Storing Cookies (See : http://ec.europa.eu/ipg/basics/legal/cookies/index_en.htm ) help us to bring you our services at overunity.com . If you use this website and our services you declare yourself okay with using cookies .More Infos here:
https://overunity.com/5553/privacy-policy/
If you do not agree with storing cookies, please LEAVE this website now. From the 25th of May 2018, every existing user has to accept the GDPR agreement at first login. If a user is unwilling to accept the GDPR, he should email us and request to erase his account. Many thanks for your understanding

User Menu

Custom Search

Author Topic: Joule Thief 101  (Read 926680 times)

MileHigh

  • Hero Member
  • *****
  • Posts: 7600
Re: Joule Thief 101
« Reply #105 on: February 05, 2016, 11:35:25 PM »
I think the problem is that any investigation into a JT circuit should be done in two steps.  The first step should be to understand how it operates in normal mode.  The second step is then to explore any possible self-resonant modes.  However, if it is running in some kind of self-resonant mode, then it is not really a JT any more, it's an oscillator.

Then there is another problem.  If you take a JT circuit and play with component values and turn it into an oscillator, then your claims of it running better and longer than a regular JT circuit would have to be proven on the bench.  A transistor operating in the linear region means that the transistor is acting like a resistor and continuously dissipating power.  Likewise the LED is continuously dissipating power.  One would think that the JT has an advantage here because it is a switching circuit where for most of the duty cycle the LED and transistor are not dissipating power.  Presumably oscillator operation demands that the battery still be capable of outputting some minimum voltage under load, whereas the whole idea behind the JT is that it can operate at very low minimum battery voltages, presumably lower than that of the comparable oscillator.  So the proof has to be in the pudding.

Also to be more accurate, most of the time the transistor is switched ON in a JT circuit and it is energizing the main coil and dissipating power.  However during this time the power consumption of the input side of the transistor is quite low.  Then when the transistor switches OFF the main coil dumps its stored energy though the LED to light it up.   So there is definitely near-continuous power dissipation associated with a JT circuit.

I always thought an interesting comparison would be between a CMOS 555 timer circuit with very carefully selected components lighting an LED and a JT circuit.  The CMOS 555 timer circuit can't operate at the very low voltages of a JT circuit but it probably would have a lower power overhead to keep it operating compared to a JT circuit.  You wonder which circuit would give you better overall performance in the long run.

Pirate88179

  • elite_member
  • Hero Member
  • ******
  • Posts: 8366
Re: Joule Thief 101
« Reply #106 on: February 06, 2016, 03:13:10 AM »
I think the problem is that any investigation into a JT circuit should be done in two steps.  The first step should be to understand how it operates in normal mode.  The second step is then to explore any possible self-resonant modes.  However, if it is running in some kind of self-resonant mode, then it is not really a JT any more, it's an oscillator.

Then there is another problem.  If you take a JT circuit and play with component values and turn it into an oscillator, then your claims of it running better and longer than a regular JT circuit would have to be proven on the bench.  A transistor operating in the linear region means that the transistor is acting like a resistor and continuously dissipating power.  Likewise the LED is continuously dissipating power.  One would think that the JT has an advantage here because it is a switching circuit where for most of the duty cycle the LED and transistor are not dissipating power.  Presumably oscillator operation demands that the battery still be capable of outputting some minimum voltage under load, whereas the whole idea behind the JT is that it can operate at very low minimum battery voltages, presumably lower than that of the comparable oscillator.  So the proof has to be in the pudding.

Also to be more accurate, most of the time the transistor is switched ON in a JT circuit and it is energizing the main coil and dissipating power.  However during this time the power consumption of the input side of the transistor is quite low.  Then when the transistor switches OFF the main coil dumps its stored energy though the LED to light it up.   So there is definitely near-continuous power dissipation associated with a JT circuit.

I always thought an interesting comparison would be between a CMOS 555 timer circuit with very carefully selected components lighting an LED and a JT circuit.  The CMOS 555 timer circuit can't operate at the very low voltages of a JT circuit but it probably would have a lower power overhead to keep it operating compared to a JT circuit.  You wonder which circuit would give you better overall performance in the long run.

I think a decent JT circuit would win this competition with the 555.  ONLY because it takes some energy to run the 555...and that is the only reason.
There is no magic to the JT BUT, we who experimented with them in the early days were told that conventional electronic theory explained them.  This, of course, is true.  My complaint was, then, why were they not being used in our electronics stuff?  Shortly thereafter, we saw those led garden lights, as well as many other items that actually were using a JT type circuit to its advantages.  Now I am happy.  Now we have chips that do this with minimum input and they are being used in everyday devices.  This answers my question from back then.

Still, no magic, no free lunch, no output more than input...just a good way to use most of the energy in a battery and get more light from leds than you otherwise could.

Damn MH, you and I used to argue about this all of the time and now I have to admit that you were right.  There is no "magic".
Son of a bitch, ha ha.

Bill

I still have not seen anyone else light 400 leds from a 'dead" AA battery like I have done.  No magic there either, just the right circuit for the desired outcome.


