Joule Thief - still not working....

[default]

On Thu, 23 Jul 2009 15:53:44 -0700 (PDT), fungus
<openglMYSOCKS@artlum.com> wrote:

On Jul 23, 11:54 pm, default <defa...@defaulter.net> wrote:
On Thu, 23 Jul 2009 14:30:02 -0700 (PDT), fungus

openglMYSO...@artlum.com> wrote:
On Jul 23, 2:03 pm, default <defa...@defaulter.net> wrote:

The coil can be wound with a tap or separately.  If you use a tap and
keep winding in the same direction it will be phased right.

I assume the winding after the tap has to go over the top of the
previous winding, right?

No.  They can be side by side.


I'm not sure I'm being clear (my bad), I mean do
they have to be like this:

ABABABABABABAB
=================
| BABABABABABABA
tap

or can they be:

AAAAAAA BBBBBBBB
==================
AAAAAAA | BBBBBBBB
tap

(view with monospace font...)

On the rusty nail he uses a twisted pair

Either way. The top example might produce slightly better results
but I couldn't say for sure. As the winding spreads out, inductance
per turn goes down and resistance goes up (all other things being
equal again) so the Q might be slightly lower, for any benefit gained
by keeping the leakage lower.

No advantage, I can think of, to using a twisted pair.

You mentioned a ferrite rod or bar? Leave a little bit of ferrite rod
with no coil over the ends (about 1/2 -2 X the diameter of the rod)
don't wind right to the ends.

When I was a kid we'd use center tapped filament transformers for
blocking oscillators, and then use them to step up voltage to get
500-1000 volts from a battery supply. I tried that with a tapped
modem transformer recently and even though the primary to secondary
ratio was only 1:2 the output voltage was ~350 volts with 12V input
(because the wave form contains a large spike).
--

 

[fungus]

On Jul 23, 5:13 pm, Jon Kirwan <j...@infinitefactors.org> wrote:

...a 'speed-up' capacitor.
...
You made me curious enough to try a simulation with 200uH on each half
of the transformer, a 2N2222 BJT, a 1N5819 freewheeling diode, 10uF
output cap (across the 6 LEDs with Vfwd=3, Ron=12.6), and 3 fresh
batteries.  The Spice results look like:

Battery     C       BJT       LEDS
 436mW      0pF     28mW     19.9mA
 428mW     22pF     16mW     20.0mA
 425mW    220pF     19mW     19.7mA
 392mW    2.2nF     50mW     16.5mA
 173mW     22nF    142mW      0.0mA

So battery power does go down.  But BJT power goes up and they
eventually cross and the whole endevour is a waste.  In other words,
the BJT starts eating up all the power and eventually eats up
everything there is.

Ok, that's scary. I'll stick to playing with the number
of turns of wire for a while. I'm going to try a ferrite rod
instead of a ring because they're much quicker to add
and remove wire (and you can even do it while the
circuit is running!)

 

[fungus]

On Jul 23, 4:36 pm, "petrus bitbyter"
<pieterkraltlaatdit...@enditookhccnet.nl> wrote:

The diode you added makes things worse....

Wasn't me that added it...

So you need to move the diode. Place it between R1 and the base of the
transistor.

I'll try that, thanks.

Next step is adding more turns to the coil as others stated already. Guess
you will need 1.5-2 times the original number of turns.

Yes, this seems to be the place to experiment.

 

[fungus]

On Jul 23, 2:03 pm, default <defa...@defaulter.net> wrote:

The coil can be wound with a tap or separately.  If you use a tap and
keep winding in the same direction it will be phased right.

I assume the winding after the tap has to go over the top of the
previous winding, right?

 

[John Fields]

On Thu, 23 Jul 2009 16:06:41 -0700, John Larkin
<jjlarkin@highNOTlandTHIStechnologyPART.com> wrote:

On Thu, 23 Jul 2009 20:32:20 GMT, Jon Kirwan
jonk@infinitefactors.org> wrote:

On Thu, 23 Jul 2009 13:23:15 -0700, John Larkin
jjlarkin@highNOTlandTHIStechnologyPART.com> wrote:

On Thu, 23 Jul 2009 19:04:43 GMT, Jon Kirwan
jonk@infinitefactors.org> wrote:

On Thu, 23 Jul 2009 09:24:55 -0700, John Larkin
jjlarkin@highNOTlandTHIStechnologyPART.com> wrote:

On Thu, 23 Jul 2009 04:20:21 -0700 (PDT), fungus
openglMYSOCKS@artlum.com> wrote:

I just got some proper parts to start making joule thieves but I'm
still
having problems.

The circuit is this: http://www.artlum.com/jt/joulethief.gif

Except I have R1 and L1 one the other way around (as in the original
web page at http://www.emanator.demon.co.uk/bigclive/joule.htm )

The problem is that my transistors keep on overheating and dying.
Why should this be? I'm using a 2N2222 in metal can (as shown here
http://en.wikipedia.org/wiki/2N2222 ). These can switch at hundreds
of megahertz so I don't think it's because of slow switching.

I measured the current at point X and it seems high - over 100mA.
Could this be the cause of the overheating? Even if it isn't the
problem
it seems wasteful. I tried putting in a resistor there but the circuit
shuts down.
.
I also tried a honking big "high speed switching" transistor pulled
out of a PSU but it made the LEDs go very dim.

Any ideas?

