U.S. Pat 6,366,028 LED Flashlight Circuit

  • Thread starter Watson A.Name - "Watt Sun
  • Start date
In article <seMyb.275196$ao4.946277@attbi_s51>,
postmaster@BulkingPro.com mentioned...
"Mjolinor" <mjolinor@hotmail.com> wrote in message
news:RVvyb.1411$Lv1.32@newsfep3-gui.server.ntli.net...

"Watson A.Name - Watt Sun, Dark Remover" <alondra101@hotmail.com> wrote in
message news:MPG.1a340519c30de629899ab@news.dslextreme.com...
I built the circuit in the above patent (www.uspto.gov) that CMG uses
in their Infinity flashlight. Essentially it's an asymmetrical
astable multivibrator F-F, with one collector load being an inductor.
I used a 100 uH instead of the 220 uH that the schem calls for. The
transistors weren't given, so I used BC337-25 for both.

My problem is that it starts to oscillate a .65V, and if I go over
1.14VDC, it quits. The freq is 822(!) kHz, very high for these
circuits, but the patent says it's supposed to be about 500 kHz. I'm
gonna try a larger inductance to see what happens. I'm using the
values for the two caps of 1000 pF and 330 pF.

Right now, I can't even get it to work on a single AA Ni-MH cell,
because the voltage is too high. :-( I'm gonna increase some of the
resistor values to see if it helps and slows it down to below the AM
BCB. It comes in loud and clear on the radio.


http://www.elektor-electronics.co.uk/miniproj/frmipro.htm

june 2002, LED torch

That works well on one AA battery with white LEDS, you can pretty much get
it to drive whatever you want from whatever source you want and I think it
does the job as well as any I've seen.

Oddly enough, the big blue LEDs I have seem to have lots of capacitance.
Thus, if you put a diode in series with the LED, the efficiency seems to go
way up.

Here is the circuit I'm using; I built it, and its working quite well...
According to simulations, the efficiency is better than 50%, and the current
spikes reach about 100mA...

http://home.comcast.net/~rcmonsen/misc/ledtorch.jpg

Should I patent it? I mean, the part about using the extra diode? :)
Click to expand...
I'm puzzled by how a 1N4148 could help. This is a signal diode, not
made for high current. So high current pulses should only make it a
higher voltage drop, thus wasting more power. I would use a 1N5817 or
other schottky diode that has low V drop at high current. One thing
you might try is to parallel two 1N4148s to see if it helps reduce
losses. If it does, then you should use a higher conductance, lower V
drop diode.

As for patenting, you have to spend a sizable chunk of change to do
it, and there's a very good chance that the method is already prior
art, so not patentable. Or possibly it's already patented. And
before you do try, you should thoroughly investgate the circuit to
make sure it's really doing what you claim.


The original circuit abuses the transistors as well by generating large
negative spikes. This one does somewhat better. Maybe it was the capacitance
of the LED, although I don't really understand the effect.

Regards,
Bob Monsen
Click to expand...

--
@@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 <at> 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@@
 
"Watson A.Name - Watt Sun, Dark Remover" wrote:
In article <3FCA933B.44C8B884@bellatlantic.net>,
ehsjr@bellatlantic.net mentioned...


"Watson A.Name - Watt Sun, Dark Remover" wrote:

I built the circuit in the above patent (www.uspto.gov) that CMG uses
in their Infinity flashlight. Essentially it's an asymmetrical
astable multivibrator F-F, with one collector load being an inductor.
I used a 100 uH instead of the 220 uH that the schem calls for. The
transistors weren't given, so I used BC337-25 for both.

My problem is that it starts to oscillate a .65V, and if I go over
1.14VDC, it quits. The freq is 822(!) kHz, very high for these
circuits, but the patent says it's supposed to be about 500 kHz. I'm
gonna try a larger inductance to see what happens. I'm using the
values for the two caps of 1000 pF and 330 pF.

Right now, I can't even get it to work on a single AA Ni-MH cell,
because the voltage is too high. :-( I'm gonna increase some of the
resistor values to see if it helps and slows it down to below the AM
BCB. It comes in loud and clear on the radio.

I get "No such Patent Number found, please check the number
and try again" at www.uspto.gov when searching on 6366028

Well, I'm holding the original printout and that's the right number.
Did you do the right things, like including the commas?
Click to expand...
It won't work on this old browser (Netscrape 4.75).
I used my other PC with IE and it worked fine. Thanks.
 
In article <seMyb.275196$ao4.946277@attbi_s51>,
postmaster@BulkingPro.com mentioned...
"Mjolinor" <mjolinor@hotmail.com> wrote in message
news:RVvyb.1411$Lv1.32@newsfep3-gui.server.ntli.net...

"Watson A.Name - Watt Sun, Dark Remover" <alondra101@hotmail.com> wrote in
message news:MPG.1a340519c30de629899ab@news.dslextreme.com...
I built the circuit in the above patent (www.uspto.gov) that CMG uses
in their Infinity flashlight. Essentially it's an asymmetrical
astable multivibrator F-F, with one collector load being an inductor.
I used a 100 uH instead of the 220 uH that the schem calls for. The
transistors weren't given, so I used BC337-25 for both.

My problem is that it starts to oscillate a .65V, and if I go over
1.14VDC, it quits. The freq is 822(!) kHz, very high for these
circuits, but the patent says it's supposed to be about 500 kHz. I'm
gonna try a larger inductance to see what happens. I'm using the
values for the two caps of 1000 pF and 330 pF.

Right now, I can't even get it to work on a single AA Ni-MH cell,
because the voltage is too high. :-( I'm gonna increase some of the
resistor values to see if it helps and slows it down to below the AM
BCB. It comes in loud and clear on the radio.


http://www.elektor-electronics.co.uk/miniproj/frmipro.htm

june 2002, LED torch

That works well on one AA battery with white LEDS, you can pretty much get
it to drive whatever you want from whatever source you want and I think it
does the job as well as any I've seen.

Oddly enough, the big blue LEDs I have seem to have lots of capacitance.
Thus, if you put a diode in series with the LED, the efficiency seems to go
way up.

Here is the circuit I'm using; I built it, and its working quite well...
According to simulations, the efficiency is better than 50%, and the current
spikes reach about 100mA...

http://home.comcast.net/~rcmonsen/misc/ledtorch.jpg
Click to expand...
I built your circuit, with three minor diffs. I used a 47 uH
inductor, which shouldn't make much diff. I used a 510 pF cap, not
much diff. Also, I included a 0.1 uF byp cap across the batt. I was
also gonna put a 47 uF across the batt, but I didn't get to that yet.

With one white LED (which is really a blue LED) instead, I get 47 mA
at 1.5VDC. When I short across t he 1N4148, the current jumps up to
57 mA, and the LED gets appreciably brighter.

