Switching transistor using radiant current spike?

[sucahyo]

Here is a dissapointing attempt to switch transistor using radiant spike. I use the radiant output of radiant oscillator to switch a transistor. In isolated circuit this transistor used to produce radiant current too. The result is dissapointing, the secondary circuit output is much less than the main circuit. secondary produce 5V output using 0.6A input at 11.5V, while the main circuit produce 200V output using 0.7A input at 10.5V. I don't know if signal quality is different if mesured with a scope.

YouTube - Switching transistor of radiant circuit with radiant current

[elias]

What a synchronicity

This is an idea which traversed my mind yesterday.

[sucahyo]

This is an idea which traversed my mind yesterday.

Wow, I guess mind can really heard others who thinking out loud. Remind me of Dale Pond sympathetic vibration theory.

[nvisser]

Why not try to use a small capasitor between base and emmiter to try and get some energy on the base to switch it?

[sucahyo]

Why not try to use a small capasitor between base and emmiter to try and get some energy on the base to switch it?

Maybe it would improve the power, but I think also make the transistor shut off slower. I would choose to add second transistor.

I am thinking that switching using radiant current has very fast switched off. Since I don't have a scope I just based my conclution on radiant output and feeling. From feeling (temperature etc.) it feels the same as direct 555 driven one. I forgot that transistor base to emitter may have resistance too, and coil collpase speed depend much on load resistance.

[Inquorate]

Transistor

Since general transistors switch based on current (i think) you will probably have more luck switching with a field effect transistor which switches due to voltage.

Otherwise, a 2n2222 will probably provide enough of a signal to switch a 2n3055.

Field-effect transistor

Love and light

[sucahyo]

Thanks, unfortunately I still has no luck in using FET. Still wondering wether I put it backward or not. How do you connect TO-220 package? is it the same as 3055?

If 555 fall time is faster with high resistance output, I guess I may use 4 transistor in cascade.

[Inquorate]

How to

Search for the part number here, ie 2n2222 (not the package case number, ie t0-220) and the data sheet will say which one is the gate or base, etc.

ALLDATASHEET.COM - Datasheet search site, Datasheet search site for Electronic Components and Semiconductors and other semiconductors.

Otherwise I usually do a google search for the part #, and the word 'schematic' to see how it is used.

Love and light

[jeanna]

Hi sucahyo,

I tried that and I think the problem is that the switching transistor of the joule thief (your second circuit) needs a steady current/voltage for the coils to have an effect on the transistor.
I make a long series of tests and learned that a very effective way to increase the output is to add inductor coils IN SERIES with the wire that is feeding the pulses.
So, you only need one set of pulses which you can vary later.
A capacitor helps too, but the inductor is the more effective element.

edit: I think what I am saying is that you already have what you need because you have the pulses. We all need to find a way to make those pulses do more and more work which I believe I showed is possible.

I made some videos about this after I started to get results.
here is one
[URL="http://www.youtube.com/watch?v=BztnQB2sgg0"[/URL]
and here is another from my more recent success. also, please remember, I purposely kept the input small because I wanted to see the effects of my actions.
[URL="http://www.youtube.com/watch?v=XzhbsLBwc54"[/URL]

jeanna

[boguslaw]

Use mosfets and transient diodes.Use high self-inductance coil(s) to store energy when mosfet is on, then recover it into capacitor when mosfet is off bypassing mosfet using transient diode.Then discharge capacitor into other coil and catch vibration as resonant wave on secondary with stationary electric wave bouncing of secondary ends.Each discharge then accumulate on secondary.
I feel it is possible , just I need to work out the correct circuit.
The key is in synchronization of all parts. Coil has to store enough power in magnetic field to completely recharge capacitor in single inductive discharge.
The goal is to have many many thousands disruptive discharges of capacitor per second. Tesla did it without troubles but he choose mechanical interrupter and high voltage capacitors. That was easy, because such capacitors can be rather small in capacitance while storing enough power.

[sucahyo]

Search for the part number here, ie 2n2222 (not the package case number, ie t0-220) and the data sheet will say which one is the gate or base, etc.

I have some datasheet, what puzzle me is whenever I replace my transistor with FET, the FET just heat up and do not switch. I figure this maybe the same reason for why my previous two 555 died, because I have to limit something. I still don't know what is wrong.

I tried that and I think the problem is that the switching transistor of the joule thief (your second circuit) needs a steady current/voltage for the coils to have an effect on the transistor.

Ok.

I make a long series of tests and learned that a very effective way to increase the output is to add inductor coils IN SERIES with the wire that is feeding the pulses.

I have coil with no induction, I think I will try it with my circuit while charging nicad, thanks for the idea .

Just try no induction coil and 24V transformer, I don't see the difference in output. Maybe because I use 10Kohm base resistance.

Since I use the circuit for charging battery, efficiency is not my primary goal. More output may not be better in my case. My circuit run between 0.3A to 2A at 10.5V. The output currently between 0.1A to 1A. The primary voltage is between 1V to 200V limited by spark shorting out the coil secondary.

I judge the signal characteristic with heat from the nicad. I can do 10 minutes charging nicad but with considerable heat, this is not what I want. I prefer 1 to 2 hour of charging without any heat detected on the battery.