That's a horrible circuit. Too many conflicting parameters depend on
the value of R1. A proper blocking oscillator uses an RC time constant
to set the rep rate, and a separate resistor to limit the base
current.

ftp://jjlarkin.lmi.net/BlockOsc.JPG

Would you care to provide some sample values and analyze that circuit
for us?

No, too much work.

Hmm.

Just to goose things along, for the joule thief circuit I get
something like this for the frequency:

(Vbattery - Vsat) * (Vout + Vfreewheeldiode - Vbattery)
f = -------------------------------------------------------
Ic_peak * L_collector * (Vout + Vfreewheeldiode - Vsat)

Ic_peak may require an iteration or two with a datasheet to
approximate. I just go in with an assumed Ic, look up a beta estimate
for that on one curve and then grab the Vbe estimate from another
curve, and apply them into:

Ic_peak = beta*(Nratio*(Vbattery - Vsat) + Vbattery - Vbe))/Rbase

That Ic_peak is then used to repeat the process. When it settles,
I've usually got a reasonable figure that I can use to compute 'f'.
(Nratio is the turns ratio, usually just 1.) I tend to use Vsat=0.2V.

If your suggestion is so nicely designable, can't you at least provide
an approximate equation?

I see the RC node moving towards a bias point, but not really
setting the frequency at which the BJT goes on and off. But I haven't
sat down more than to glance over it, yet.

In general, "on" pulse width is set by the volt-second saturation of
the inductor (although a small value of C can make it shorter.)

So in your circuit case, it does depend on saturation of the core.
What would happen in an air core case?

The classic tube "blocking oscillator" had its ON time determined by
inductor saturation. If it can't saturate, the ON interval ends when
the transistor runs out of beta (or the tube out of plate current), or
when C runs out of charge to drive the base/grid. The "blocking" part
was the negative swing on the tube grid from grid current charging the
cap; it fired again when R1 charged the grid the other way, back to
the turnon threshold.


Base
current is limited by R2 (the one connected to the base.) While the
transistor's on, the base current charges up the cap, and that charge
will back-bias the transistor until R1 recharges the cap back up to
+0.7 volts, at which it fires again.

Something like that.

Try R1=1K, R2=100 C=100nF as very rough starting points. A lot depends
on the inductor. It won't Spice unless the model includes inductor
saturation.

Yes. I gather.

Unless L can't saturate, of course. Then it's not an official
"blocking oscillator."


It's probebly easier to use a Tiny Logic schmitt-trigger oscillator to
drive the transistor, and just use a single-winding inductor. Blocking
oscillators are tricky.

Single BJTs are cheap and, if you saw one of the web sites mentioned
some time back in the related thread, you'd have seen that the whole
thing is tiny enough to place inside a small flashlight bulb base.

If you don't mind the 2-winding coil, and the additional futzing, the
blocking oscillator is potentially cheap.


...

Since you write, "That's a horrible circuit. Too many conflicting
parameters depend on the value of R1. A proper blocking oscillator
uses an RC time constant to set the rep rate, and a separate resistor
to limit the base current," shouldn't it be the case that you can tell
me how to compute the frequency with ease? Isn't that the entire
point of saying all that? Or did I miss your point, here?

As I said, a blocking oscillator is complex. I can't define the
frequency "with ease." But having separate control over base drive and
rep-rate helps orthogonalize things. Having one part control two
circuit parameters can get awkward. Three is a nightmare.

The MIT RadLab books are full of blocking oscillator theory and
circuits, especially vol 19. Tube radars were full of them, as
oscillators, comparators, pulse regenerators, and frequency dividers.

Some texts referred to rf squegging circuits as blocking oscillators.

---
Translation:

I don't have a clue.

JF

 

[Jon Kirwan]

On Thu, 23 Jul 2009 16:06:41 -0700, John Larkin
<jjlarkin@highNOTlandTHIStechnologyPART.com> wrote:

On Thu, 23 Jul 2009 20:32:20 GMT, Jon Kirwan
jonk@infinitefactors.org> wrote:

On Thu, 23 Jul 2009 13:23:15 -0700, John Larkin
jjlarkin@highNOTlandTHIStechnologyPART.com> wrote:

On Thu, 23 Jul 2009 19:04:43 GMT, Jon Kirwan
jonk@infinitefactors.org> wrote:

On Thu, 23 Jul 2009 09:24:55 -0700, John Larkin
jjlarkin@highNOTlandTHIStechnologyPART.com> wrote:

On Thu, 23 Jul 2009 04:20:21 -0700 (PDT), fungus
openglMYSOCKS@artlum.com> wrote:

I just got some proper parts to start making joule thieves but I'm
still
having problems.

The circuit is this: http://www.artlum.com/jt/joulethief.gif

Except I have R1 and L1 one the other way around (as in the original
web page at http://www.emanator.demon.co.uk/bigclive/joule.htm )

The problem is that my transistors keep on overheating and dying.
Why should this be? I'm using a 2N2222 in metal can (as shown here
http://en.wikipedia.org/wiki/2N2222 ). These can switch at hundreds
of megahertz so I don't think it's because of slow switching.

I measured the current at point X and it seems high - over 100mA.
Could this be the cause of the overheating? Even if it isn't the
problem
it seems wasteful. I tried putting in a resistor there but the circuit
shuts down.
.
I also tried a honking big "high speed switching" transistor pulled
out of a PSU but it made the LEDs go very dim.