TinselKoala

  • Hero Member
  • *****
  • Posts: 13958
Re: Joule Thief 101
« Reply #107 on: February 06, 2016, 06:56:48 AM »

sm0ky2

  • Hero Member
  • *****
  • Posts: 3948
Re: Joule Thief 101
« Reply #108 on: February 06, 2016, 01:08:34 PM »

Also to be more accurate, most of the time the transistor is switched ON in a JT circuit and it is energizing the main coil and dissipating power. 

You are forgetting the reluctance factor of the inductor in self-resonance.

MileHigh

  • Hero Member
  • *****
  • Posts: 7600
Re: Joule Thief 101
« Reply #109 on: February 06, 2016, 05:00:47 PM »
You are forgetting the reluctance factor of the inductor in self-resonance.

I am not sure of what you mean by that.  The more information the better when discussing electronics.

MileHigh

  • Hero Member
  • *****
  • Posts: 7600
Re: Joule Thief 101
« Reply #110 on: February 06, 2016, 05:10:40 PM »
I think a decent JT circuit would win this competition with the 555.  ONLY because it takes some energy to run the 555...and that is the only reason.

I am assuming that it would take tens of microwatts to run the 555 including the carefully selected timing components (the "overhead" not counting flashing the LED) and the equivalent overhead for the JT would be on the order of milliwatts.  So if you are talking about very long run times it may be an interesting competition.  Of course you would have to put the actual LED flashing on a level playing field.

There was a chip from the mid 1970s, the LM3909, that was also a very efficient LED flasher.  You can look it up.  In fact, it looks to me like it may be more efficient because it doesn't use any inductive components.  It might make for a three-way horse race.

TinselKoala

  • Hero Member
  • *****
  • Posts: 13958
Re: Joule Thief 101
« Reply #111 on: February 06, 2016, 11:55:16 PM »
What are they made of, gold and diamond dust? The cheapest I can find LM3909 is around 4 dollars each, from china, and even in quantity. Some people want much more than that. My favorite Chinese vendor UTSource wants $31.25 for a lot of 5 (but has another listing for 1 for $4.00). Go figure.

A couple of sellers in the UK want over $17 for _one_ (but with free shipping.)

So I'm not going to be testing one any time soon, unless someone else donates one or two to me.

TinselKoala

  • Hero Member
  • *****
  • Posts: 13958
Re: Joule Thief 101
« Reply #112 on: February 07, 2016, 12:32:12 AM »
Rob Paisley has "reverse engineered" the LM3909 so that it can be simulated with discrete components.

http://home.cogeco.ca/~rpaisley4/LM3909.html

Looks like an interesting project, but I still don't think it will compete with a JT. The Data Sheet for the LM3909 gives an "LED Boost" circuit that will drive an LED from 1.5 volt battery at 2 kHz... but it says it draws 4 mA.  Presumably that is an average, or equivalent-continuous, current drain. In which case, unless the LED is unusually brilliant.... well, it's not really that great.

Pirate88179

  • elite_member
  • Hero Member
  • ******
  • Posts: 8366
Re: Joule Thief 101
« Reply #113 on: February 07, 2016, 05:57:19 AM »
http://www.youtube.com/watch?v=wU5x8T2UkuI

 8)

Nice video TK, wow that is really small.  Very cool.

https://www.youtube.com/watch?v=iHmTc0PwiyY

Here is a vid from Ludic Science making a JT without a transistor.  He said he is replicating Lidmotor's circuit...I had forgotten about that one  This is pretty simple and cool.  I suppose it is really just a very simple and basic relay/solenoid  type device but man, I would have never thought of this in a million years.

Bill

Here is Lidmotor's original video:  https://www.youtube.com/watch?v=VjqBRXU3XnU

TinselKoala

  • Hero Member
  • *****
  • Posts: 13958
Re: Joule Thief 101
« Reply #114 on: February 07, 2016, 08:27:16 AM »
Even smaller:


sm0ky2

  • Hero Member
  • *****
  • Posts: 3948
Re: Joule Thief 101
« Reply #115 on: February 07, 2016, 09:45:47 AM »
I am not sure of what you mean by that.  The more information the better when discussing electronics.


ummm

ok

Ampere Turns per Weber
Reluctance (R) = MagnetoMotive Force (Ampere-turns) divided by Flux in Webers.
Hopkinson's law

Length of the wire divided by the cross-sectional area times the magnetic permeability of the material....

It translates directly to Henries, of magnitude, and in self-resonance, reluctance becomes = 0.
The capacitive counterpart also disappears and the capacitor itself takes on a purely inductive behavior.

https://translate.google.com/translate?hl=en&sl=zh-TW&u=http://140.114.17.97/circuit/ch14.htm&prev=search


In the future if you want to play ignorant with me,.. please go back and delete all your posts where you discuss the details of that which you claim to not know......

MileHigh

  • Hero Member
  • *****
  • Posts: 7600
Re: Joule Thief 101
« Reply #116 on: February 07, 2016, 11:45:01 AM »
No, I am simply asking you what kind of point you are trying to make, I was not asking you for definitions.  I made a point that the JT circuit was basically continuously drawing power from the supply battery under normal operation and your response was, "You are forgetting the reluctance factor of the inductor in self-resonance."

