Yes, shunting the base bias to negative is not the most efficient way to turn off the JT. I do it several ways in my recent blog. You may have never thought that a red LED could turn off the JT. It's at rustybolt dot info and click on the link to my blog.
hi i need your help as i can see your very good with solar power lights things i need help i have a project that im doing for my back yard i need to make a solar power light with a lot more led kinda like a led Christmas light i need it to turn on at night and have a power switch also . how would i do this please help me if u dont fully under stand my question pm me and ill be more descriptive . thank you very much
Hi Jeri, interesting problem to solve. I had a go at redesigning it last night in LTSpiceIV and came up with the following circuit: screencast com/t/fuiEmTVyMg It is essentially a PNP BJT that switches off the center tap of the transformer when the solar cell is within battery - .65V - IR of the base resistor. Shottky from cell to batt. Adj. base resistor for better efficiency, I was really boosting the output. High gain BJT's or possibly FETs would be better. -Brett (FightCube)
I wonder if a Panasonic 1381/13811/13812 voltage detector would work in this circuit to replace the NPN on the left? They're used in a number of BEAM solar related robotic circuits. The 13812 has a CMOS inverted output in a TO-92 package.
@yellowmetalcyborg The trouble with outdoor solar night lights is that, unless they are in a secure area, unlike the wired variety, people can just walk up and pluck them from the ground like flowers. Perhaps someone will come up with a mounting stalk that can be driven into the ground and the, with a twist or some other mechanical trigger, cause spikes to radiate outward, anchoring them more permanently. (Oooh! I just thought of a way, but-I-won't-tell-you).
@CampKohler I was talking about supercapacitors for storing electrical charge, I have do idea how that relates to spikes and anchors.
Anyhow, here in the United States I'm allowed to shoot trespassers in my backyard. I just so happen to have a dart gun, so no need to make an anchor for nightlights. What a lovely country! (not).
@CampKohler I was commenting on the use of outdoor lights, per se, not the caps. But I don't think you could get much time from even whole-farad caps.
With respect to backyards, here in CA, it would get you prison time for sure unless you could show a threat to your safety; merely saving property would not justify shooting someone. (Put a knife in his hand!)
It's similar to what I'm working on for a school project. I used a bigger 6V solar panel, a battery from an old camera that is charged through a diode. The only problem is that I have to use a switch to turn the light on and off but looks like the simple methods are the best.
@stelian95 If you want to blow the everyone's minds, put a small window shade roller next to the solar cell and arrange a rope with a knob on it passing through a guide loop, so they only have to pull the knob to cover the cell, else it retracts. Now here's the good part: build a hidden switch (reed & magnet?) into the shade that operates the light. When they cover the cell, the light comes on! Only the truly geek will think that this is absolutly whack and start looking to find out why.
@stelian95 It's brilliant only if you pull the shade, otherwise it's dim. :-) Seriously, you could explain the shade by saying that it is there to simulate nighttime and that the light is not intended to come on during the day, which would waste the charge stored. What they will not be able to figure out from your schematic (if you leave the switch part out) is how stopping the charge causes the light to come on. The secret goal is to see if they are smart enough to ask that question.
@CampKohler And now for the real fun. Assuming you can stand nearby the project, you keep notes on what visitors do with the shade, how they react and if they ask "the question." The project is not the dumb light, but the results of the statistics about the visitors' behavior. In other words, it's a psychology project, not a science project! So, even if the light is a total bust, you tell the teacher(s) what the REAL project is and they will think YOU are brilliant. Sneaky is as sneaky does!
You can burn a bit less current keeping the switcher off by reducing it's base bias needs, it need only be biased enough to start up if the oscillator bootstraps itself. Maybe bypass part of the bias resistor with capacitance?