Any ideas?

That's a horrible circuit. Too many conflicting parameters depend on
the value of R1. A proper blocking oscillator uses an RC time constant
to set the rep rate, and a separate resistor to limit the base
current.

ftp://jjlarkin.lmi.net/BlockOsc.JPG

Would you care to provide some sample values and analyze that circuit
for us?

No, too much work.

Hmm.

Just to goose things along, for the joule thief circuit I get
something like this for the frequency:

(Vbattery - Vsat) * (Vout + Vfreewheeldiode - Vbattery)
f = -------------------------------------------------------
Ic_peak * L_collector * (Vout + Vfreewheeldiode - Vsat)

Ic_peak may require an iteration or two with a datasheet to
approximate. I just go in with an assumed Ic, look up a beta estimate
for that on one curve and then grab the Vbe estimate from another
curve, and apply them into:

Ic_peak = beta*(Nratio*(Vbattery - Vsat) + Vbattery - Vbe))/Rbase

That Ic_peak is then used to repeat the process. When it settles,
I've usually got a reasonable figure that I can use to compute 'f'.
(Nratio is the turns ratio, usually just 1.) I tend to use Vsat=0.2V.

If your suggestion is so nicely designable, can't you at least provide
an approximate equation?

I see the RC node moving towards a bias point, but not really
setting the frequency at which the BJT goes on and off. But I haven't
sat down more than to glance over it, yet.

In general, "on" pulse width is set by the volt-second saturation of
the inductor (although a small value of C can make it shorter.)

So in your circuit case, it does depend on saturation of the core.
What would happen in an air core case?

The classic tube "blocking oscillator" had its ON time determined by
inductor saturation. If it can't saturate, the ON interval ends when
the transistor runs out of beta (or the tube out of plate current), or
when C runs out of charge to drive the base/grid. The "blocking" part
was the negative swing on the tube grid from grid current charging the
cap; it fired again when R1 charged the grid the other way, back to
the turnon threshold.


Base
current is limited by R2 (the one connected to the base.) While the
transistor's on, the base current charges up the cap, and that charge
will back-bias the transistor until R1 recharges the cap back up to
+0.7 volts, at which it fires again.

Something like that.

Try R1=1K, R2=100 C=100nF as very rough starting points. A lot depends
on the inductor. It won't Spice unless the model includes inductor
saturation.

Yes. I gather.

Unless L can't saturate, of course. Then it's not an official
"blocking oscillator."

Are the saturation of cores more predictable than BJT beta -- keeping
in mind that we are talking about the same part number AND
manufacturer in both cases?

It's probebly easier to use a Tiny Logic schmitt-trigger oscillator to
drive the transistor, and just use a single-winding inductor. Blocking
oscillators are tricky.

Single BJTs are cheap and, if you saw one of the web sites mentioned
some time back in the related thread, you'd have seen that the whole
thing is tiny enough to place inside a small flashlight bulb base.

If you don't mind the 2-winding coil, and the additional futzing, the
blocking oscillator is potentially cheap.


...

Since you write, "That's a horrible circuit. Too many conflicting
parameters depend on the value of R1. A proper blocking oscillator
uses an RC time constant to set the rep rate, and a separate resistor
to limit the base current," shouldn't it be the case that you can tell
me how to compute the frequency with ease? Isn't that the entire
point of saying all that? Or did I miss your point, here?

As I said, a blocking oscillator is complex. I can't define the
frequency "with ease." But having separate control over base drive and
rep-rate helps orthogonalize things. Having one part control two
circuit parameters can get awkward. Three is a nightmare.

But the existing schematic (the joule thief thing) does that, within
bounds. Assuming fixed battery voltage and fixed winding ratio of the
transformer, the base resistor sets the Ib. The beta then establishes
the peak Ic. I'm not sure of any advantages in the new arrangement
you suggest, yet. (And I suspect it's behavior is harder to analyze,
besides.)

The MIT RadLab books are full of blocking oscillator theory and
circuits, especially vol 19. Tube radars were full of them, as
oscillators, comparators, pulse regenerators, and frequency dividers.

I think someone posted some of that, some time back, in
sci.electronics.design. I'll see if I can track any of that down.
sadly, other than that possibility, I don't have ready access.

Some texts referred to rf squegging circuits as blocking oscillators.

I'll look to see if I can find a lucid description.

Thanks,
Jon

 

[fungus]

On Jul 23, 10:18 pm, "bw" <bweg...@hotmail.com> wrote:

Start with the original circuit, and get it to work on one battery.

then go from there to what you want to do.

That's what I'm doing...

The original circuit lights up a LED but the current
is very low - about 5mA.

To drive six LEDs at 20mA with one battery you'd
have to get the frequency up into the mHz (which
isn't going to happen).

The solution seems to be to raise the input voltage
so that's what I'm trying to get working. Unfortunately
I never studied electronics beyond what they told me
in high-school physics class so I'm at a disadvantage.

 

[David Eather]

fungus wrote:

On Jul 23, 1:52 pm, David Eather <eat...@tpg.com.au> wrote:
fungus wrote:
Any ideas?
Yes. Figure out what you want to do and state it explicitly and exactly.
Then work to that goal in steps you understand.

a) I want to light up some LEDs (eg. six of them) using batteries, eg.
three AAAs. Circuit is decorative and has to be small because I want
to hide it.

b) I want them to be as bright as possible - the full 20mA or as close
to it as I can get.

c) It's a battery ... so voltage is going to drop over time (from 4.6V
to
3.3V), this makes part (b) problematic. I accept that current will
drop
a bit, but if it can stay in the range 15-20mA then that's Ok.

