Joule Thief

[TheNOP]

@all, My initial concern is that we have huge spikes in flux, and these translate to voltages beyond the capabilities of most transistors' V_ceo rating.  Then, allow me to ask again,

Since the juice is flowing through the led's or what have you, wouldn't it be safe to say that there's not to much danger to the transistor when we run a diode off the collector and use this path for the boosted voltage? ...wouldn't the extra components just sink the current and voltage? ...this is not in reference to setups with pickup coils.

V_ceo is not the only think to be wary about, V_be too.
transient surge voltage is often stated in transistor data sheets but not always.
the higher the transient surge the higher the transistor life time risk to be diminish.

i don't think we can use diode to protect the transistor in the way you and i would like.
it will simply get in the way, no boost.
in faq, only half protection can be achive, but at the expense of half the power.

i am sure of this, but don't take my words for it, test it yourself.

[altrez]

I have just made another JT and the basic circuit works just fine. I can light 10 LEDS from an old AA battery. I still can not for whatever reason get the pickup coil to work, even with 25 turns of 22 AWG mag wire. That took what felt like forever to wrap lol. I am not giving up, I plan on ordering some better toroids to see what happens.

What puzzles me is that everyone else has such great success with the pickup coil and I have yet to get a flicker from any pickup coil I have added. I have tested with 3 different types of wire and even with 6 different JT builds 1 turn up to 25 on the secondary no luck. 

Thanks for all the help!

[jadaro2600]

You may try twisting your two primary wires together, winding while spreading the turns evenly around the toroid.  Then place your pickup coils over this if you have to.  You may have a high permeability toroid.

What we need to use is low permeability - as someone pointed out a few pages back.

[xee2]

 jadaro2600

My initial concern is that we have huge spikes in flux, and these translate to voltages beyond the capabilities of most transistors' V_ceo rating.  Then, allow me to ask again,

Since the juice is flowing through the led's or what have you, wouldn't it be safe to say that there's not to much danger to the transistor when we run a diode off the collector and use this path for the boosted voltage? ...wouldn't the extra components just sink the current and voltage? ...this is not in reference to setups with pickup coils.

Yes. Putting an LED from the collector to the emitter will limit the spike to a few volts. This is fine for lighting LEDs located between collector and emitter. But the small spike voltage will not produce much voltage in a secondary coil. Using a neon will limit the spike to about 100 volts. This will also limit voltage produced in secondary but not by very much. Using a single diode is not good since the current will flow from battery through it instead of going through transistor because it only takes 0.7 volts for current to flow through it. Whereas, current will not flow through LED unless voltage is over 2.5 volts which only happens when there is a spike. However, three or more diodes in series could be used since then it would take over two volts for current to flow through them and battery voltage is only 1.5 volts. So current would only flow through them when there was a spike.

Using a secondary will allow you to get output voltages exceeding the Vceo of the transistor.

[dog812]

Hey thanks for the images about coils.. I see that it is a second coil on the same ferrite..

But what does the secondary coil do? Why add a second coil?