The circuit is drawing 70 to 85 mW total, which is less than the LED
itself can handle. Figuring an efficiency of 67%, the total current
for one LED should be about 160 to 180 mA. So it needs to have more
power out to the LED to get it to its full brightness. The circuit
needs lower Vce(sat) transistor for Q2. I'll try that later. The
freq measured 187 kHz, which is okay. Most of my other circuits were
under 100 kHz. I could get your circuit to toggle on and off at below
0.8Vdc, which is good and stable, no funnyness like the one I
discussed earlier in this thread. I can crank it up to 2VDC, and it
sucks over 100 mA, which is good, but the reason why it has to have a
higher V is because the 2N4401 doesn't have low enough Vce)(sat) at
high currents. I think your circuit would be better when operated
from 2 AA cells, with a change in the resistors to keep the current
down to maximum 30 mA to the LED.

The circuit I used quite a bit was the one on the left at
http://www.belza.cz/ledlight/ledm.htm. I also used the one on the
right, with the schottky and cap. With a NTE11 for the output tran, I
can easily drive three LEDs and suck a half amp from the batt. Now,
_that's_ a light! </Croc Dundee> ;-)

Back to experimenting..

I put another 1k across R3 1k, and the total current jumped up to 80
mA, and the LED got substantially brighter. So this is a good way to
get the Q2 to put out more. I shorted across the diode and the
total current is almost 100 mA, which makes the LED put out a really
bright beam. This is about where the LED should be run at, where it's
putting out some serious light. ;-)

Thanks for the well-behaved circuit.

[snip]
Regards,
Bob Monsen
Click to expand...

--
@@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 <at> 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 <ijRyb.275226$9E1.1431278@attbi_s52>,
postmaster@BulkingPro.com mentioned...
[snip]
: Regards,
: Bob Monsen

Thank you Bob. I did discover that they were chokes/coils,
as I had suspected, but your explanation of their construction
and how they are formulated made it very clear. I now know
what to call them and how to find out what their ratings are
when I reverse-engineer a circuit. Incidentally, is there an easy
method for taking a reading on them to find out their value?

Yes, you can buy inductance meters. I've heard a rumor that this is a good
one:

http://www.aade.com/lcm2binst/LC2Binst.htm

If you have access to a signal generator, you can test them directly by
observing the resonance frequency of a circuit with an inductor and a
capacitor of known value, using the equation

L = 1/((2*PI*f)^2 * C)

There are also markings on them to tell you what their inductance value is.
They use the same color scheme as resistors, except that the value depicted
is in micro-henries (uH) rather than ohms. So, a 33uH inductor will be
orange orange black, and then a color band for precision. There are other
schemes for marking as well, see a book such as "Practical Electronics for
Inventors" by Sherz at your library for more details on inductor markings.
There are also websites that have this information.

Regards,
Bob Monsen
Click to expand...
I have the above meter, and I think it's a good buy - a hundred
dollars for the kit. Or add $30 and get it assembled.

There are as many different looking inductors as there are capacitors.
Some are obviously coils of wire on a core. Others look like blobs of
dipped plastic, AKA gumdrops, with three or four colored dots on the
top. Others have the values printed in letters, which can be cryptic
until you begin to understand it. Like he said, some look like
resistors, and they can be tricky because they may be wound with very
fine wire, which gives them a certain resistance if you measure them
with a DMM.

Some radiofrequency coils are just wire, which was wound on a dowel
and then the dowel was removed. Sometimes the maker adjusted the
coils at the factory by spreading the turns, so it looks like some kid
got in there and messed things up. Or the coil may have wax poured
over it to keep it from changing value.

And then there's _the_ ultimate coil, the TC, short for Tesla Coil.
Don't even bring that topic up, you get those fanatics started and
they'll never end! ;-) See URLs
http://www.stargazing.net/Astroman/TCPhotos.html
http://www.rpi.edu/~cleara/hv/tesla8pics.htm


--
@@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 <at> 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@@
 
"Watson A.Name - Watt Sun, Dark Remover" <alondra101@hotmail.com> wrote in
message news:MPG.1a359ed7e00ede19899bb@news.dslextreme.com...
In article <seMyb.275196$ao4.946277@attbi_s51>,
postmaster@BulkingPro.com mentioned...
"Mjolinor" <mjolinor@hotmail.com> wrote in message
news:RVvyb.1411$Lv1.32@newsfep3-gui.server.ntli.net...

"Watson A.Name - Watt Sun, Dark Remover" <alondra101@hotmail.com
wrote in
message news:MPG.1a340519c30de629899ab@news.dslextreme.com...
I built the circuit in the above patent (www.uspto.gov) that CMG
uses
in their Infinity flashlight. Essentially it's an asymmetrical
astable multivibrator F-F, with one collector load being an
inductor.
I used a 100 uH instead of the 220 uH that the schem calls for. The
transistors weren't given, so I used BC337-25 for both.

My problem is that it starts to oscillate a .65V, and if I go over
1.14VDC, it quits. The freq is 822(!) kHz, very high for these
circuits, but the patent says it's supposed to be about 500 kHz.
I'm
gonna try a larger inductance to see what happens. I'm using the
values for the two caps of 1000 pF and 330 pF.

Right now, I can't even get it to work on a single AA Ni-MH cell,
because the voltage is too high. :-( I'm gonna increase some of
the
resistor values to see if it helps and slows it down to below the AM
BCB. It comes in loud and clear on the radio.


http://www.elektor-electronics.co.uk/miniproj/frmipro.htm

june 2002, LED torch

That works well on one AA battery with white LEDS, you can pretty much
get
it to drive whatever you want from whatever source you want and I
think it
does the job as well as any I've seen.

Oddly enough, the big blue LEDs I have seem to have lots of capacitance.
Thus, if you put a diode in series with the LED, the efficiency seems to
go
way up.

Here is the circuit I'm using; I built it, and its working quite well...
According to simulations, the efficiency is better than 50%, and the
current
spikes reach about 100mA...

http://home.comcast.net/~rcmonsen/misc/ledtorch.jpg

Should I patent it? I mean, the part about using the extra diode? :)

I'm puzzled by how a 1N4148 could help. This is a signal diode, not
made for high current. So high current pulses should only make it a
higher voltage drop, thus wasting more power. I would use a 1N5817 or
other schottky diode that has low V drop at high current. One thing
you might try is to parallel two 1N4148s to see if it helps reduce
losses. If it does, then you should use a higher conductance, lower V
drop diode.
Click to expand...
I believe the thing that was happening with the LEDs that I have is that
there was some capacitance across the LED. When I put the diode in series,
it stopped behaving badly, and the efficiency of the simulated circuit
doubled.

Also, oddly enough, the voltage between the LED and the diode got 'pumped
up' similarly to what would happen if you had a cap/resistor in parallel in
place of the LED.

This may just be some artifact of the blue LEDs I'm using (and simulating
with.)