I've figured out that a Joule Thief is much closer to these
characteristics
than a simple resistor circuit doesn't. See the graph I plotted here:
http://www.artlum.com/jt/jt_vs_res.gif

But ... at the moment it's eating up transistors.

Also what is the resistance value of R1?

In the interim try this:

1 - increase R1 by a factor of 10 and next by a factor of 100, does that
reduce the heating of the 2n2222? (if you are over-driving the base of
the transistor and the inductance of the transformer is too small this
would cause the problem you are seeing - I think JK is also thinking
along these lines)

2 - increase the number of turns in both l1 and l2 by the same ratio e.g
.. double the number of turns on both or triple the number of turns on
both. This should make the transformer more efficient (and you probably
want to do that anyway)

 

[fungus]

On Jul 23, 4:27 pm, Jon Kirwan <j...@infinitefactors.org> wrote:

BJT temperature is related (obviously) to its power dissipation.  That
power dissipation comes from a variety of possible corners from my
hobby viewpoint:  (1)  The transistor has been damaged (diode put in
later, after it was already ruined perhaps?) and isn't operating well
anymore;

Fresh transistors are quite frequent at the moment...

or, (2) the base-emitter junction current;

I've moved the resistor in between the inductor and transistor
to try to limit this (ie. R1 is now exactly as shown in the
scematic). Doesn't seem to help much.

(3) collector-emitter current times collector-emitter voltage;

I'm suspecting this at the moment.

(4) frequency of operation is too high for the reverse transit
time of the BJT.

Not sure I understand that.

So wind more turns and get the frequency near or under 100kHz, or so,
where the reverse transit will be only be a few percent or so and
won't be wasting a lot of power.

ie. Lower frequencies mean the transistor will be switched on for less
percentage of the time and more electrons will go through the load
instead of being dumped to ground via the transistor.

(Yeah, I know - http://xkcd.com/567/ ).

 

[fungus]

On Jul 23, 11:54 pm, default <defa...@defaulter.net> wrote:

On Thu, 23 Jul 2009 14:30:02 -0700 (PDT), fungus

openglMYSO...@artlum.com> wrote:
On Jul 23, 2:03 pm, default <defa...@defaulter.net> wrote:

The coil can be wound with a tap or separately.  If you use a tap and
keep winding in the same direction it will be phased right.

I assume the winding after the tap has to go over the top of the
previous winding, right?

No.  They can be side by side.

I'm not sure I'm being clear (my bad), I mean do
they have to be like this:

ABABABABABABAB
================ | BABABABABABABA
tap

or can they be:

AAAAAAA BBBBBBBB
=================AAAAAAA | BBBBBBBB
tap

(view with monospace font...)

On the rusty nail he uses a twisted pair

 

[whit3rd]

On Jul 23, 7:20 am, fungus <openglMYSO...@artlum.com> wrote:

I just got some proper parts to start making joule thieves but I'm
still
having problems.

The circuit is this:http://www.artlum.com/jt/joulethief.gif

That kind of oscillator (blocking oscillator) depends on saturation
of the core, which is quite difficult to engineer (and somewhat
energy-lossy). Unless you really need a minimum-parts solution,
it's easier to run a '555 oscillator into an autotransformer to
boost voltage. Remember to use DC blocking capacitors,
and you can get nearly half a watt from such a circuit.

 

[David Eather]

David Eather wrote:

fungus wrote:
On Jul 23, 1:52 pm, David Eather <eat...@tpg.com.au> wrote:
fungus wrote:
Any ideas?
Yes. Figure out what you want to do and state it explicitly and exactly.
Then work to that goal in steps you understand.

a) I want to light up some LEDs (eg. six of them) using batteries, eg.
three AAAs. Circuit is decorative and has to be small because I want
to hide it.

b) I want them to be as bright as possible - the full 20mA or as close
to it as I can get.

c) It's a battery ... so voltage is going to drop over time (from 4.6V
to
3.3V), this makes part (b) problematic. I accept that current will
drop
a bit, but if it can stay in the range 15-20mA then that's Ok.

I've figured out that a Joule Thief is much closer to these
characteristics
than a simple resistor circuit doesn't. See the graph I plotted here:
http://www.artlum.com/jt/jt_vs_res.gif

But ... at the moment it's eating up transistors.

Also what is the resistance value of R1?

In the interim try this:

forget about point 1 - I made a wrong assumption about your circuit. I
assumed you had a resistor connected from the base of the transistor to
the junction l1 and d7. If you want to try it put a 1k resistor in -you
may need to change this value upwards if the transistor still runs too
hot or downwards if the circuit stops working.

1 - increase R1 by a factor of 10 and next by a factor of 100, does that
reduce the heating of the 2n2222? (if you are over-driving the base of
the transistor and the inductance of the transformer is too small this
would cause the problem you are seeing - I think JK is also thinking
along these lines)

2 - increase the number of turns in both l1 and l2 by the same ratio e.g
. double the number of turns on both or triple the number of turns on
both. This should make the transformer more efficient (and you probably
want to do that anyway)

 

[fungus]

On Jul 24, 1:18 am, David Eather <eat...@tpg.com.au> wrote:

What voltage are you running the joule thief on?

