Here I show you the concept:
https://youtu.be/3ju0WJfpwso
In response to criticism over some of my reactive circuits, I have put together a working demonstration that shows how we can achieve the same results through alternative methods. The key is to limit the current to its lowest necessary value to save on energy usage. The previous x-reactance method was just one approach among many.
In this demonstration, I use a low voltage square wave generator that operates at around 5 volts to trigger a base resistor of an NPN transistor. The transistor switches a 9-volt battery into a home-built 1.9 ohm air core coil, quickly wound with telephone cable of about 300 feet. We use diodes Bedini-style to feed the high voltage back EMF and quickly charge a 10uF capacitor to 100 volts with very little current. The sharp low 10% duty cycle holds back much of the current from leaving the 9-volt battery while producing loads of back EMF for quick charging and discharging of the 10uF capacitor at 100 volts, thanks to the SCR neon capacitor dump. We store this charge into a set of 12-volt supercapacitors, giving us steady DC in return, enough to run a small high-frequency inverter and drive the lamp at full lumens, which normally requires 15 watts at 60 hertz AC.
This circuit is more complicated, but it proves the concept further. It shows what can be done with very little current if utilized correctly. It's clear that this little 9-volt battery cannot produce "15" watts on its own without the help of additional systems such as pulsing, electrets, supercaps, high frequency, etc., thanks to running the circuit mostly as an open loop.
Thanks for the insight, Joel.
My question may probably sound stupid, but why is it needed to have a secondary capacitor (a capacitor bank in this case)? Why didn't you use the first capacitor as a source for the inverter? Why didn't you use just one capacitor bank as the first capacitor?
My confusion comes from the following: You have a very low 10 Hz frequency while the output lamp does not flicker which means at the end you have enough power for each impulse for the lamp which is usually at least 40-50 Hz.
To me it sounds like you are filling a bucket of water for 1 second and then you are pouring it into a container that has a volume of 6 buckets but your load requires pulses of 4 buckets every second. Your example works which means my math is probably wrong. I've tried to discharge one capacitor into another and the difference is the higher current I get from the discharged capacitor which charges the other one in an instant. T. H. Moray says that he discharges capacitors into others, but I still can't see what's the advantage besides the quicker charge time.
Again, sorry if my question sounds too stupid.
By the way, I've tried to use a simple Tesla transformer output by rectifying it, storing it into a capacitor and discharging it through a gas spark gap (I'm using a 90 V arrester like this one:
https://ae01.alicdn.com/kf/HTB1odbyafjsK....jpg_.webp ). The discharged current goes through a small 5 W incandescent car light bulb. It flashes from time to time, but the frequency is rather slow (once every 4-5 seconds). The Tesla transformer was deliberately low powered (9V; 80 mA) in order to see how much I can squeeze out of it. The voltage was about 1.5 kV. I've ordered a few neon lights to see if discharging through them would be different.
I'm probably comparing apples to oranges right now, because with the Tesla transformer I got almost no current and a sinusoidal wave (but at very high frequency) while with the back-EMF peaks there is more current, even though the voltage is much lower (100 V in your case). The higher current is probably the difference.
I'm still wondering why you didn't discharge the first capacitor through the invertor but used an intermediate capacitor dump which was with a very low frequency.
Hi Joel.
Thanks for this info and video!
I don't think you realise just how important this is!
You showed your cap bank charging while running a mini inverter under a load.
I've been trying to do this for months with no success... Just batteries that end up needing recharging.
Pretty sure I get over 1000 volts BEMF from my 12 volt joule thief, so not sure why we need the extra cap to pump the driving cap bank..
Unless it is to help increase amperage??
Anyway, I was surprised to find the signal generator for only $2.52 NZD, so ordered 2, plus neon bulbs, scr's and gas discharge tubes..
Can't wait for them to arrive to try this out!!
Also will feed back to the start up battery to have a self runner... (As you showed too much voltage growing..)
You're doing a great job sharing this info Joel!
Here's the link to the signal generator for anyone looking or it..
https://vi.aliexpress.com/item/100500335...9385473%21
it's a public holiday here today. So what better to do than work on this??
So now I have my Joule Thief powered by an old cell phone charger..
That's only 4 volts.. filling 2x 25 volt, 47000 uf caps in parallel.. Running the 100 watt mini inverter, which lights the 220 volt 7 watt LED..
The caps voltage is maintained at 11 volts...
Next step is to try and make it self looping...
[
attachment=278]
[
attachment=279]
ChatGPT admits this system could be self sustaining
I asked if I could add an inductor in series with the bulb, add a secondary to extract 4 volts to store in a cap and run the joule thief from that...
if I stored the 4 volts from the secondary in a capacitor and used it to run the joule thief, would that not work? I could do away with the cell phone charger and touch a battery to the capacitor to start it..
ChatGPT
Storing the 4 volts generated by the secondary coil in a capacitor and using that stored energy to run a joule thief or another low-power circuit is a feasible approach. In this setup, you could potentially eliminate the need for the cell phone charger as a continuous power source and use the capacitor to kickstart the system when needed. Here's how it might work:
Charging the Capacitor: The 4-volt output from the secondary coil can be used to charge a capacitor. The capacitor stores the electrical energy until it reaches a sufficient charge level.
Joule Thief Circuit: The joule thief circuit can be powered by the energy stored in the capacitor. When the voltage across the capacitor reaches a certain threshold, you can use a switching mechanism (like a transistor) to enable the joule thief circuit, which can then run the 220V to 4V conversion process.
External Battery Kickstart: If you need to start the system, you can touch a separate battery to the capacitor to initiate the charging process. This would be a manual kickstart to begin the energy storage.
Keep in mind the following considerations:
The capacity (in farads) of the capacitor will determine how much energy it can store and how long the joule thief circuit can operate without an external source.
You'll need to ensure the joule thief circuit can efficiently convert the capacitor's voltage to run your 220V to 4V conversion when enabled.
The efficiency of this setup may still be relatively low, as energy storage and conversion processes often involve some losses.
Managing the kickstart process with an external battery requires manual intervention. To make it fully self-sustaining, you would need to explore mechanisms to automatically initiate the charging process when the stored energy drops below a certain level.
It's an interesting experiment, but developing a reliable self-sustaining energy system with high efficiency can be quite challenging. You may need to iterate and fine-tune your design to optimize its performance.
Well my parts finally arrived..
But then I realised that there is no schematic for Joel's setup.
I'm no expert and can't figure it out from the video, so I can't progress any further unless someone has a schematic they can post??
(10-23-2023, 08:20 AM)MrTD Wrote: [ -> ]it's a public holiday here today. So what better to do than work on this??
So now I have my Joule Thief powered by an old cell phone charger..
That's only 4 volts.. filling 2x 25 volt, 47000 uf caps in parallel.. Running the 100 watt mini inverter, which lights the 220 volt 7 watt LED..
The caps voltage is maintained at 11 volts...
Next step is to try and make it self looping...
Thank you very much for posting your design, i found your design of using joule thiefs back emf very easy to build and efficient, but joels version was necessary for us to understand it completely. i think using a hvhf generator with an av plug to charge caps is hard compared to using joule thiefs back emf. this is by far the easiest way to extract more amps with just 4v. thanks again.