@jeriellsworth another idea would be by cutting the supply of the inductor(the center tap) with pnp and voltage divider on the base so that it would start only below threshold voltage. hope youget the idea. :)
@vk2zay Thanks for the tips. I have a similiar circuit in one of my projects and what I did to kick start the oscillator is to use an CR integrator to trigger a pulse on the gate of an SCR. The SCR keeps the load to turn on as long as the PV cell is blocked. Once the load current drops below the holding current of the SCR (light is on) the entire circuit shuts off and no leakage is measured. Im not sure will it help, Jeriellsowrth. Anyway, I will also try to play with vk2zay's idea. Thanks!
I just wish nano solar would hurry up and release their thin film nano ink panels. Then we could all enjoy cheap abundant energy and stop polluting the enviroment, maybe even shed foriegn oil dependance
Nice project. Actually a correctly built JT circuit is very efficient. Some of mine run for months 24/7 on an AA. Dr. Jones has measured possibly COP 8 with his circuit on oudotcom. Others in our group have circuits that will run for a year. Very efficient and flexible little circuit. You can really get the amp draw very low with a little effort. Very nice video.
Jeri got me lookin'. If you Google "joule thief engine", you will find a puzzling (to me) circuit at beam-wiki that is supposed to cause an LED to blink, but I'll be ding-donged if I can see the power source. Anybody?
Here's an idea. Feel free to poke holes in it. Rewind the torid so it has 120 turns with the tap still at 20 turns from one end. Now you will have 5x the feedback voltage and can use a 5K resistor instead of a 1K. That will cut your off state current losses to 1/5 what they were before.
@PhattyMo Words to live by: "Green energy is for fools, Jeri!" Seriously, if solar power was the answer, they would be covering our roofs with free panels just so they could us sell the power. So far the money cost kills it; solar earns less than other investments. And who's going to cough up the big bux to fix a break in the wiring on the roof or an inverter that craps out? Jeri needs to invent an AC-output solar cell that plugs in an outlet that they can sell at the 99-cent store.
@CampKohler solar power could easily solve all the worlds power needs, that is if the "invention secrecy act of 1951" was abolished. At the time anything over 20% efficiency was subject to restriction (based on being - detrimental to the national security), and the patents taken from the inventor.. Best solar panel you can buy now is 23%..You better believe there is much better, whether you can buy it, is another story. Nanosolar? Know any place to get them?.. Neither do I..
@Wavefront101 I must admit I hadn't heard of the act, but I am aware of it now. I guess the only way out if one thinks it might be invoked is to forego patenting and try to be the "firstest with the mostest" to market. I wonder if there is any prohibition on getting a foreign patent first, and, once the idea was so exposed, there would be little excuse to invoke the act?
@CampKohler problem is, most if not all corporations won't do it without a patent, otherwise they have no protection on their product, and therefor would not make any money, which is the driving force of corporations. You could get foreign patents, but that will only draw attention to yourself. Most know what happened to Stan Meyer. The one and only loophole has always been open-source and giving the schematic away for free. Surprisingly no one thus far has been successful in doing that.
@CampKohler Its just the cost of good cells. Every technology starts off expensive and in below par designs. In time it will be cheaper to make efficient panels. The question is will something else over take it in that time...
@MrPieChee There's more to it than just the cells; there is the cost of the structure to support and protect them, the weatherproof wiring to connect them and the inverter that converts DC to AC and synchronizes the latter to allow connection to the grid. Also ther is the cost of installation, and, as I mentioned, ongoing maintenance. It all adds up, and, if the investment yields less than others, solar would have to be done on principle (green is good) rather than for monitary reasons.
@Insignificantful It turns off the switcher by grounding the base of the powersupply transistor when there is enough light. You're thinking about it backwards.
Hi yes that would work, I think this is the solution PNP as a switch to the center tap, with the base of the transistor on the anode of the schottky, through a resistor of course.. Try it out Jeri :) And Thanks for all the cool videos..
@Claude77 That was my thought too, then I thought of writing it, and then I said to myself that someone whould already thought and write it, so I just have to find the comment and vote it up...! Internet is weird...