Right now? I just measured 3.8V.

 

[fungus]

On Jul 24, 2:06 am, David Eather <eat...@tpg.com.au> wrote:

fungus wrote:
On Jul 23, 1:52 pm, David Eather <eat...@tpg.com.au> wrote:
fungus wrote:
Any ideas?
Yes. Figure out what you want to do and state it explicitly and exactly.
Then work to that goal in steps you understand.

a) I want to light up some LEDs (eg. six of them) using batteries, eg.
three AAAs. Circuit is decorative and has to be small because I want
to hide it.

b) I want them to be as bright as possible - the full 20mA or as close
to it as I can get.

c) It's a battery ... so voltage is going to drop over time (from 4.6V
to
3.3V), this makes part (b) problematic. I accept that current will
drop
a bit, but if it can stay in the range 15-20mA then that's Ok.

I've figured out that a Joule Thief is much closer to these
characteristics
than a simple resistor circuit doesn't. See the graph I plotted here:
http://www.artlum.com/jt/jt_vs_res.gif

But ... at the moment it's eating up transistors.

Also what is the resistance value of R1?

1k

In the interim try this:

1 - increase R1 by a factor of 10 ... does that
reduce the heating of the 2n2222?

Yes, but the current going through the LEDs
dropped from 18mA to 5mA.

 

[John Larkin]

On Thu, 23 Jul 2009 23:34:04 GMT, Jon Kirwan
<jonk@infinitefactors.org> wrote:

On Thu, 23 Jul 2009 16:06:41 -0700, John Larkin
jjlarkin@highNOTlandTHIStechnologyPART.com> wrote:

On Thu, 23 Jul 2009 20:32:20 GMT, Jon Kirwan
jonk@infinitefactors.org> wrote:

On Thu, 23 Jul 2009 13:23:15 -0700, John Larkin
jjlarkin@highNOTlandTHIStechnologyPART.com> wrote:

On Thu, 23 Jul 2009 19:04:43 GMT, Jon Kirwan
jonk@infinitefactors.org> wrote:

On Thu, 23 Jul 2009 09:24:55 -0700, John Larkin
jjlarkin@highNOTlandTHIStechnologyPART.com> wrote:

On Thu, 23 Jul 2009 04:20:21 -0700 (PDT), fungus
openglMYSOCKS@artlum.com> wrote:

I just got some proper parts to start making joule thieves but I'm
still
having problems.

The circuit is this: http://www.artlum.com/jt/joulethief.gif

Except I have R1 and L1 one the other way around (as in the original
web page at http://www.emanator.demon.co.uk/bigclive/joule.htm )

The problem is that my transistors keep on overheating and dying.
Why should this be? I'm using a 2N2222 in metal can (as shown here
http://en.wikipedia.org/wiki/2N2222 ). These can switch at hundreds
of megahertz so I don't think it's because of slow switching.

I measured the current at point X and it seems high - over 100mA.
Could this be the cause of the overheating? Even if it isn't the
problem
it seems wasteful. I tried putting in a resistor there but the circuit
shuts down.
.
I also tried a honking big "high speed switching" transistor pulled
out of a PSU but it made the LEDs go very dim.

Any ideas?

That's a horrible circuit. Too many conflicting parameters depend on
the value of R1. A proper blocking oscillator uses an RC time constant
to set the rep rate, and a separate resistor to limit the base
current.

ftp://jjlarkin.lmi.net/BlockOsc.JPG

Would you care to provide some sample values and analyze that circuit
for us?

No, too much work.

Hmm.

Just to goose things along, for the joule thief circuit I get
something like this for the frequency:

(Vbattery - Vsat) * (Vout + Vfreewheeldiode - Vbattery)
f = -------------------------------------------------------
Ic_peak * L_collector * (Vout + Vfreewheeldiode - Vsat)

Ic_peak may require an iteration or two with a datasheet to
approximate. I just go in with an assumed Ic, look up a beta estimate
for that on one curve and then grab the Vbe estimate from another
curve, and apply them into:

Ic_peak = beta*(Nratio*(Vbattery - Vsat) + Vbattery - Vbe))/Rbase

That Ic_peak is then used to repeat the process. When it settles,
I've usually got a reasonable figure that I can use to compute 'f'.
(Nratio is the turns ratio, usually just 1.) I tend to use Vsat=0.2V.

If your suggestion is so nicely designable, can't you at least provide
an approximate equation?

I see the RC node moving towards a bias point, but not really
setting the frequency at which the BJT goes on and off. But I haven't
sat down more than to glance over it, yet.

In general, "on" pulse width is set by the volt-second saturation of
the inductor (although a small value of C can make it shorter.)

So in your circuit case, it does depend on saturation of the core.
What would happen in an air core case?

The classic tube "blocking oscillator" had its ON time determined by
inductor saturation. If it can't saturate, the ON interval ends when
the transistor runs out of beta (or the tube out of plate current), or
when C runs out of charge to drive the base/grid. The "blocking" part
was the negative swing on the tube grid from grid current charging the
cap; it fired again when R1 charged the grid the other way, back to
the turnon threshold.


Base
current is limited by R2 (the one connected to the base.) While the
transistor's on, the base current charges up the cap, and that charge
will back-bias the transistor until R1 recharges the cap back up to
+0.7 volts, at which it fires again.

