.

....or you could run the PIC output into an LM3914 and
get that to regulate the LED current for you

I suppose someone could come up with a PIC coded up to allow
non-programmers to create special effects with it without having to
program

But the programming's the easy part!

 

[Jon Kirwan]

On Wed, 12 Aug 2009 05:33:41 -0700 (PDT), fungus
<openglMYSOCKS@artlum.com> wrote:

On Aug 12, 2:10 am, Jon Kirwan <j...@infinitefactors.org> wrote:

The boost itself could be driven by the PIC, of course.  The special
effects would be fun, too.

...or you could run the PIC output into an LM3914 and
get that to regulate the LED current for you

I was talking about 'voltage boost' here. The LM3914 doesn't do that.

I suppose someone could come up with a PIC coded up to allow
non-programmers to create special effects with it without having to
program

But the programming's the easy part!

In that case... we should talk about that kind of schematic, shouldn't
we? PICs and some other micros can operate off of 2V or so. Call it
2 AA's, for something from 2.2V to 3.1V operation range. If you
select a down-to-2V microcontroller, you can do the voltage boost
control AND the PWM intensity control as well as all the special
effects from a single IC and a handful of cheap external parts.

Ready for that?

Jon

 

[Paul E. Schoen]

"fungus" <openglMYSOCKS@artlum.com> wrote in message
news:406f3eaa-e59d-4e13-b037-8ebfa48414d1@q23g2000yqn.googlegroups.com...
On Aug 11, 11:56 pm, "Paul E. Schoen" <p...@peschoen.com> wrote:

Kind of, except that R1 is also sourcing current to the chain of BJTs,
as well.

Surely it's a tiny amount, not enough to affect the
voltage level at the transistor base...

Yes, the BJT base current is only 100 uA or so. The high current in R1 is
to make sure the diodes are working in a more stable region of forward
voltage.

I tried to make the circuit as simple as possible while allowing it to
work
at the minimum possible battery voltage. The 100 ohm resistor allowed it
to
work at lower voltage but at the expense of excess current at the high
end.
A better design would be a simple current regulator into the two diodes,
or
even a regulated current into a resistor which would provide a stable
reference voltage.

There was one in the middle of all this that seemed
to work best - it had a transistor where the two
diodes are and a direct feedback loop. I'm sure the
extra transistor is worth it.

The main problem I see with with this circuit is that
a 'deadbug' version for six LEDs would be quite messy.

Yes, since each LED needs its own current source transistor and sense
resistor, and it still suffers from a minimum required voltage and poor
efficiency.

The most efficient circuit would have several LEDs in series, and step up
the battery voltage enough to drive them. A charge pump and a PWM drive
into a small inductor can approach 90% efficiency, but requires some
relatively complex circuitry using ICs.

The PICAXE mentioned earlier can set PWM output on
its pins and might be ideal for this setup - you'd
even have a couple of pins left over for special
effects and an analogue input for sensing.

For special effects, such as sound modulation, more circuitry must be
added, and the effects need to be defined. ... use a PIC to do all the
"dirty work". That way you can just make a simple circuit and then use
programming to achieve all the effects, and make changes as you desire
without even touching the wrong end of a slobbering iron.

I don't think a PICAXE is powerful enough to be able
to do a frequency analysis of an input sound signal...

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

It can be done with a PIC, but it is rather complex. I found this:
http://www.piclist.com/techref/microchip/fft/picspect.htm

Also check the color organ using a PIC16F84 on:
http://mondo-technology.com/

It is probably better to use analog bandpass filters to get several bands
of desired frequencies, rectified and filtered and measured by A/D
converters. Then use PWM outputs to drive the strings of LEDs. Actually,
the current regulated circuit I posted could be connected to the rectified
filter output and the LED intensity would be modulated without using a PIC.

I have made a simulation for a voltage-controlled LED boost driver that
works from about 2 V to 5 V and is modulated by a 100 mV RMS AC source
which could also be a DC level from a bandpass filter and rectifier with
proper voltage offset which would also be a reference brightness control.
The inverting modulation amplifier increases brightness to about 35 mA for
positive input levels and down to near zero. I added an LED and resistor as
a crude 3V voltage regulator to make the modulation amplifier less
sensitive to power supply voltage. It doubles as a power-on light.

I am modulating it with a 150 Hz sine wave, but a color organ would
modulate according to changes of sound volume in the bandpass, which may be
in the order of 1 Hz. The bandpass filter can be made from another
transistor and a few resistors and capacitors. A diode and capacitor can
produce a DC signal for modulation and it can be adjusted to provide a
fast-attack response to stretch out quick peaks of sound as desired.