I didn't have any low vf diodes, so I couldn't try those. I haven't had any
luck.

As for patenting, you have to spend a sizable chunk of change to do
it, and there's a very good chance that the method is already prior
art, so not patentable. Or possibly it's already patented. And
before you do try, you should thoroughly investgate the circuit to
make sure it's really doing what you claim.
Click to expand...
The patent comment was a joke... I was making fun of the fact that somebody
patented the circuit you originally posted. :)

The original circuit abuses the transistors as well by generating large
negative spikes. This one does somewhat better. Maybe it was the
capacitance
of the LED, although I don't really understand the effect.

Regards,
Bob Monsen


--
@@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 <at> 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@@
Click to expand...
 
"Watson A.Name - Watt Sun, Dark Remover" <alondra101@hotmail.com> wrote in
message news:MPG.1a35da3c977f866e9899bc@news.dslextreme.com...
In article <seMyb.275196$ao4.946277@attbi_s51>,
postmaster@BulkingPro.com mentioned...
"Mjolinor" <mjolinor@hotmail.com> wrote in message
news:RVvyb.1411$Lv1.32@newsfep3-gui.server.ntli.net...

"Watson A.Name - Watt Sun, Dark Remover" <alondra101@hotmail.com
wrote in
message news:MPG.1a340519c30de629899ab@news.dslextreme.com...
I built the circuit in the above patent (www.uspto.gov) that CMG
uses
in their Infinity flashlight. Essentially it's an asymmetrical
astable multivibrator F-F, with one collector load being an
inductor.
I used a 100 uH instead of the 220 uH that the schem calls for. The
transistors weren't given, so I used BC337-25 for both.

My problem is that it starts to oscillate a .65V, and if I go over
1.14VDC, it quits. The freq is 822(!) kHz, very high for these
circuits, but the patent says it's supposed to be about 500 kHz.
I'm
gonna try a larger inductance to see what happens. I'm using the
values for the two caps of 1000 pF and 330 pF.

Right now, I can't even get it to work on a single AA Ni-MH cell,
because the voltage is too high. :-( I'm gonna increase some of
the
resistor values to see if it helps and slows it down to below the AM
BCB. It comes in loud and clear on the radio.


http://www.elektor-electronics.co.uk/miniproj/frmipro.htm

june 2002, LED torch

That works well on one AA battery with white LEDS, you can pretty much
get
it to drive whatever you want from whatever source you want and I
think it
does the job as well as any I've seen.

Oddly enough, the big blue LEDs I have seem to have lots of capacitance.
Thus, if you put a diode in series with the LED, the efficiency seems to
go
way up.

Here is the circuit I'm using; I built it, and its working quite well...
According to simulations, the efficiency is better than 50%, and the
current
spikes reach about 100mA...

http://home.comcast.net/~rcmonsen/misc/ledtorch.jpg

I built your circuit, with three minor diffs. I used a 47 uH
inductor, which shouldn't make much diff. I used a 510 pF cap, not
much diff. Also, I included a 0.1 uF byp cap across the batt. I was
also gonna put a 47 uF across the batt, but I didn't get to that yet.
Click to expand...
I found that the bypass cap didn't do anything at all. In the simulations,
there was no energy getting transferred through it. Thats why I dumped it.

With one white LED (which is really a blue LED) instead, I get 47 mA
at 1.5VDC. When I short across t he 1N4148, the current jumps up to
57 mA, and the LED gets appreciably brighter.
Click to expand...
Must be some difference with the LEDs. When I breadboarded it, the opposite
happened (ie, it got more efficient, and the LED brightened using the diode)

The circuit is drawing 70 to 85 mW total, which is less than the LED
itself can handle. Figuring an efficiency of 67%, the total current
for one LED should be about 160 to 180 mA. So it needs to have more
power out to the LED to get it to its full brightness. The circuit
needs lower Vce(sat) transistor for Q2. I'll try that later.
Click to expand...
Are you soldering or using a solderless breadboard? I find that I get
oscillations at a bit more than a MHz with the circuit as posted, at least
according to my oscilloscope.

The
freq measured 187 kHz, which is okay. Most of my other circuits were
under 100 kHz. I could get your circuit to toggle on and off at below
0.8Vdc, which is good and stable, no funnyness like the one I
discussed earlier in this thread. I can crank it up to 2VDC, and it
sucks over 100 mA, which is good, but the reason why it has to have a
higher V is because the 2N4401 doesn't have low enough Vce)(sat) at
high currents. I think your circuit would be better when operated
from 2 AA cells, with a change in the resistors to keep the current
down to maximum 30 mA to the LED.
Click to expand...
Unfortunately, if you increase the voltage to above the LED voltage, it
doesn't oscillate. The voltage is increased by using the additional diode,
so if you added in a couple of diodes, it might oscillate.

I don't have any low Vce sat transistors. The 2N4401 has a Vce(sat) of .75V
at 50mA.

Maybe I'll try to get it working with n-mosfets or jfets. That might improve
the efficiency as well.

The circuit I used quite a bit was the one on the left at
http://www.belza.cz/ledlight/ledm.htm. I also used the one on the
right, with the schottky and cap. With a NTE11 for the output tran, I
can easily drive three LEDs and suck a half amp from the batt. Now,
_that's_ a light! </Croc Dundee> ;-)

Back to experimenting..

I put another 1k across R3 1k, and the total current jumped up to 80
mA, and the LED got substantially brighter. So this is a good way to
get the Q2 to put out more. I shorted across the diode and the
total current is almost 100 mA, which makes the LED put out a really
bright beam. This is about where the LED should be run at, where it's
putting out some serious light. ;-)

Thanks for the well-behaved circuit.
Click to expand...
BTW, its not really 'my' circuit, except for the diode, which you apparently
aren't using. It is the one posted by mjolinor with a much smaller
inductor...

Regards,
Bob Monsen


--
@@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 <at> 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@@
Click to expand...
 
"Watson A.Name - Watt Sun, Dark Remover" <alondra101@hotmail.com>
wrote in message
news:MPG.1a35e2214b8268e39899bd@news.dslextreme.com...
: In article <ijRyb.275226$9E1.1431278@attbi_s52>,
: postmaster@BulkingPro.com mentioned...
: [snip]
: > > : Regards,
: > > : Bob Monsen
: > >
: > > Thank you Bob. I did discover that they were chokes/coils,
: > > as I had suspected, but your explanation of their
construction
: > > and how they are formulated made it very clear. I now know
: > > what to call them and how to find out what their ratings
are
: > > when I reverse-engineer a circuit. Incidentally, is there
an easy
: > > method for taking a reading on them to find out their
value?
: >
: > Yes, you can buy inductance meters. I've heard a rumor that
this is a good
: > one:
: >
: > http://www.aade.com/lcm2binst/LC2Binst.htm
: >
: > If you have access to a signal generator, you can test them
directly by
: > observing the resonance frequency of a circuit with an
inductor and a
: > capacitor of known value, using the equation
: >
: > L = 1/((2*PI*f)^2 * C)
: >
: > There are also markings on them to tell you what their
inductance value is.
: > They use the same color scheme as resistors, except that the
value depicted
: > is in micro-henries (uH) rather than ohms. So, a 33uH
inductor will be
: > orange orange black, and then a color band for precision.
There are other
: > schemes for marking as well, see a book such as "Practical
Electronics for
: > Inventors" by Sherz at your library for more details on
inductor markings.
: > There are also websites that have this information.
: >
: > Regards,
: > Bob Monsen
:
: I have the above meter, and I think it's a good buy - a hundred
: dollars for the kit. Or add $30 and get it assembled.