Something like that.

Try R1=1K, R2=100 C=100nF as very rough starting points. A lot depends
on the inductor. It won't Spice unless the model includes inductor
saturation.

Yes. I gather.

Unless L can't saturate, of course. Then it's not an official
"blocking oscillator."

Are the saturation of cores more predictable than BJT beta -- keeping
in mind that we are talking about the same part number AND
manufacturer in both cases?

It's probebly easier to use a Tiny Logic schmitt-trigger oscillator to
drive the transistor, and just use a single-winding inductor. Blocking
oscillators are tricky.

Single BJTs are cheap and, if you saw one of the web sites mentioned
some time back in the related thread, you'd have seen that the whole
thing is tiny enough to place inside a small flashlight bulb base.

If you don't mind the 2-winding coil, and the additional futzing, the
blocking oscillator is potentially cheap.


...

Since you write, "That's a horrible circuit. Too many conflicting
parameters depend on the value of R1. A proper blocking oscillator
uses an RC time constant to set the rep rate, and a separate resistor
to limit the base current," shouldn't it be the case that you can tell
me how to compute the frequency with ease? Isn't that the entire
point of saying all that? Or did I miss your point, here?

As I said, a blocking oscillator is complex. I can't define the
frequency "with ease." But having separate control over base drive and
rep-rate helps orthogonalize things. Having one part control two
circuit parameters can get awkward. Three is a nightmare.

But the existing schematic (the joule thief thing) does that, within
bounds. Assuming fixed battery voltage and fixed winding ratio of the
transformer, the base resistor sets the Ib. The beta then establishes
the peak Ic. I'm not sure of any advantages in the new arrangement
you suggest, yet. (And I suspect it's behavior is harder to analyze,
besides.)

In the OP's ascii-art circuit, R1 determines ON base current (and
perhaps ON time, if the inductor doesn't saturate first) and also sets
OFF time, as part of the L/R decay. It may be hard to pick one value
that does both right, namely produces an efficient duty cycle, as
witnessed by the many blown up transistors.

The big advantage of the circuit I posted is that rep-rate can be set
independent of pulse width... two knobs to turn. That allows low duty
cycles which won't fry transistors. And brightness control, if you
want it.

The MIT RadLab books are full of blocking oscillator theory and
circuits, especially vol 19. Tube radars were full of them, as
oscillators, comparators, pulse regenerators, and frequency dividers.

I think someone posted some of that, some time back, in
sci.electronics.design. I'll see if I can track any of that down.
sadly, other than that possibility, I don't have ready access.

Some texts referred to rf squegging circuits as blocking oscillators.

I'll look to see if I can find a lucid description.

Terman's "Radio Engineering" texts refer to squeggers as blocking
oscillators. But Terman was a little weird.

Millman&Taub's classic "Pulse and Digital Circuits" has a whole
chapter "Pulse Transformers and Blocking Oscillators", with a bunch of
analysis. All tubes.

John

 

[Jon Kirwan]

On Thu, 23 Jul 2009 20:25:03 -0700, John Larkin
<jjlarkin@highNOTlandTHIStechnologyPART.com> wrote:

On Thu, 23 Jul 2009 23:34:04 GMT, Jon Kirwan
jonk@infinitefactors.org> wrote:

On Thu, 23 Jul 2009 16:06:41 -0700, John Larkin
jjlarkin@highNOTlandTHIStechnologyPART.com> wrote:

On Thu, 23 Jul 2009 20:32:20 GMT, Jon Kirwan
jonk@infinitefactors.org> wrote:

On Thu, 23 Jul 2009 13:23:15 -0700, John Larkin
jjlarkin@highNOTlandTHIStechnologyPART.com> wrote:

On Thu, 23 Jul 2009 19:04:43 GMT, Jon Kirwan
jonk@infinitefactors.org> wrote:

On Thu, 23 Jul 2009 09:24:55 -0700, John Larkin
jjlarkin@highNOTlandTHIStechnologyPART.com> wrote:

On Thu, 23 Jul 2009 04:20:21 -0700 (PDT), fungus
openglMYSOCKS@artlum.com> wrote:

I just got some proper parts to start making joule thieves but I'm
still
having problems.

The circuit is this: http://www.artlum.com/jt/joulethief.gif

Except I have R1 and L1 one the other way around (as in the original
web page at http://www.emanator.demon.co.uk/bigclive/joule.htm )

The problem is that my transistors keep on overheating and dying.
Why should this be? I'm using a 2N2222 in metal can (as shown here
http://en.wikipedia.org/wiki/2N2222 ). These can switch at hundreds
of megahertz so I don't think it's because of slow switching.

I measured the current at point X and it seems high - over 100mA.
Could this be the cause of the overheating? Even if it isn't the
problem
it seems wasteful. I tried putting in a resistor there but the circuit
shuts down.
.
I also tried a honking big "high speed switching" transistor pulled
out of a PSU but it made the LEDs go very dim.

Any ideas?

That's a horrible circuit. Too many conflicting parameters depend on
the value of R1. A proper blocking oscillator uses an RC time constant
to set the rep rate, and a separate resistor to limit the base
current.

ftp://jjlarkin.lmi.net/BlockOsc.JPG

Would you care to provide some sample values and analyze that circuit
for us?

No, too much work.

Hmm.