Paul

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

Version 4
SHEET 1 1120 696
WIRE 64 -80 -144 -80
WIRE 208 -80 144 -80
WIRE 304 -80 208 -80
WIRE 384 -80 304 -80
WIRE 512 -80 384 -80
WIRE 304 -32 304 -80
WIRE -496 0 -512 0
WIRE -432 0 -496 0
WIRE -144 0 -144 -80
WIRE -144 0 -432 0
WIRE -32 0 -64 0
WIRE 16 0 -32 0
WIRE 32 0 16 0
WIRE 384 0 384 -80
WIRE 512 0 512 -80
WIRE 32 16 32 0
WIRE 208 16 208 -80
WIRE -432 32 -432 0
WIRE -144 32 -144 0
WIRE -64 32 -64 0
WIRE -512 112 -512 0
WIRE -32 112 -32 0
WIRE 32 112 32 80
WIRE -144 128 -144 112
WIRE -64 128 -64 96
WIRE -64 128 -144 128
WIRE -432 176 -432 112
WIRE -416 176 -432 176
WIRE -320 176 -336 176
WIRE -224 176 -240 176
WIRE -208 176 -224 176
WIRE -432 208 -432 176
WIRE 32 208 32 176
WIRE 384 224 384 80
WIRE 384 224 160 224
WIRE -224 240 -224 176
WIRE -176 240 -224 240
WIRE -144 256 -144 224
WIRE -32 256 -32 176
WIRE -32 256 -144 256
WIRE 384 256 384 224
WIRE 512 256 512 80
WIRE 592 256 512 256
WIRE 720 256 656 256
WIRE 512 304 512 256
WIRE 512 304 448 304
WIRE -512 320 -512 192
WIRE -432 320 -432 272
WIRE -432 320 -512 320
WIRE -272 320 -432 320
WIRE -224 320 -224 304
WIRE -224 320 -272 320
WIRE -144 320 -144 256
WIRE -144 320 -224 320
WIRE -64 320 -144 320
WIRE 32 320 32 272
WIRE 32 320 16 320
WIRE -176 352 -176 240
WIRE -64 352 -64 320
WIRE -96 400 -112 400
WIRE -64 400 -96 400
WIRE 32 400 32 320
WIRE 32 400 16 400
WIRE 384 400 384 352
WIRE 512 400 512 304
WIRE 720 400 720 256
WIRE -272 448 -272 320
WIRE -176 448 -272 448
WIRE -96 496 -96 400
WIRE -64 496 -96 496
WIRE 160 496 160 224
WIRE 160 496 16 496
WIRE -272 656 -272 448
WIRE 208 656 208 80
WIRE 208 656 -272 656
WIRE 304 656 304 32
WIRE 304 656 208 656
WIRE 384 656 384 480
WIRE 384 656 304 656
WIRE 512 656 512 480
WIRE 512 656 384 656
WIRE 720 656 720 480
WIRE 720 656 512 656
FLAG -64 352 0
FLAG 16 0 out
FLAG -496 0 in
SYMBOL npn -208 128 R0
SYMATTR InstName Q1
SYMATTR Value 2N3904
SYMBOL res -448 16 R0
SYMATTR InstName R1
SYMATTR Value 5k
SYMBOL cap -448 208 R0
SYMATTR InstName C1
SYMATTR Value 10n
SYMBOL ind2 -128 128 R180
WINDOW 0 36 80 Left 0
WINDOW 3 36 40 Left 0
SYMATTR InstName L1
SYMATTR Value 100ľ
SYMATTR Type ind
SYMBOL ind2 -336 192 R270
WINDOW 0 32 56 VTop 0
WINDOW 3 5 56 VBottom 0
SYMATTR InstName L2
SYMATTR Value 100ľ
SYMATTR Type ind
SYMBOL LED 16 112 R0
SYMATTR InstName D1
SYMATTR Value AOT-2015
SYMATTR Description Diode
SYMATTR Type diode
SYMBOL LED 16 208 R0
SYMATTR InstName D2
SYMATTR Value AOT-2015
SYMATTR Description Diode
SYMATTR Type diode
SYMBOL LED 16 16 R0
WINDOW 0 24 1 Left 0
SYMATTR InstName D3
SYMATTR Value AOT-2015
SYMATTR Description Diode
SYMATTR Type diode
SYMBOL voltage -512 96 R0
WINDOW 123 0 0 Left 0
WINDOW 39 0 0 Left 0
SYMATTR InstName V1
SYMATTR Value 3
SYMBOL res -432 192 R270
WINDOW 0 32 56 VTop 0
WINDOW 3 0 56 VBottom 0
SYMATTR InstName R2
SYMATTR Value 1k
SYMBOL schottky -208 304 R180
WINDOW 0 24 72 Left 0
WINDOW 3 18 5 Left 0
SYMATTR InstName D4
SYMATTR Value 1N5818
SYMATTR Description Diode
SYMATTR Type diode
SYMBOL schottky -48 96 R180
WINDOW 0 32 68 Left 0
WINDOW 3 16 -2 Left 0
SYMATTR InstName D5
SYMATTR Value 1N5818
SYMATTR Description Diode
SYMATTR Type diode
SYMBOL cap -48 112 R0
SYMATTR InstName C2
SYMATTR Value 5ľ
SYMBOL npn -112 352 M0
SYMATTR InstName Q2
SYMATTR Value 2N3904
SYMBOL res -80 336 R270
WINDOW 0 32 56 VTop 0
WINDOW 3 5 63 VBottom 0
SYMATTR InstName R3
SYMATTR Value 20
SYMBOL res -80 416 R270
WINDOW 0 32 56 VTop 0
WINDOW 3 0 56 VBottom 0
SYMATTR InstName R4
SYMATTR Value 500
SYMBOL res -80 512 R270
WINDOW 0 32 56 VTop 0
WINDOW 3 0 56 VBottom 0
SYMATTR InstName R5
SYMATTR Value 500
SYMBOL voltage 720 384 R0
WINDOW 123 24 132 Left 0
WINDOW 39 0 0 Left 0
SYMATTR InstName V2
SYMATTR Value SINE(0 .15 150 .001 0 0 1000)
SYMBOL npn 448 256 M0
WINDOW 0 11 -55 Left 0
WINDOW 3 -34 -15 Left 0
SYMATTR InstName Q3
SYMATTR Value 2N3904
SYMBOL res 368 384 R0
SYMATTR InstName R6
SYMATTR Value 100
SYMBOL res 368 -16 R0
SYMATTR InstName R7
SYMATTR Value 500
SYMBOL res 496 -16 R0
SYMATTR InstName R8
SYMATTR Value 15k
SYMBOL res 496 384 R0
SYMATTR InstName R9
SYMATTR Value 10k
SYMBOL cap 656 240 R90
WINDOW 0 0 32 VBottom 0
WINDOW 3 32 32 VTop 0
SYMATTR InstName C3
SYMATTR Value 1ľ
SYMBOL LED 192 16 R0
SYMATTR InstName D6
SYMATTR Value NSPW500BS
SYMATTR Description Diode
SYMATTR Type diode
SYMBOL res 48 -64 R270
WINDOW 0 32 56 VTop 0
WINDOW 3 0 56 VBottom 0
SYMATTR InstName R10
SYMATTR Value 100
SYMBOL cap 288 -32 R0
SYMATTR InstName C4
SYMATTR Value 10ľ
TEXT -352 64 Left 0 !K1 L1 L2 1
TEXT -544 344 Left 0 !.tran 30m startup