Aren't there any I can make for about $25?
I am such a cheap-skate. ;-)

: There are as many different looking inductors as there are
capacitors.
: Some are obviously coils of wire on a core. Others look like
blobs of
: dipped plastic, AKA gumdrops, with three or four colored dots
on the
: top. Others have the values printed in letters, which can be
cryptic
: until you begin to understand it. Like he said, some look like
: resistors, and they can be tricky because they may be wound
with very
: fine wire, which gives them a certain resistance if you measure
them
: with a DMM.

I had a few choke coils laying around that I had
taken from some Apple II disk drive analog controllers.
None of them gave any reading on the ohms scale.
They are around 18 gauge solid wrapped around some
kind of ring of unknown material painted yellow, with
shrink-tube over the whole thing. The wires terminated
at the same place on the ring but on opposite edges.
Only 5/8" diameter and 3/8" thick, but fairly heavy.

: Some radiofrequency coils are just wire, which was wound on a
dowel
: and then the dowel was removed. Sometimes the maker adjusted
the
: coils at the factory by spreading the turns, so it looks like
some kid
: got in there and messed things up. Or the coil may have wax
poured
: over it to keep it from changing value.

I've seen these and wondered if they were a mistake,
or placed there to trick people. So if the coils lose
their shape, it can affect the circuit operation?

: And then there's _the_ ultimate coil, the TC, short for Tesla
Coil.
: Don't even bring that topic up, you get those fanatics started
and
: they'll never end! ;-) See URLs
: http://www.stargazing.net/Astroman/TCPhotos.html
: http://www.rpi.edu/~cleara/hv/tesla8pics.htm

I won't say a word, except for this:
Why would anyone make that a serious hobby?, and,
I'd think as a junior high project it would have
been fun, once, maybe. Gee, hope I didn't stir up
anything. ;-)

Bill @ GarberStreet Enterprizez };-)
Web Site - http://garberstreet.netfirms.com
Email - willy4SPAM6pa@comXcast.net
Remove - SPAM and X to contact me




---
This email ain't infected, dude!

Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.544 / Virus Database: 338 - Release Date: 11/27/03
 
"Robert Monsen" <postmaster@BulkingPro.com> wrote in message
news:lGYyb.278133$9E1.1461852@attbi_s52...
"Watson A.Name - Watt Sun, Dark Remover" <alondra101@hotmail.com> wrote in
message news:MPG.1a35da3c977f866e9899bc@news.dslextreme.com...
In article <seMyb.275196$ao4.946277@attbi_s51>,
postmaster@BulkingPro.com mentioned...
"Mjolinor" <mjolinor@hotmail.com> wrote in message
news:RVvyb.1411$Lv1.32@newsfep3-gui.server.ntli.net...

"Watson A.Name - Watt Sun, Dark Remover" <alondra101@hotmail.com
wrote in
message news:MPG.1a340519c30de629899ab@news.dslextreme.com...
I built the circuit in the above patent (www.uspto.gov) that CMG
uses
in their Infinity flashlight. Essentially it's an asymmetrical
astable multivibrator F-F, with one collector load being an
inductor.
I used a 100 uH instead of the 220 uH that the schem calls for.
The
transistors weren't given, so I used BC337-25 for both.

My problem is that it starts to oscillate a .65V, and if I go over
1.14VDC, it quits. The freq is 822(!) kHz, very high for these
circuits, but the patent says it's supposed to be about 500 kHz.
I'm
gonna try a larger inductance to see what happens. I'm using the
values for the two caps of 1000 pF and 330 pF.

Right now, I can't even get it to work on a single AA Ni-MH cell,
because the voltage is too high. :-( I'm gonna increase some of
the
resistor values to see if it helps and slows it down to below the
AM
BCB. It comes in loud and clear on the radio.


http://www.elektor-electronics.co.uk/miniproj/frmipro.htm

june 2002, LED torch

That works well on one AA battery with white LEDS, you can pretty
much
get
it to drive whatever you want from whatever source you want and I
think it
does the job as well as any I've seen.

Oddly enough, the big blue LEDs I have seem to have lots of
capacitance.
Thus, if you put a diode in series with the LED, the efficiency seems
to
go
way up.

Here is the circuit I'm using; I built it, and its working quite
well...
According to simulations, the efficiency is better than 50%, and the
current
spikes reach about 100mA...

http://home.comcast.net/~rcmonsen/misc/ledtorch.jpg

I built your circuit, with three minor diffs. I used a 47 uH
inductor, which shouldn't make much diff. I used a 510 pF cap, not
much diff. Also, I included a 0.1 uF byp cap across the batt. I was
also gonna put a 47 uF across the batt, but I didn't get to that yet.


I found that the bypass cap didn't do anything at all. In the simulations,
there was no energy getting transferred through it. Thats why I dumped it.

With one white LED (which is really a blue LED) instead, I get 47 mA
at 1.5VDC. When I short across t he 1N4148, the current jumps up to
57 mA, and the LED gets appreciably brighter.

Must be some difference with the LEDs. When I breadboarded it, the
opposite
happened (ie, it got more efficient, and the LED brightened using the
diode)
Click to expand...

CORRECTION:

Actually, this is wrong. Its based on simulations.

When I put a wire across the diode, it does get a bit brighter, but the
current goes way up, not proportional to the amount of brightness increase,
so I guess its probably getting much less efficient, as in the simulations.

Again, in the simulations, the difference in efficiency with the diode was
remarkable, going from 20% to above 60%. I'm still not sure if this is an
artifact of the simulation. I guess I'll have to build a power meter again
to measure it.

The circuit is drawing 70 to 85 mW total, which is less than the LED
itself can handle. Figuring an efficiency of 67%, the total current
for one LED should be about 160 to 180 mA. So it needs to have more
power out to the LED to get it to its full brightness. The circuit
needs lower Vce(sat) transistor for Q2. I'll try that later.

Are you soldering or using a solderless breadboard? I find that I get
oscillations at a bit more than a MHz with the circuit as posted, at least
according to my oscilloscope.