Just to goose things along, for the joule thief circuit I get
something like this for the frequency:

(Vbattery - Vsat) * (Vout + Vfreewheeldiode - Vbattery)
f = -------------------------------------------------------
Ic_peak * L_collector * (Vout + Vfreewheeldiode - Vsat)

Ic_peak may require an iteration or two with a datasheet to
approximate. I just go in with an assumed Ic, look up a beta estimate
for that on one curve and then grab the Vbe estimate from another
curve, and apply them into:

Ic_peak = beta*(Nratio*(Vbattery - Vsat) + Vbattery - Vbe))/Rbase

That Ic_peak is then used to repeat the process. When it settles,
I've usually got a reasonable figure that I can use to compute 'f'.
(Nratio is the turns ratio, usually just 1.) I tend to use Vsat=0.2V.

If your suggestion is so nicely designable, can't you at least provide
an approximate equation?

I see the RC node moving towards a bias point, but not really
setting the frequency at which the BJT goes on and off. But I haven't
sat down more than to glance over it, yet.

In general, "on" pulse width is set by the volt-second saturation of
the inductor (although a small value of C can make it shorter.)

So in your circuit case, it does depend on saturation of the core.
What would happen in an air core case?

The classic tube "blocking oscillator" had its ON time determined by
inductor saturation. If it can't saturate, the ON interval ends when
the transistor runs out of beta (or the tube out of plate current), or
when C runs out of charge to drive the base/grid. The "blocking" part
was the negative swing on the tube grid from grid current charging the
cap; it fired again when R1 charged the grid the other way, back to
the turnon threshold.


Base
current is limited by R2 (the one connected to the base.) While the
transistor's on, the base current charges up the cap, and that charge
will back-bias the transistor until R1 recharges the cap back up to
+0.7 volts, at which it fires again.

Something like that.

Try R1=1K, R2=100 C=100nF as very rough starting points. A lot depends
on the inductor. It won't Spice unless the model includes inductor
saturation.

Yes. I gather.

Unless L can't saturate, of course. Then it's not an official
"blocking oscillator."

Are the saturation of cores more predictable than BJT beta -- keeping
in mind that we are talking about the same part number AND
manufacturer in both cases?

It's probebly easier to use a Tiny Logic schmitt-trigger oscillator to
drive the transistor, and just use a single-winding inductor. Blocking
oscillators are tricky.

Single BJTs are cheap and, if you saw one of the web sites mentioned
some time back in the related thread, you'd have seen that the whole
thing is tiny enough to place inside a small flashlight bulb base.

If you don't mind the 2-winding coil, and the additional futzing, the
blocking oscillator is potentially cheap.


...

Since you write, "That's a horrible circuit. Too many conflicting
parameters depend on the value of R1. A proper blocking oscillator
uses an RC time constant to set the rep rate, and a separate resistor
to limit the base current," shouldn't it be the case that you can tell
me how to compute the frequency with ease? Isn't that the entire
point of saying all that? Or did I miss your point, here?

As I said, a blocking oscillator is complex. I can't define the
frequency "with ease." But having separate control over base drive and
rep-rate helps orthogonalize things. Having one part control two
circuit parameters can get awkward. Three is a nightmare.

But the existing schematic (the joule thief thing) does that, within
bounds. Assuming fixed battery voltage and fixed winding ratio of the
transformer, the base resistor sets the Ib. The beta then establishes
the peak Ic. I'm not sure of any advantages in the new arrangement
you suggest, yet. (And I suspect it's behavior is harder to analyze,
besides.)

In the OP's ascii-art circuit, R1 determines ON base current (and
perhaps ON time, if the inductor doesn't saturate first)

Yes. R1 determines the ON base drive, together of course with the
battery voltage, less something for the Vbe. The base drive
determines and BJT beta determines the peak sustainable Ic (leaving
out core satuation considerations), which then determines the ON
_time_.

and also sets
OFF time, as part of the L/R decay.

This, I do NOT see. Off time is determined by the required voltage
across the collector winding when the BJT goes OFF and the field
collapses, reversing the polarity. I really do NOT see how R1 plays
into that calculation, at all. The base winding really isn't tapping
much field energy -- most of which is being either driving out through
the LEDs or else via the freewheeling diode (1N5819?) and cap on the
output I suggested elsewhere. (I suggested also a diode to protect
the BJT base, but that's a separate issue.)

Can you explain how R1 plays into an L/R decay time? I'd like to hear
about it.

It may be hard to pick one value
that does both right, namely produces an efficient duty cycle, as
witnessed by the many blown up transistors.

Well, until I gather your point about the OFF time's L/R, I have to
withhold further comment.

The big advantage of the circuit I posted is that rep-rate can be set
independent of pulse width... two knobs to turn. That allows low duty
cycles which won't fry transistors. And brightness control, if you
want it.

I need to first fathom your L/R point and then I need to spend more
time with the suggested circuit you gave before I can agree. I
apologize for my ignorance on this, but that's where I'm at right now.

The MIT RadLab books are full of blocking oscillator theory and
circuits, especially vol 19. Tube radars were full of them, as
oscillators, comparators, pulse regenerators, and frequency dividers.

I think someone posted some of that, some time back, in
sci.electronics.design. I'll see if I can track any of that down.
sadly, other than that possibility, I don't have ready access.

Some texts referred to rf squegging circuits as blocking oscillators.

I'll look to see if I can find a lucid description.