 

[fungus]

On Aug 12, 10:43 pm, "Paul E. Schoen" <p...@peschoen.com> wrote:

  I don't think a PICAXE is powerful enough to be able
  to do a frequency analysis of an input sound signal...


It can be done with a PIC, but it is rather complex. I found this...

That's done in PIC assembly language, not PICAXE basic.
Assembly language is a lot more work/learning and you'd
need a full PIC programmer to do it.

It is probably better to use analog bandpass filters to get several bands
of desired frequencies, rectified and filtered and measured by A/D
converters.

That's my next job but I'm a bit busy this week and
haven't had any time to research it. I'd be happy
with just low/high note detection.

I want to put LEDs in a drum like this one:

http://www.artesanum.com/foto-instrumentos_egipcios-5921-0.html

The idea I have in my head is four red LEDs
which are permanently on, then:

a) If you hit the drum in the middle (low note)
turn on two extra blue LEDs.

b) If you hit the drum near the edge (high note)
turn on an extra green one.

I already made a setup where I connected five
LEDs to the five lowest pins of my LM3914 and
two LEDs to the highest pins (leaving a safety
gap in the middle).

I connected my microphone to the sensor input
with a couple of resistors (one to add DC offset
and one to match the microphone impedance) and
the top two resistors responded to sound quite
nicely. That's as far as I got that day -just a
quick experiment and not bad for just the chip,
three resistors and a microphone.

For high note/low note detection I'm thinking
the chip's sensor could be part of the low note
detector and I'd need an external transistor to
switch one of the low LEDs for the high note
detector.

 

[Paul E. Schoen]

"fungus" <openglMYSOCKS@artlum.com> wrote in message
news:f329054c-90a2-4e47-84a9-a61d1441e7f7@k6g2000yqn.googlegroups.com...
On Aug 12, 10:43 pm, "Paul E. Schoen" <p...@peschoen.com> wrote:

I don't think a PICAXE is powerful enough to be able
to do a frequency analysis of an input sound signal...


It can be done with a PIC, but it is rather complex. I found this...