The
freq measured 187 kHz, which is okay. Most of my other circuits were
under 100 kHz. I could get your circuit to toggle on and off at below
0.8Vdc, which is good and stable, no funnyness like the one I
discussed earlier in this thread. I can crank it up to 2VDC, and it
sucks over 100 mA, which is good, but the reason why it has to have a
higher V is because the 2N4401 doesn't have low enough Vce)(sat) at
high currents. I think your circuit would be better when operated
from 2 AA cells, with a change in the resistors to keep the current
down to maximum 30 mA to the LED.

Unfortunately, if you increase the voltage to above the LED voltage, it
doesn't oscillate. The voltage is increased by using the additional diode,
so if you added in a couple of diodes, it might oscillate.

I don't have any low Vce sat transistors. The 2N4401 has a Vce(sat) of
..75V
at 50mA.

Maybe I'll try to get it working with n-mosfets or jfets. That might
improve
the efficiency as well.


The circuit I used quite a bit was the one on the left at
http://www.belza.cz/ledlight/ledm.htm. I also used the one on the
right, with the schottky and cap. With a NTE11 for the output tran, I
can easily drive three LEDs and suck a half amp from the batt. Now,
_that's_ a light! </Croc Dundee> ;-)

Back to experimenting..

I put another 1k across R3 1k, and the total current jumped up to 80
mA, and the LED got substantially brighter. So this is a good way to
get the Q2 to put out more. I shorted across the diode and the
total current is almost 100 mA, which makes the LED put out a really
bright beam. This is about where the LED should be run at, where it's
putting out some serious light. ;-)

Thanks for the well-behaved circuit.


BTW, its not really 'my' circuit, except for the diode, which you
apparently
aren't using. It is the one posted by mjolinor with a much smaller
inductor...

[snip]
Regards,
Bob Monsen


--
@@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 <at> 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@@
Click to expand...
 
In article <uvYyb.280052$ao4.990457@attbi_s51>,
Robert Monsen <postmaster@BulkingPro.com> wrote:
I believe the thing that was happening with the LEDs that I have is that
there was some capacitance across the LED. When I put the diode in series,
it stopped behaving badly, and the efficiency of the simulated circuit
doubled.

Also, oddly enough, the voltage between the LED and the diode got 'pumped
up' similarly to what would happen if you had a cap/resistor in parallel in
place of the LED.

This may just be some artifact of the blue LEDs I'm using (and simulating
with.)
Click to expand...
I've not looked at your circuit, but it might be "reverse recovery time".
where a diode looks like a short circuit for a microsecond or so after
the applied voltage changes, until the charge carriers get swept out
of the diode junction (or something like that). It depends on the
size/construction of the diode and small signal diodes are supposed to
avoid or minimize it.

Or it could be the reverse breakdown voltage.

I don't trust a simulator get get either of these correct, off the shelf.

Mark Zenier mzenier@eskimo.com Washington State resident
 
In article <3FCC2ECF.C7104E5D@bellatlantic.net>,
ehsjr@bellatlantic.net mentioned...
"Watson A.Name - Watt Sun, Dark Remover" wrote:

In article <3FCA933B.44C8B884@bellatlantic.net>,
ehsjr@bellatlantic.net mentioned...


"Watson A.Name - Watt Sun, Dark Remover" wrote:

I built the circuit in the above patent (www.uspto.gov) that CMG uses
in their Infinity flashlight. Essentially it's an asymmetrical
astable multivibrator F-F, with one collector load being an inductor.
I used a 100 uH instead of the 220 uH that the schem calls for. The
transistors weren't given, so I used BC337-25 for both.

My problem is that it starts to oscillate a .65V, and if I go over
1.14VDC, it quits. The freq is 822(!) kHz, very high for these
circuits, but the patent says it's supposed to be about 500 kHz. I'm
gonna try a larger inductance to see what happens. I'm using the
values for the two caps of 1000 pF and 330 pF.

Right now, I can't even get it to work on a single AA Ni-MH cell,
because the voltage is too high. :-( I'm gonna increase some of the
resistor values to see if it helps and slows it down to below the AM
BCB. It comes in loud and clear on the radio.

I get "No such Patent Number found, please check the number
and try again" at www.uspto.gov when searching on 6366028

Well, I'm holding the original printout and that's the right number.
Did you do the right things, like including the commas?


It won't work on this old browser (Netscrape 4.75).
I used my other PC with IE and it worked fine. Thanks.
Click to expand...
I went from 4.75, 4.79, 4.8 and finally got smart and d/l'd Mozilla.
I was having trouble viewing some websites that used the newer html.
So get rid of it, and d/l at www.mozilla.org. It's free.


--
@@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 <at> 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@@
 
"Watson A.Name - Watt Sun, Dark Remover" <alondra101@hotmail.com> wrote in
message news:MPG.1a3785855eea89c89899c5@news.dslextreme.com...
In article <3FCC2ECF.C7104E5D@bellatlantic.net>,
ehsjr@bellatlantic.net mentioned...


"Watson A.Name - Watt Sun, Dark Remover" wrote:

In article <3FCA933B.44C8B884@bellatlantic.net>,
ehsjr@bellatlantic.net mentioned...


"Watson A.Name - Watt Sun, Dark Remover" wrote:

I built the circuit in the above patent (www.uspto.gov) that CMG
uses
in their Infinity flashlight. Essentially it's an asymmetrical
astable multivibrator F-F, with one collector load being an
inductor.
I used a 100 uH instead of the 220 uH that the schem calls for.
The
transistors weren't given, so I used BC337-25 for both.

My problem is that it starts to oscillate a .65V, and if I go over
1.14VDC, it quits. The freq is 822(!) kHz, very high for these
circuits, but the patent says it's supposed to be about 500 kHz.
I'm
gonna try a larger inductance to see what happens. I'm using the
values for the two caps of 1000 pF and 330 pF.

Right now, I can't even get it to work on a single AA Ni-MH cell,
because the voltage is too high. :-( I'm gonna increase some of
the
resistor values to see if it helps and slows it down to below the
AM
BCB. It comes in loud and clear on the radio.
Click to expand...

Found another one you may like to try, seems easy to get working, mind you
it should be with 1 transistor 1 resistor and 1 inductor:)

http://users.cableaz.com/~cappels/dproj/ledpage/leddrv.htm

Next problem: If I want to put more than 1 LED on is it a circuit each. I
have real difficulty actually putting figures on how much light there is
particularly if I have two LEDs lit. I can easily fasten one LED to an LDR
and get a comparative brightness for different voltage / current /
frequencies / inductors but, for example if I parallel two LEDs then the
current only goes up by a ma or so but both LEDs are obviously lit though
each LED is slightly dimmer than it was with only one. How do I tell if the
overall light output is more than it was with one LED?

Lots of variables and no solutions.
 