What kind of a connection are you making and what kind of point are you trying to make?  I know what self-resonance of a coil is but I don't know what you mean by "reluctance factor" here.  I know what the reluctance of a magnetic flux path is.  A coil with a core material in self-resonance will act as an AC short-circuit and you are left with the DC resistance of the wire.  Presumably the core material will also burn off a certain amount of power due to hysteresis.

If I assume a standard JT circuit but no explanation for how it is operating as an oscillator and how self-resonance of the coil ties into all of this and what a "reluctance factor" is and having no timing diagrams, I am simply not sure what you are meaning and what kind of point you are trying to get across.

Call it a little pet peeve of mine if you want, but the forums have thousands of "discussions" about electronics by people with limited knowledge of electronics with no schematics to reference and no timing diagrams.  In almost every case they are pretentious nonsense discussions that don't really mean anything and are in essence unworkable wild speculation in the from of a fake back-and-forth dialogue between two posters that is disconnected from the reality of the circuit.

tinman

  • Hero Member
  • *****
  • Posts: 5365

Pirate88179

  • elite_member
  • Hero Member
  • ******
  • Posts: 8366
Re: Joule Thief 101
« Reply #118 on: February 07, 2016, 03:35:40 PM »
https://www.youtube.com/watch?v=ekPh9p4YECE

Brad:

Nice.  I have about 40 of those led garden light circuit boards lying around here, those chips they use make a decent JT.
Is that button cell 3 volts or 1.5?  Will it run those down to low voltage as well?

Bill

sm0ky2

  • Hero Member
  • *****
  • Posts: 3948
Re: Joule Thief 101
« Reply #119 on: February 07, 2016, 06:25:26 PM »
No, I am simply asking you what kind of point you are trying to make, I was not asking you for definitions.  I made a point that the JT circuit was basically continuously drawing power from the supply battery under normal operation and your response was, "You are forgetting the reluctance factor of the inductor in self-resonance."

What kind of a connection are you making and what kind of point are you trying to make?  I know what self-resonance of a coil is but I don't know what you mean by "reluctance factor" here.  I know what the reluctance of a magnetic flux path is.  A coil with a core material in self-resonance will act as an AC short-circuit and you are left with the DC resistance of the wire.  Presumably the core material will also burn off a certain amount of power due to hysteresis.
yes. That IS the point I am trying to make. magnetic reluctance becomes non-effective in the circuit. There is no effective "resistance" placed on the wire by the induction of the core material.
Yet, the core material still becomes energized, and the resultant field collapse represents itself through the coils inductance. Hysteresis is minimized in resonant operation of the core. The flux graph is sinusoidal as well, 90-degrees to the electric.

Quote

If I assume a standard JT circuit but no explanation for how it is operating as an oscillator and how self-resonance of the coil ties into all of this and what a "reluctance factor" is and having no timing diagrams, I am simply not sure what you are meaning and what kind of point you are trying to get across.

Call it a little pet peeve of mine if you want, but the forums have thousands of "discussions" about electronics by people with limited knowledge of electronics with no schematics to reference and no timing diagrams.  In almost every case they are pretentious nonsense discussions that don't really mean anything and are in essence unworkable wild speculation in the from of a fake back-and-forth dialogue between two posters that is disconnected from the reality of the circuit.

to that I agree.

The "reality" of the circuit, was presented by Edwin Armstrong in 1912. This is designed to be a resonant tank circuit. It was used exclusively in the Steven Mark TPU, as the prime exciter.
The same technology which powered radios in the 1940's before we had commercial batteries.
these radios (although not very loud) required no external power source. Not only that, they were well known at the time, for building up radio interference waves that would disrupt signals for miles around the device. Which ultimately led to its' replacement by a less invasive technology.

"Constructive Interference", "Positive Feedback"

If you somehow missed this point during the 136,000 pages of Joule Thief discussion, then I apologize for jumping on your case. I assumed you already knew what we were talking about here.
-------------------------------------------------------------------------------------------------------------------

Modern day "joule thieves" place switching transistors in digital mode with inefficient diodes, successfully destroying the resonant effect.

There is still "some effect", because of the natural SRF of the circuit being the dominant factor between the inductor and the tank. but it is disrupted during each cycle, thus a heavier drain on the source than a resonant LRC would or should represent in ideal operation.

This is why a modern day analysis of a JT circuit, observes a wide range of inefficiencies in the circuit.

It is a simple concept, which Americans are indoctrinated to NOT observe.
they teach us these things are bad in circuits, and every way to get RID of this effect.
simply reverse your training to do the opposite.

invite these extra energy levels to build up all they want to :)
like strumming a string over the resonant cavity of the guitar.
-------------------------------------------------------------------------------------------------------------------
Here are two images of transistors switching. the one on the left is a transistor operating outside of linear mode. There no be resonance when the transistor scope shot looks like this.

The image on the right shows a transistor operating in the range of its' linear mode of operation. due to the particular circuit it is in, it is biased slightly below the actual linear value, but still within the range.
Notice, that the signal shown on the scope is the input signal, NOT the transistor function.