Terman's "Radio Engineering" texts refer to squeggers as blocking
oscillators. But Terman was a little weird.

Millman&Taub's classic "Pulse and Digital Circuits" has a whole
chapter "Pulse Transformers and Blocking Oscillators", with a bunch of
analysis. All tubes.

I'm modestly familiar with vacuum tubes, load lines, grid leak
resistors (what a pain I had with those), etc. So I may somehow hope
to be able to follow along.

Jon

 

[David Eather]

fungus wrote:

On Jul 24, 2:06 am, David Eather <eat...@tpg.com.au> wrote:
fungus wrote:
On Jul 23, 1:52 pm, David Eather <eat...@tpg.com.au> wrote:
fungus wrote:
Any ideas?
Yes. Figure out what you want to do and state it explicitly and exactly.
Then work to that goal in steps you understand.
a) I want to light up some LEDs (eg. six of them) using batteries, eg.
three AAAs. Circuit is decorative and has to be small because I want
to hide it.
b) I want them to be as bright as possible - the full 20mA or as close
to it as I can get.
c) It's a battery ... so voltage is going to drop over time (from 4.6V
to
3.3V), this makes part (b) problematic. I accept that current will
drop
a bit, but if it can stay in the range 15-20mA then that's Ok.
I've figured out that a Joule Thief is much closer to these
characteristics
than a simple resistor circuit doesn't. See the graph I plotted here:
http://www.artlum.com/jt/jt_vs_res.gif
But ... at the moment it's eating up transistors.
Also what is the resistance value of R1?


1k

In the interim try this:

1 - increase R1 by a factor of 10 ... does that
reduce the heating of the 2n2222?


Yes, but the current going through the LEDs
dropped from 18mA to 5mA.

Iff (if and only if) you are in the mood for experimenting, more turns
on L2 might increase this current through the LED while R1 is 10k.

Also the range of currents 18ma to 5ma (4:1) vs the range of resistance
1k to 10k (1:10) suggests that some value between 1k and 10k will reduce
power consumption of the transistor with a much smaller decrease in
current to the LED. Perhaps there is a suitable compromise.

A different subject - I am seeking information.

How long does this thing have to run on one set of batteries? and if you
can how much current is coming out of the batteries when the LED's are
getting their 18ma? (if the 2n2222 is getting hot then this is a
missing piece of information.)

 

[David Eather]

fungus wrote:

On Jul 23, 10:18 pm, "bw" <bweg...@hotmail.com> wrote:
Start with the original circuit, and get it to work on one battery.

then go from there to what you want to do.

That's what I'm doing...

The original circuit lights up a LED but the current
is very low - about 5mA.

To drive six LEDs at 20mA with one battery you'd
have to get the frequency up into the mHz (which
isn't going to happen).

No, the frequency thing is not correct. You can increase the output
voltage by putting more turns on L2. It is the ratio of turns between L1
and L2 that mostly determines the output voltage. To go from 1 LED to 6
and upping the voltage by 3 times try doubling the turns on L2 and
increasing R1 to about 3k (2.7k or 3.3 would both be fine)

If you really wanted to nail it, triple the turns on L1 and put 6 times
the number of turns you already have on L2 and change R1, but just L2
and R1 will get you closer to where you want to go.

The solution seems to be to raise the input voltage
so that's what I'm trying to get working. Unfortunately
I never studied electronics beyond what they told me
in high-school physics class so I'm at a disadvantage.

 

[greg]

fungus wrote:

The problem is that my transistors keep on overheating and dying.
Why should this be?

My first suspicion would be switching losses in the
transistor.

What does the voltage waveform across the emitter and
collector of the transistor look like? Does it have
nice sharp vertical edges?

If not, then you need to lower the operating frequency
until the switching time is negligible compared to the
on and off times. Increasing the number of turns on
the inductor should do that.

I'm using a 2N2222 in metal can (as shown here
http://en.wikipedia.org/wiki/2N2222 ). These can switch at hundreds
of megahertz

Like Jon said, f_T is NOT the relevant parameter here,
it's only about amplifying small signals. You can't
expect the transistor to switch between full-on and full-off
at anywhere near that speed.

I measured the current at point X and it seems high - over 100mA.

From the Wikipedia article, the 2N2222 should be able to
handle that, so I don't think it's a concern, provided the
transistor is switching efficiently.

I also tried a honking big "high speed switching" transistor pulled
out of a PSU but it made the LEDs go very dim.

If it's designed to handle very high currents, it may not
work very efficiently at the modest power levels you're
concerned with here.

From what Jon said, it sounds like the 2N2222 should be
capable of the job as long as you don't try to run it
at more than 100kHz or so.

You really need to look at things with a scope and find
out what's going on. Just fiddling randomly with the
parameters is likely to lead to a lot of frustration and
dead transistors.

--
Greg

 

[greg]

fungus wrote:

I don't see
how the number of turns would be related to the transistor
temperature.

Increasing the number of turns increases the inductance,
so the current rises at a lower rate when the transistor
is on. Assuming the base current remains the same (which
it will be if you increase the number of base winding
turns by the same amount and keep the base resistor the
same) then more time will be taken for the collector
current to reach its maximum value and trigger a turn-off.

This means that the operating frequency will be lower,
and the switching time of the transistor will be a
smaller fraction of the on-time, leading to less power
dissipation and a cooler transistor.

That's the theory, anyway.

--
Greg