That's done in PIC assembly language, not PICAXE basic.
Assembly language is a lot more work/learning and you'd
need a full PIC programmer to do it.

Actually the PICkit and MPLAB will program the PIC, but it is in assembly.
There are free C compilers for PICs, and possibly also BASIC.

It is probably better to use analog bandpass filters to get several bands
of desired frequencies, rectified and filtered and measured by A/D
converters.

That's my next job but I'm a bit busy this week and
haven't had any time to research it. I'd be happy
with just low/high note detection.

I want to put LEDs in a drum like this one:

http://www.artesanum.com/foto-instrumentos_egipcios-5921-0.html

The idea I have in my head is four red LEDs
which are permanently on, then:

a) If you hit the drum in the middle (low note)
turn on two extra blue LEDs.

b) If you hit the drum near the edge (high note)
turn on an extra green one.

I already made a setup where I connected five
LEDs to the five lowest pins of my LM3914 and
two LEDs to the highest pins (leaving a safety
gap in the middle).

I connected my microphone to the sensor input
with a couple of resistors (one to add DC offset
and one to match the microphone impedance) and
the top two resistors responded to sound quite
nicely. That's as far as I got that day -just a
quick experiment and not bad for just the chip,
three resistors and a microphone.

For high note/low note detection I'm thinking
the chip's sensor could be part of the low note
detector and I'd need an external transistor to
switch one of the low LEDs for the high note
detector.

The circuit below is a simple bandpass filter at about 500 Hz. Its response
is mostly between 200 Hz and 2 kHz with about 1 or 2 volts applied. You can
adjust C1 and C2 for different frequency bands. Without rectification the
LED will be modulated at the applied frequency which might give an
interesting visual effect for low notes.

Paul

--------------------------------------------------
Version 4
SHEET 1 880 680
WIRE 160 48 -160 48
WIRE 288 48 160 48
WIRE 288 64 288 48
WIRE 160 80 160 48
WIRE -160 160 -160 48
WIRE 288 160 288 128
WIRE 352 160 288 160
WIRE 368 160 352 160
WIRE -48 208 -64 208
WIRE -32 208 -48 208
WIRE 128 208 112 208
WIRE 160 208 160 160
WIRE 160 208 128 208
WIRE 192 208 160 208
WIRE 224 208 192 208
WIRE -64 256 -64 208
WIRE 128 256 128 208
WIRE 192 256 192 208
WIRE 288 272 288 256
WIRE -160 368 -160 240
WIRE -64 368 -64 336
WIRE -64 368 -160 368
WIRE 128 368 128 320
WIRE 128 368 -64 368
WIRE 192 368 192 336
WIRE 192 368 128 368
WIRE 288 368 288 352
WIRE 288 368 192 368
WIRE -160 384 -160 368
FLAG -160 384 0
FLAG -48 208 in
FLAG 352 160 out
SYMBOL npn 224 160 R0
SYMATTR InstName Q1
SYMATTR Value 2N3904
SYMBOL LED 272 64 R0
WINDOW 0 28 3 Left 0
WINDOW 3 78 56 Left 0
SYMATTR InstName D1
SYMATTR Value QTLP690C
SYMATTR Description Diode
SYMATTR Type diode
SYMBOL res 272 256 R0
SYMATTR InstName R1
SYMATTR Value 10
SYMBOL res 176 240 R0
SYMATTR InstName R2
SYMATTR Value 5k
SYMBOL cap 32 192 R90
WINDOW 0 0 32 VBottom 0
WINDOW 3 32 32 VTop 0
SYMATTR InstName C2
SYMATTR Value 100n
SYMBOL Misc\\battery -160 144 R0
WINDOW 123 0 0 Left 0
WINDOW 39 0 0 Left 0
SYMATTR InstName V1
SYMATTR Value 3
SYMBOL voltage -64 240 R0
WINDOW 0 34 36 Left 0
WINDOW 3 -74 150 Left 0
WINDOW 123 14 97 Left 0
WINDOW 39 0 0 Left 0
SYMATTR InstName V2
SYMATTR Value SINE(0 2 600 0 0 0 100)
SYMATTR Value2 AC 2 0
SYMBOL res 144 64 R0
SYMATTR InstName R3
SYMATTR Value 20k
SYMBOL cap 144 320 R180
WINDOW 0 24 64 Left 0
WINDOW 3 24 8 Left 0
SYMATTR InstName C1
SYMATTR Value 50n
SYMBOL res 128 192 R90
WINDOW 0 0 56 VBottom 0
WINDOW 3 32 56 VTop 0
SYMATTR InstName R4
SYMATTR Value 5k
TEXT 174 392 Left 0 !;tran .05 startup
TEXT -48 32 Left 0 !.ac oct 5 10 20k