Cool video. Very instructive. Would using a capacitor across the output of the flyback section give you a more constant output. I understand that the flyback section gives you pulsing DC on the order of at least some khz.
@jeriellsworth Thanks for the feedback. I have a few of these "thieves", I'll try doing that with one of mine.I didn't have the sense to think that by adding the capacitor it's now a tank circuit. Thanks for the reply, Great work keep it up.
Don't last very long meaning.... the battery is overcharged and damaged? or the battery never gets a full charge because the circuit is wasteful and the the LED just doesn't light for long? sry, I has an electronics noob.
Yes, shunting the base bias to negative is not the most efficient way to turn off the JT. I do it several ways in my recent blog. You may have never thought that a red LED could turn off the JT. It's at rustybolt dot info and click on the link to my blog.
acmefixer1 1 month ago
what size is the wire on that joule?
jayson01982 7 months ago
This has been flagged as spam show
hi i need your help as i can see your very good with solar power lights things i need help i have a project that im doing for my back yard i need to make a solar power light with a lot more led kinda like a led Christmas light i need it to turn on at night and have a power switch also . how would i do this please help me if u dont fully under stand my question pm me and ill be more descriptive . thank you very much
Matta212121 7 months ago
Can't you add one of those coil things to the solar cell to boost the charging power to?
AKSoapy29 7 months ago
Very well detailed, Intelligent, Well explained. I like it.
PoirierMike 8 months ago
can you charge your phone with this?
ltkipras 8 months ago
Use a darington transistor to reduce the waste of the current.
MrThomassss 9 months ago
Thanks for the explanation. I nearly understand how it works now. :)
shades2 9 months ago
You need to build a panel that will work at night.
12b4u2 10 months ago
Hi Jeri, interesting problem to solve. I had a go at redesigning it last night in LTSpiceIV and came up with the following circuit: screencast com/t/fuiEmTVyMg It is essentially a PNP BJT that switches off the center tap of the transformer when the solar cell is within battery - .65V - IR of the base resistor. Shottky from cell to batt. Adj. base resistor for better efficiency, I was really boosting the output. High gain BJT's or possibly FETs would be better. -Brett (FightCube)
MonkeyFCoconut 10 months ago
I wonder if a Panasonic 1381/13811/13812 voltage detector would work in this circuit to replace the NPN on the left? They're used in a number of BEAM solar related robotic circuits. The 13812 has a CMOS inverted output in a TO-92 package.
roboanalogtom 10 months ago
You should try to use some supercaps, they're perfect for nightlights.
yellowmetalcyborg 10 months ago
@yellowmetalcyborg The trouble with outdoor solar night lights is that, unless they are in a secure area, unlike the wired variety, people can just walk up and pluck them from the ground like flowers. Perhaps someone will come up with a mounting stalk that can be driven into the ground and the, with a twist or some other mechanical trigger, cause spikes to radiate outward, anchoring them more permanently. (Oooh! I just thought of a way, but-I-won't-tell-you).
CampKohler 10 months ago
@CampKohler I was talking about supercapacitors for storing electrical charge, I have do idea how that relates to spikes and anchors.
Anyhow, here in the United States I'm allowed to shoot trespassers in my backyard. I just so happen to have a dart gun, so no need to make an anchor for nightlights. What a lovely country! (not).
yellowmetalcyborg 10 months ago
@CampKohler I was commenting on the use of outdoor lights, per se, not the caps. But I don't think you could get much time from even whole-farad caps.
With respect to backyards, here in CA, it would get you prison time for sure unless you could show a threat to your safety; merely saving property would not justify shooting someone. (Put a knife in his hand!)
CampKohler 10 months ago
It's similar to what I'm working on for a school project. I used a bigger 6V solar panel, a battery from an old camera that is charged through a diode. The only problem is that I have to use a switch to turn the light on and off but looks like the simple methods are the best.
stelian95 10 months ago
@stelian95 If you want to blow the everyone's minds, put a small window shade roller next to the solar cell and arrange a rope with a knob on it passing through a guide loop, so they only have to pull the knob to cover the cell, else it retracts. Now here's the good part: build a hidden switch (reed & magnet?) into the shade that operates the light. When they cover the cell, the light comes on! Only the truly geek will think that this is absolutly whack and start looking to find out why.