In article <MPG.1a35da3c977f866e9899bc@news.dslextreme.com>, Watson
A.Name - "Watt Sun, Dark Remover" <alondra101@hotmail.com>
mentioned...
[snip]

Here is the circuit I'm using; I built it, and its working quite well...
According to simulations, the efficiency is better than 50%, and the current
spikes reach about 100mA...

http://home.comcast.net/~rcmonsen/misc/ledtorch.jpg

I built your circuit, with three minor diffs. I used a 47 uH
inductor, which shouldn't make much diff. I used a 510 pF cap, not
much diff. Also, I included a 0.1 uF byp cap across the batt. I was
also gonna put a 47 uF across the batt, but I didn't get to that yet.

With one white LED (which is really a blue LED) instead, I get 47 mA
at 1.5VDC. When I short across t he 1N4148, the current jumps up to
57 mA, and the LED gets appreciably brighter.

The circuit is drawing 70 to 85 mW total, which is less than the LED
itself can handle. Figuring an efficiency of 67%, the total current
for one LED should be about 160 to 180 mA. So it needs to have more
power out to the LED to get it to its full brightness. The circuit
needs lower Vce(sat) transistor for Q2. I'll try that later. The
freq measured 187 kHz, which is okay. Most of my other circuits were
under 100 kHz. I could get your circuit to toggle on and off at below
0.8Vdc, which is good and stable, no funnyness like the one I
discussed earlier in this thread. I can crank it up to 2VDC, and it
sucks over 100 mA, which is good, but the reason why it has to have a
higher V is because the 2N4401 doesn't have low enough Vce)(sat) at
high currents. I think your circuit would be better when operated
from 2 AA cells, with a change in the resistors to keep the current
down to maximum 30 mA to the LED.

The circuit I used quite a bit was the one on the left at
http://www.belza.cz/ledlight/ledm.htm. I also used the one on the
right, with the schottky and cap. With a NTE11 for the output tran, I
can easily drive three LEDs and suck a half amp from the batt. Now,
_that's_ a light! </Croc Dundee> ;-)

Back to experimenting..

I put another 1k across R3 1k, and the total current jumped up to 80
mA, and the LED got substantially brighter. So this is a good way to
get the Q2 to put out more. I shorted across the diode and the
total current is almost 100 mA, which makes the LED put out a really
bright beam. This is about where the LED should be run at, where it's
putting out some serious light. ;-)

Thanks for the well-behaved circuit.
Click to expand...
I built a second one using BC337-25s for the two transistors, a 100 uH
coil, and a few other resistance changes. It's well-behaved down to
..75VDC, and puts out mucho light at 1.5VDC, using 105 mA total. The
freq is a bit over 100 kHz. This is a good, decent circuit for
driving a single LED. With surface mount parts, each LED could have
its own circuit, and any number of LEDs could be added. I haven't
seen a surface mount transistor equivalent to the NTE11, which can
handle 5 Amps. The BC817 will do 2A, so it's not a bad choice. With
the NTE11 or 2SD965 the circuit might be able to drive three LEDs in
parallel.



Regards,
Bob Monsen
Click to expand...
--
@@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 <at> 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@@
 
"Mark Zenier" <mzenier@eskimo.com> wrote in message
news:bql70d$c8p$1@eskinews.eskimo.com...
In article <uvYyb.280052$ao4.990457@attbi_s51>,
Robert Monsen <postmaster@BulkingPro.com> wrote:
I believe the thing that was happening with the LEDs that I have is that
there was some capacitance across the LED. When I put the diode in
series,
it stopped behaving badly, and the efficiency of the simulated circuit
doubled.

Also, oddly enough, the voltage between the LED and the diode got 'pumped
up' similarly to what would happen if you had a cap/resistor in parallel
in
place of the LED.

This may just be some artifact of the blue LEDs I'm using (and simulating
with.)

I've not looked at your circuit, but it might be "reverse recovery time".
where a diode looks like a short circuit for a microsecond or so after
the applied voltage changes, until the charge carriers get swept out
of the diode junction (or something like that). It depends on the
size/construction of the diode and small signal diodes are supposed to
avoid or minimize it.

Or it could be the reverse breakdown voltage.

I don't trust a simulator get get either of these correct, off the shelf.

Mark Zenier mzenier@eskimo.com Washington State resident
Click to expand...
Out of curiosity, I built a shunt around the diode controlled by a
pushbutton, and scoped the junction at the inductor. With the diode, the
scope looks just like the simulation. Without the diode, the scope looks
completely different. So I'm guessing that, as you said, the LED simulation
is simply wrong. On the scope, the only real difference is that the peak is
higher with the diode, which makes sense, since the inductor has to bring up
the voltage to cross two diode junctions instead of one.

Thanks for looking at this.

Regards,
Bob Monsen
 
In article <lGYyb.278133$9E1.1461852@attbi_s52>,
postmaster@BulkingPro.com mentioned...

[snip]

http://home.comcast.net/~rcmonsen/misc/ledtorch.jpg
[snip]
Click to expand...

Are you soldering or using a solderless breadboard? I find that I get
oscillations at a bit more than a MHz with the circuit as posted, at least
according to my oscilloscope.
Click to expand...
All the ones I'm building are soldered on a piece of predrilled board.
Size is .8" by 1".

The
freq measured 187 kHz, which is okay. Most of my other circuits were
under 100 kHz. I could get your circuit to toggle on and off at below
0.8Vdc, which is good and stable, no funnyness like the one I
discussed earlier in this thread. I can crank it up to 2VDC, and it
sucks over 100 mA, which is good, but the reason why it has to have a
higher V is because the 2N4401 doesn't have low enough Vce)(sat) at
high currents. I think your circuit would be better when operated
from 2 AA cells, with a change in the resistors to keep the current
down to maximum 30 mA to the LED.

Unfortunately, if you increase the voltage to above the LED voltage, it
doesn't oscillate. The voltage is increased by using the additional diode,
so if you added in a couple of diodes, it might oscillate.
Click to expand...
The white LEDs are always over 3V. Maybe I should try two LEDs in
series. I did that on another one, with a red, green and yellow.

I don't have any low Vce sat transistors. The 2N4401 has a Vce(sat) of .75V
at 50mA.
Click to expand...
Well, according to the Fairchild data sheet, the Vce(sat) is .75V at
500 mA Ic, 50 mA Ib. But both of the values given are maximum, and
those values are usually well over the typical values. I'd guess a
third of that would be more typical.

But the 2N4401s are not as capable as some other really inexpensive
transistors like the BC337 (PNP complement is BC327), or the BC639,
which will do .8A and 1A. What does the job best is a transistor that
has good beta holdup at high currents, so it will saturate with less
base currrent, which wastes less total current.

And the best one for this is the 2SD965 or NTE11, which are good for 5
Amps.

