Joule Thief

[resonanceman]

@resonanceman

Hello resonanceman,
I maybe interpreter your intentions wrong, but the way
i understand you, first of al you want to use a led as an overcharge
protection for a supercap.
This can be done but i woudn't go this way and i will tell
you why not. When you feedback the energy held by the supercap the led turns on,
this works great but the led will then take (depending on the led specs) 5mA up
to 20 mA (unless you take a low current one 2 mA).
So that is all loss. I would rather go for a zener in series with 2 parallel
resistors; one going to the negative of course and one going to the base
of a transistor or mosfet. This way you can controle the voltage level of the
supercap much more accurate and you will lose a lot less current
(for really low losses you can even darlington the transistors!!!).

The second i read from your story is that when the charging from the supercap
starts the input should stop (if i understand correctly).
This can be done quit easily; just place an extra transistor or mosfet
acting like a switch.

But now for the real problem; the way i can make up out of your words your JT
works just fine, but when you put a secondary winding on and you want to tap
energy of the JT crashes and doesn't work anymore!
Well this is a problem which sounds very familiar.
I can't tell you exactly what you should do because i don't know what kind of
circuit you're using, but i can explaine.
What probably is happening is you have a JT circuit with a secondary running
some leds well this works just fine but now you take out the leds and put in
a supercap therefor the capacitive load of the circuit changes and therefor the
resonance frequenty changes, when these two are out of sync the JT doesn't do
a thing (or crashes if you like).
You will have to see a JT circuit like a RCL circuit.
I would recomment to use koolers BJT circuit and use a 4k7 or 10k potentiometer
to tweak the circuit to find resonance.

Reagards,

Flux 4 Energizer

Flux4Energizer

I only have a few minutes to reply today.

I see  we have a little misscomunication here.

I disagree  that my  circuit would  be less efficient than yours.

Perhaps my usage  of the word crashing   caused some confusion.

I think  the biggest  difference  between my  circuit and yours..... ( your  hand drawn  circuit ) ......you interface  your circuit  with a chip......I  connect  mine directly to the JT
Most of  the parts of my circuit are similar to  your hand drawn circuit.....   
You have  your chip running all the time...my circuit is completely passive  until the U cap is nearly fully charged ........ then the excess power is used to crash or inhibit the efficiency of the charger

My explaining  about  the  the JT crashing when power is taken  off  of an extra primary is not  a problem I am having........ it is an observation  of an effect that I plan on  using  to control my JT
In other words.........using my transistor to ground one end of an extra primary winding should  reduce the output of the  JT or complertely  shut it down  depending on how strong the signal is.

  gary

[nievesoliveras]

Hi @resonanceman!

Are you going to post your circuit?

Jesus

[dllabarre]

Hi @resonanceman!

Are you going to post your circuit?

Jesus

resonanceman!

I'd like to see a copy of your circuit as well.

DonL

[jeanna]

sorry Jeanna,  I meant just using only 10-25 mA draw from the battery powering hopefully 10-15 bright LED's.

OK,
This is what I answered.
You should figure 7 spiky volts per led to start and this 60v output should give you what you are looking for.

I have found that the 10-25mA draw will be drawn when there are no leds or many. It is the basic joule thief that takes that, and the leds being 'excited' by the secondary seem not to affect the draw. (Sometimes, the draw seems to go down.)

Bottom line:
7 spiky volts per series led,
and
450 spiky volts minimum for a cfl

I hope that clears up what I am trying to say!

thank you,

jeanna

[Flux4Energizer]

Flux4Energizer

I only have a few minutes to reply today.

I see  we have a little misscomunication here.

I disagree  that my  circuit would  be less efficient than yours.

Perhaps my usage  of the word crashing   caused some confusion.

I think  the biggest  difference  between my  circuit and yours..... ( your  hand drawn  circuit ) ......you interface  your circuit  with a chip......I  connect  mine directly to the JT
Most of  the parts of my circuit are similar to  your hand drawn circuit.....   
You have  your chip running all the time...my circuit is completely passive  until the U cap is nearly fully charged ........ then the excess power is used to crash or inhibit the efficiency of the charger

My explaining  about  the  the JT crashing when power is taken  off  of an extra primary is not  a problem I am having........ it is an observation  of an effect that I plan on  using  to control my JT
In other words.........using my transistor to ground one end of an extra primary winding should  reduce the output of the  JT or complertely  shut it down  depending on how strong the signal is.

  gary

@resonanceman (Gary),

Oke, i see what you mean now.
Well i can be very brief in this case; the circuit you discribe won't work!!!
I understand where you are going with this but you oversee one little problem; as soon as you will use the red led as a
voltage protection for your 2.7 Volt supercap you will distroy your supercap!!!
You made a little miscalculation here; let me explain: the red led starts conducting when a voltage of 2 volts (or so) is reached.
This means you have a zero volt negative and a 2 volt positive. When you hook up a led directly on 12 Volt you will burn out the led right???
Well this isn't really true -> the led will burn out when +12V and -0V is used, BUT when +12 Volt is used as the positive and +10 Volt as a negative
the led will glow very nice and need (i think you get where i´m going at!?!).

Now in your circuit you'll have the drive battery (i think 1.2 up to 1.5 Volt) and the charge ultra cap (2.7 Volt as you said).
So in the worsed case your led will start conducting at 1.5 + 2.0 = 3.5 Volt.
This is why i say it will not work because it's not reliable.
It only protects your ultra cap if your drive battery is at 0.6 up to 0.7 volts!!!
There are still more reasons why this won't work but i will stop here because this is the main issue.

I hope you agree, but if not i would like to hear your thoughts about it.

P.S.
I don't use a zener diode in my JT circuit.
The circuit shown is a picprogrammer circuit JDM style!!!

Regards,

Flux 4 Energizer