CampKohler 10 months ago
@CampKohler That's brilliant ! I could try this.
stelian95 10 months ago
@stelian95 It's brilliant only if you pull the shade, otherwise it's dim. :-) Seriously, you could explain the shade by saying that it is there to simulate nighttime and that the light is not intended to come on during the day, which would waste the charge stored. What they will not be able to figure out from your schematic (if you leave the switch part out) is how stopping the charge causes the light to come on. The secret goal is to see if they are smart enough to ask that question.
CampKohler 10 months ago
@CampKohler And now for the real fun. Assuming you can stand nearby the project, you keep notes on what visitors do with the shade, how they react and if they ask "the question." The project is not the dumb light, but the results of the statistics about the visitors' behavior. In other words, it's a psychology project, not a science project! So, even if the light is a total bust, you tell the teacher(s) what the REAL project is and they will think YOU are brilliant. Sneaky is as sneaky does!
CampKohler 10 months ago
Yay, another Jeri-video! Moar!
juopotski 10 months ago
You can burn a bit less current keeping the switcher off by reducing it's base bias needs, it need only be biased enough to start up if the oscillator bootstraps itself. Maybe bypass part of the bias resistor with capacitance?
vk2zay 10 months ago
@vk2zay Interesting idea. Wonder if it would start.
jeriellsworth 10 months ago
@jeriellsworth another idea would be by cutting the supply of the inductor(the center tap) with pnp and voltage divider on the base so that it would start only below threshold voltage. hope youget the idea. :)
culerz 10 months ago
@vk2zay Thanks for the tips. I have a similiar circuit in one of my projects and what I did to kick start the oscillator is to use an CR integrator to trigger a pulse on the gate of an SCR. The SCR keeps the load to turn on as long as the PV cell is blocked. Once the load current drops below the holding current of the SCR (light is on) the entire circuit shuts off and no leakage is measured. Im not sure will it help, Jeriellsowrth. Anyway, I will also try to play with vk2zay's idea. Thanks!
SniperPIKACHU 9 months ago
I just wish nano solar would hurry up and release their thin film nano ink panels. Then we could all enjoy cheap abundant energy and stop polluting the enviroment, maybe even shed foriegn oil dependance
prankmypants 10 months ago
aren't joule thief's incredible inefficient?
MrPieChee 10 months ago
@MrPieChee Yes. That is what I'm saying in the video.
jeriellsworth 10 months ago
@jeriellsworth
Nice project. Actually a correctly built JT circuit is very efficient. Some of mine run for months 24/7 on an AA. Dr. Jones has measured possibly COP 8 with his circuit on oudotcom. Others in our group have circuits that will run for a year. Very efficient and flexible little circuit. You can really get the amp draw very low with a little effort. Very nice video.
Bill
Pirate88179 8 months ago
@MrPieChee
About 40 to 70 percent, from what I've measured; typically 50 percent.
acmefixer1 1 month ago
Jeri got me lookin'. If you Google "joule thief engine", you will find a puzzling (to me) circuit at beam-wiki that is supposed to cause an LED to blink, but I'll be ding-donged if I can see the power source. Anybody?
CampKohler 10 months ago
use a pnp for the trigerring with a voltage divider.
culerz 10 months ago
It's gona be a update for this solar garden light?
I wana built without joule thief,it s possible?
Regards jerrie
macobt 10 months ago
@macobt use a 3v battery
culerz 10 months ago
Here's an idea. Feel free to poke holes in it. Rewind the torid so it has 120 turns with the tap still at 20 turns from one end. Now you will have 5x the feedback voltage and can use a 5K resistor instead of a 1K. That will cut your off state current losses to 1/5 what they were before.
omegageek64 10 months ago
I'm impressed by your skills.