Maybe I'll try to get it working with n-mosfets or jfets. That might improve
the efficiency as well.
Click to expand...
Well, the problem is that you need 3 or more Volts to turn on an enh.
mode FET, and you only have 1.5VDC. So you would have to make a
starter circuit like the one here.
http://www.ece.uvic.ca/~jbornema/Journals/064a-97ia-jmd.pdf
This is fine for some circuit that's more complicated and expensive,
but is too complicated for a flashlight. This circuit has a dozen or
less parts, so it's just fine for the job. Some designs use a current
sensor to keep from overdriving the LED, but that's not really
necessary. See the second pic at URL
http://www.geocities.co.jp/Technopolis-Mars/2881/LED02.html

The circuit I used quite a bit was the one on the left at
http://www.belza.cz/ledlight/ledm.htm. I also used the one on the
right, with the schottky and cap. With a NTE11 for the output tran, I
can easily drive three LEDs and suck a half amp from the batt. Now,
_that's_ a light! </Croc Dundee> ;-)

Back to experimenting..

I put another 1k across R3 1k, and the total current jumped up to 80
mA, and the LED got substantially brighter. So this is a good way to
get the Q2 to put out more. I shorted across the diode and the
total current is almost 100 mA, which makes the LED put out a really
bright beam. This is about where the LED should be run at, where it's
putting out some serious light. ;-)

Thanks for the well-behaved circuit.


BTW, its not really 'my' circuit, except for the diode, which you apparently
aren't using. It is the one posted by mjolinor with a much smaller
inductor...
Click to expand...
I've built a lot of the circuits that have been posted, and I don't
remember seeing it before. But great, it's a viable circuit design,
without any funnyness. I think I'd try to get it to drive three LEDs
with one of the above transistors, the 2SD965, and see how efficient
it is.

I bought a bunch of 1N5817s from Mouser, but I'd like to get a bunch
of the schottky used in that last URL, the 11EQS04, which is half the
size of the 1N5817, and rated for a half amp.

Regards,
Bob Monsen
Click to expand...

--
@@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 <at> 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 <4k5zb.401386$Fm2.410986@attbi_s04>,
postmaster@BulkingPro.com mentioned...
"Robert Monsen" <postmaster@BulkingPro.com> wrote in message
news:lGYyb.278133$9E1.1461852@attbi_s52...

"Watson A.Name - Watt Sun, Dark Remover" <alondra101@hotmail.com> wrote in
message news:MPG.1a35da3c977f866e9899bc@news.dslextreme.com...
In article <seMyb.275196$ao4.946277@attbi_s51>,
postmaster@BulkingPro.com mentioned...
"Mjolinor" <mjolinor@hotmail.com> wrote in message
news:RVvyb.1411$Lv1.32@newsfep3-gui.server.ntli.net...

"Watson A.Name - Watt Sun, Dark Remover" <alondra101@hotmail.com
wrote in
message news:MPG.1a340519c30de629899ab@news.dslextreme.com...
I built the circuit in the above patent (www.uspto.gov) that CMG
uses
in their Infinity flashlight. Essentially it's an asymmetrical
astable multivibrator F-F, with one collector load being an
inductor.
I used a 100 uH instead of the 220 uH that the schem calls for.
The
transistors weren't given, so I used BC337-25 for both.

My problem is that it starts to oscillate a .65V, and if I go over
1.14VDC, it quits. The freq is 822(!) kHz, very high for these
circuits, but the patent says it's supposed to be about 500 kHz.
I'm
gonna try a larger inductance to see what happens. I'm using the
values for the two caps of 1000 pF and 330 pF.

Right now, I can't even get it to work on a single AA Ni-MH cell,
because the voltage is too high. :-( I'm gonna increase some of
the
resistor values to see if it helps and slows it down to below the
AM
BCB. It comes in loud and clear on the radio.


http://www.elektor-electronics.co.uk/miniproj/frmipro.htm

june 2002, LED torch

That works well on one AA battery with white LEDS, you can pretty
much
get
it to drive whatever you want from whatever source you want and I
think it
does the job as well as any I've seen.

Oddly enough, the big blue LEDs I have seem to have lots of
capacitance.
Thus, if you put a diode in series with the LED, the efficiency seems
to
go
way up.

Here is the circuit I'm using; I built it, and its working quite
well...
According to simulations, the efficiency is better than 50%, and the
current
spikes reach about 100mA...

http://home.comcast.net/~rcmonsen/misc/ledtorch.jpg

I built your circuit, with three minor diffs. I used a 47 uH
inductor, which shouldn't make much diff. I used a 510 pF cap, not
much diff. Also, I included a 0.1 uF byp cap across the batt. I was
also gonna put a 47 uF across the batt, but I didn't get to that yet.


I found that the bypass cap didn't do anything at all. In the simulations,
there was no energy getting transferred through it. Thats why I dumped it.

With one white LED (which is really a blue LED) instead, I get 47 mA
at 1.5VDC. When I short across t he 1N4148, the current jumps up to
57 mA, and the LED gets appreciably brighter.

Must be some difference with the LEDs. When I breadboarded it, the
opposite
happened (ie, it got more efficient, and the LED brightened using the
diode)

CORRECTION:

Actually, this is wrong. Its based on simulations.

When I put a wire across the diode, it does get a bit brighter, but the
current goes way up, not proportional to the amount of brightness increase,
so I guess its probably getting much less efficient, as in the simulations.

Again, in the simulations, the difference in efficiency with the diode was
remarkable, going from 20% to above 60%. I'm still not sure if this is an
artifact of the simulation. I guess I'll have to build a power meter again
to measure it.
Click to expand...
Someone here showed me an easy way to find the current thru the LED.
Put a 1 ohm resistor in series with the LED, and bypass the 1 ohm with
a cap. Then just measure the millivolts across the 1 ohm to find the
current.

I simulate the simulator, I solder some parts together to see if the
simulator design deals with the Real World properly. ;-)
BTW, sometimes I'm disappointed.

[snip]
Regards,
Bob Monsen
Click to expand...

--
@@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 <at> 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@@
 
"Watson A.Name - Watt Sun, Dark Remover" <alondra101@hotmail.com> wrote in
message news:MPG.1a386a6eca3aa6429899c9@news.dslextreme.com...
I simulate the simulator, I solder some parts together to see if the
simulator design deals with the Real World properly. ;-)
BTW, sometimes I'm disappointed.
Click to expand...
Simulators are very fast for doing simple things like this. Unfortunately,
they let you down sometimes. The LED model on CircuitMaker that I was using
really doesn't seem to model an actual device at all.

Regards,
Bob Monsen
 
In article <lYlzb.194$F57.157@newsfep3-gui.server.ntli.net>,
mjolinor@hotmail.com mentioned...
"Watson A.Name - Watt Sun, Dark Remover" <alondra101@hotmail.com> wrote in
[snip]
Click to expand...