My requirement
- The transformer (my coil) must be cool at all time
- The battery being charged must not be warmer than body temperature. Easily detected on NiCd or NiMH.
- cool 555
- cooling my cup of water
- optional cool switching transistor & resistor


Some experience when increasing output of my primary circuit:
- if I reduce duty cycle (also increasing frequency), the output increase but the transistor and transformer become heat up significantly. Reducing duty cycle will also increase heat.
- If I remove intermediate transistor and use 555 output to directly drive the switching transistor, the output increase significantly but made the charged battery heat up significantly too.
- I currently use 10K base resistance, using 100 ohm would increase the output to 1Amp but will heat up even 12V battery.

I made some videos about this after I started to get results.
here is one
YouTube - Joule Thief with multiple Secondaries Explanation Part 4 Diagram
and here is another from my more recent success. also, please remember, I purposely kept the input small because I wanted to see the effects of my actions.
YouTube - Joule Thief with 2 Secondaries Explanation Part 6

Thanks, that is very efficient design . Did you use radiantly charged battery for that video?

BTW, you use wrong format for posting link. If you just paste the link it would show properly since the forum software will format it automatically.

Use mosfets and transient diodes.Use high self-inductance coil(s) to store energy when mosfet is on, then recover it into capacitor when mosfet is off bypassing mosfet using transient diode.Then discharge capacitor into other coil and catch vibration as resonant wave on secondary with stationary electric wave bouncing of secondary ends.Each discharge then accumulate on secondary.

Did you mean that we dump the capacitor back to the coil with the help of it's secondary when the MOSFET start to turn on?

[jeanna]

I judge the signal characteristic with heat from the nicad. I can do 10 minutes charging nicad but with considerable heat, this is not what I want. I prefer 1 to 2 hour of charging without any heat detected on the battery.

Is this not normal for fast charging?
I have a wall charger that can do a fast charge and sometimes the batteries are too hot to touch.

Some experience when increasing output of my primary circuit:
- if I reduce duty cycle (also increasing frequency), the output increase but the transistor and transformer become heat up significantly. Reducing duty cycle will also increase heat.
- If I remove intermediate transistor and use 555 output to directly drive the switching transistor, the output increase significantly but made the charged battery heat up significantly too.
- I currently use 10K base resistance, using 100 ohm would increase the output to 1Amp but will heat up even 12V battery.

This experience sounds right to me.

Maybe it is just this part where you return the spikes to the basic circuit.
I am just starting this part of the journey and have no experience, but after I get my bedini to work from the basic recipe I want to see if I can get it from just the secondary and continue to leave that primary completely alone.

Did you use radiantly charged battery for that video?

No, just a NiMH AAA with 1.2v

BTW, you use wrong format for posting link. If you just paste the link it would show properly since the forum software will format it automatically.

I will try it right here, if I may. It took me 2 days to find this thread... I may not find it again!

here is a try at a link:

http://http://www.youtube.com/result...rch_type=&aq=f


(It looks wrong to me. Thanks for the tip. )

[edit] No, it adds an extra http and firefox cannot find it I must delete the http:// from my copy before I paste it

Is there a way to find my way to the threads I have posted on? I have not yet found a way??

thank you,

jeanna

[sucahyo]

Is this not normal for fast charging?
I have a wall charger that can do a fast charge and sometimes the batteries are too hot to touch

Normal 1 hour charger will make battery hot since it would charge the battery with 2600mAh current. The input of my circuit is 700mAh at 10.5V, and yet it can charge 2500mAh NiMH or 1000mAh nicad under one hour.

link given by druzara from rcgroups, nicad spec:
http://battery.sanyo.com/en/spec/cadnica/N-1700SCR.pdf

Maybe it is just this part where you return the spikes to the basic circuit.
I am just starting this part of the journey and have no experience, but after I get my bedini to work from the basic recipe I want to see if I can get it from just the secondary and continue to leave that primary completely alone.

Ok. That would be interesting to explore. I think that kind of arrangement may work best with ferrite and air core should be avoided to improve efficiency.

[edit] No, it adds an extra http and firefox cannot find it I must delete the http:// from my copy before I paste it

Copy paste along with the text, not after you click insert url.

A try out.
YouTube - jeannacav

edit: it work, I use quoted reply and paste the link directly bellow "a try out" text. If you paste it after you click post link, delete the "http://" text first.

Is there a way to find my way to the threads I have posted on? I have not yet found a way??

There is search button on top navigation. "Advanced Search" with "exact name" of yours, "show results as post".

[jeanna]

The input of my circuit is 700mAh at 10.5V, and yet it can charge 2500mAh NiMH or 1000mAh nicad under one hour.

That is terrific.
Do you happen to know the frequency and the volts that are being presented to the battery?
I noticed a while ago that when you raise the frequency the volts go down. Maybe raising the frequency could slow down the heating.?

I think that kind of arrangement may work best with ferrite and air core should be avoided to improve efficiency.

Yes, you are probably right about this.


And, thank you a lot for your helping me to move around this forum.

jeanna

[sucahyo]

Do you happen to know the frequency and the volts that are being presented to the battery?
I noticed a while ago that when you raise the frequency the volts go down. Maybe raising the frequency could slow down the heating.?

I can't measure the frequency, but from value it calculate as 9883Hz. The current for one hour charging is around 300mA with voltage 0.1V more than battery voltage.

For frequency I think it depend on the coil. When I replicate joule thief at 12V I found that there is sweet spot frequency. Reducing frequency increase output, but there is spesific frequency (potentiometer value) where input current and output current ratio is best. Reducing frequency bellow this value would increase input but will decrease output. Increasing frequency above this value would reduce output and decrease output.

The heating seems related with duty cycle, maybe frequency too but my previous around 2000Hz circuit also has cool coil using different resistor and capacitor value.

Glad I can help .