This is a reply to Paul’s question.
After all the experimenting you have done since this was first designed, do you still favour a SJT over a conventional JT routinely, or can the extra components be circumvented by a different coil design or optimised transistor specification.
The SJT has the feedback winding connected to the circuit, whereas the conventional JT has the winding connected to the positive, which allows it to be connected to a controlling circuit such as the one in my Blue Blinky. The SJT can’t be controlled like that , however I have the SJT flasher. Point is that the SJT has some limitations that the conventional JT doesnt’t have. For a non-experimental use (flashlight or torch) the SJT can to a better job than the conventional. But I often want to experiment and then I find that the SJT circuit gets in the way, so to speak. So I often use the conventional JT for experimenting.
The components broadly tune the SJT so they can’t be circumvented as far as I know. The transistor may help but as far as I know it doesn’t change the SJT’s need for the capacitor to broadly tune it. I’ve used various coils and the SJT circuit works with most of them. The air core coil seems to have some quirkiness, if I remember correctly from Quantsuff’s website.
Some people like me drive a car because it’s a car. Some people prefer a Beemer over a Jaguar or vice versa. But in electronics, I guess I perceive some things in my own way that the average experimenter doesn’t. So perhaps, in electronics I should be compared to those that prefer a Jaguar over a Beemer, so to speak. I often try to help, but I get a lot of negative feedback from others that they don’t like or need my help. And they then burn up transistors by the gross. So let ’em keep making the same mistakes over and over again! I told you so! Sort of like the townspeoples’ opinion of Gandalf.
Sorry that was too philosophical an answer for me to understand. What would the answer be in terms of simple reliability, low voltage working, efficiency and power. As opposed to elegance, charm, élan and soul. I was trying to ask if the need for a SJT could in practice disappear, if you just used, for example the optimum transistor and coil windings in a JT, or does the SJT always seem to come out on top.
Hi,
Please take a look at the simulations made by me on my website http://flagiusz.republika.pl/joulethief/sim.html#sjt
Both actual measurements and their numerical simulations are the same.
The superiority of SJT over the traditional design of Joule Thief is a fact, and I fully agree with that.
However, a much simpler solution with accelerating capacitor and with no diode (as previously discussed on http://rustybolt.info/wordpress/?p=3131) is slightly better and a little more efficient then SJT.
I look forward to see your comparison results.
I have really enjoyed reading your blog.
You’re doing a great job!
Sincerely,
Flagiusz
Thank you for the comment. I’ll look at the info and see if I can get some good results.
i have done many experiments and modified the joule thief…Of them i hav e design that can extract power from a complete dead cell(about 0.45 volts)
but i want to build a better joule thief that will light a led for long time and maintain a constant brightness.can u help me?
I already have, by writing about many different Joule Thiefs in my blog.