Found another one you may like to try, seems easy to get working, mind you
it should be with 1 transistor 1 resistor and 1 inductor:)

http://users.cableaz.com/~cappels/dproj/ledpage/leddrv.htm
Click to expand...
This is the original circuit that I used when I started experimenting
with the white LEDs and V boost circuits. I added a byp cap across
the battery. The problem I had was to get enough current from it to
light the LED brightly. I found that with the 2N4401, about the best
it would do was maybe 3 to 6 mA. The toroids I was winding were part
of the problem, they had too many turns and too high DC resistance.
And the toroid material wasn't good for below 1 NHz. Here's a similar
circuit, in Japanese.
http://www.geocities.co.jp/Technopolis-Mars***/2881/LED02.html

Next problem: If I want to put more than 1 LED on is it a circuit each. I
have real difficulty actually putting figures on how much light there is
particularly if I have two LEDs lit. I can easily fasten one LED to an LDR
and get a comparative brightness for different voltage / current /
frequencies / inductors but, for example if I parallel two LEDs then the
current only goes up by a ma or so but both LEDs are obviously lit though
each LED is slightly dimmer than it was with only one. How do I tell if the
overall light output is more than it was with one LED?

Lots of variables and no solutions.
Click to expand...
As I was just saying in the previous post, the easy way to do it is to
put a 1 ohm, 1/4W resistor in series with each LED, and a bypass cap
across that resistor. Then just measure the V drop with the DMM, and
millivolts gives you the current in mA. Costs all of a penny and a
half for the resistor. You can take it out or short it after you're
happy with the current.

I paralleled the output transistor with a second transistor, added
another coil, and another LED. Works fine, and puts out twice as much
light. You can also put two LEDs in series across the coil.

The 2N4401 can't put out enough power at low voltage to drive more
than about a single LED, so paralleling them for multiple LEDs is a
way to 'cheat'.

--
@@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 <at> 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 <ZgAzb.415167$Tr4.1193278@attbi_s03>,
postmaster@BulkingPro.com mentioned...
"Watson A.Name - Watt Sun, Dark Remover" <alondra101@hotmail.com> wrote in
message news:MPG.1a386a6eca3aa6429899c9@news.dslextreme.com...

I simulate the simulator, I solder some parts together to see if the
simulator design deals with the Real World properly. ;-)
BTW, sometimes I'm disappointed.


Simulators are very fast for doing simple things like this. Unfortunately,
they let you down sometimes. The LED model on CircuitMaker that I was using
really doesn't seem to model an actual device at all.
Click to expand...
Well, I guess that leaves you one choice. Warm up the smoldering
iron. ;-)

Regards,
Bob Monsen
Click to expand...

--
@@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 <at> 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@@
 
"Watson A.Name - Watt Sun, Dark Remover" <alondra101@hotmail.com> wrote in
message news:MPG.1a38c53a616d1439899ca@news.dslextreme.com...
In article <lYlzb.194$F57.157@newsfep3-gui.server.ntli.net>,
mjolinor@hotmail.com mentioned...

"Watson A.Name - Watt Sun, Dark Remover" <alondra101@hotmail.com> wrote
in
[snip]

Found another one you may like to try, seems easy to get working, mind
you
it should be with 1 transistor 1 resistor and 1 inductor:)

http://users.cableaz.com/~cappels/dproj/ledpage/leddrv.htm

This is the original circuit that I used when I started experimenting
with the white LEDs and V boost circuits. I added a byp cap across
the battery. The problem I had was to get enough current from it to
light the LED brightly. I found that with the 2N4401, about the best
it would do was maybe 3 to 6 mA. The toroids I was winding were part
of the problem, they had too many turns and too high DC resistance.
And the toroid material wasn't good for below 1 NHz. Here's a similar
circuit, in Japanese.
Click to expand...
http://www.dknpowerline.com/pictures/led_1.jpg

That's it built and it draws 70ma @ 1.5 volt, plenty bright but the coil
seems to be a bit critical. If I use a smaller ferrite it will saturate, if
I use thinner wire and the same number of turns it is pretty crap. It seems
to have to be about 1mm wire in order for the current to get up and it seems
that the inductance doesn't need to be that high, 3 turns either side (6 in
total) seems to be enough but 5 either side (as in the picture) seems best.
It is also significantly better if you do 5 turns spread all the way round
the toroid then the other 5 turns all the way round on top of the first
winding. Better than 10 turns round the ferrite and a centre tap.
 
Mjolinor wrote:
"Watson A.Name - Watt Sun, Dark Remover" <alondra101@hotmail.com> wrote in
message news:MPG.1a38c53a616d1439899ca@news.dslextreme.com...

In article <lYlzb.194$F57.157@newsfep3-gui.server.ntli.net>,
mjolinor@hotmail.com mentioned...

"Watson A.Name - Watt Sun, Dark Remover" <alondra101@hotmail.com> wrote

in

[snip]


Found another one you may like to try, seems easy to get working, mind

you

it should be with 1 transistor 1 resistor and 1 inductor:)

http://users.cableaz.com/~cappels/dproj/ledpage/leddrv.htm

This is the original circuit that I used when I started experimenting
with the white LEDs and V boost circuits. I added a byp cap across
the battery. The problem I had was to get enough current from it to
light the LED brightly. I found that with the 2N4401, about the best
it would do was maybe 3 to 6 mA. The toroids I was winding were part
of the problem, they had too many turns and too high DC resistance.
And the toroid material wasn't good for below 1 NHz. Here's a similar
circuit, in Japanese.


http://www.dknpowerline.com/pictures/led_1.jpg

That's it built and it draws 70ma @ 1.5 volt, plenty bright but the coil
seems to be a bit critical. If I use a smaller ferrite it will saturate, if
I use thinner wire and the same number of turns it is pretty crap. It seems
to have to be about 1mm wire in order for the current to get up and it seems
that the inductance doesn't need to be that high, 3 turns either side (6 in
total) seems to be enough but 5 either side (as in the picture) seems best.
It is also significantly better if you do 5 turns spread all the way round
the toroid then the other 5 turns all the way round on top of the first
winding. Better than 10 turns round the ferrite and a centre tap.
Click to expand...
Some recommend winding them bifilar. Or trifilar, with two of the
windings paralleled for the collector. That allows smaller wire, and
makes it easier to wind.

The core material of the toroid is also important. I used the RFI
suppressor sleeve off of keyboard and video cables for the toroid. I
saw it in half with a ceramic tile saw, and sand it down flat on a piece
of heavy sandpaper used for sanding floors. I usually get 100 uH with
10 to 12 turns of 24 gauge solid telephone wire. Different core
materials may have 60 uH or as high as 180 uH with the same number of
turns. They all work, tho.

But the nice thing about the circuit is you can parallel 2, 3, or more
2N4401s, to get more current. But it's just easier to put a single
higher current transistor in there.


--
@@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 <at> 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@@
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
u
d
e
d

t
e
x
t

m
s
g
 
You must log in or register to reply here.

Welcome to EDABoard.com

Sponsor

Back
Top