Have a nice day =)
IceFritzLanger 10 months ago
Lets get Jeri's evil genius working on some green energy!
The whole planet will be solar powered in a month!
PhattyMo 10 months ago
@PhattyMo Words to live by: "Green energy is for fools, Jeri!" Seriously, if solar power was the answer, they would be covering our roofs with free panels just so they could us sell the power. So far the money cost kills it; solar earns less than other investments. And who's going to cough up the big bux to fix a break in the wiring on the roof or an inverter that craps out? Jeri needs to invent an AC-output solar cell that plugs in an outlet that they can sell at the 99-cent store.
CampKohler 10 months ago
@CampKohler solar power could easily solve all the worlds power needs, that is if the "invention secrecy act of 1951" was abolished. At the time anything over 20% efficiency was subject to restriction (based on being - detrimental to the national security), and the patents taken from the inventor.. Best solar panel you can buy now is 23%..You better believe there is much better, whether you can buy it, is another story. Nanosolar? Know any place to get them?.. Neither do I..
Wavefront101 10 months ago
@Wavefront101 I must admit I hadn't heard of the act, but I am aware of it now. I guess the only way out if one thinks it might be invoked is to forego patenting and try to be the "firstest with the mostest" to market. I wonder if there is any prohibition on getting a foreign patent first, and, once the idea was so exposed, there would be little excuse to invoke the act?
CampKohler 10 months ago
@CampKohler problem is, most if not all corporations won't do it without a patent, otherwise they have no protection on their product, and therefor would not make any money, which is the driving force of corporations. You could get foreign patents, but that will only draw attention to yourself. Most know what happened to Stan Meyer. The one and only loophole has always been open-source and giving the schematic away for free. Surprisingly no one thus far has been successful in doing that.
Wavefront101 10 months ago
@CampKohler Its just the cost of good cells. Every technology starts off expensive and in below par designs. In time it will be cheaper to make efficient panels. The question is will something else over take it in that time...
MrPieChee 10 months ago
@MrPieChee There's more to it than just the cells; there is the cost of the structure to support and protect them, the weatherproof wiring to connect them and the inverter that converts DC to AC and synchronizes the latter to allow connection to the grid. Also ther is the cost of installation, and, as I mentioned, ongoing maintenance. It all adds up, and, if the investment yields less than others, solar would have to be done on principle (green is good) rather than for monitary reasons.
CampKohler 10 months ago
Jeri - what's with the 21K ? Let me guess - you have an E96 resistor kit & already used up all the normal values...?
I think Linear tech have a similar problem - all their appnotes use wacky resistor values.
mikeselectricstuff 10 months ago
@mikeselectricstuff Hehe. I do have big kits of 1% resistor I feed from.
jeriellsworth 10 months ago
How does the transistor with the 21k resistor turn on if the photo-voltaic cell does not supply enough voltage when in a dark room.
Insignificantful 10 months ago
@Insignificantful It turns off the switcher by grounding the base of the powersupply transistor when there is enough light. You're thinking about it backwards.
jeriellsworth 10 months ago
Be careful, you are going to be sucked into the overunity movement..
Wavefront101 10 months ago
I always hit "Like" before the video even starts, and you haven't made a liar out of me yet. Another gem.
frac 10 months ago 8
At first glance I think a PNP working as a switch above the center tap might work. With the base somehow controlled by the panel voltage...
Orthosonic 10 months ago
@Orthosonic
Hi yes that would work, I think this is the solution PNP as a switch to the center tap, with the base of the transistor on the anode of the schottky, through a resistor of course.. Try it out Jeri :) And Thanks for all the cool videos..
Claude77 10 months ago 12
@Claude77 That was my thought too, then I thought of writing it, and then I said to myself that someone whould already thought and write it, so I just have to find the comment and vote it up...! Internet is weird...
ntomata0002 10 months ago
Cool video. Very instructive. Would using a capacitor across the output of the flyback section give you a more constant output. I understand that the flyback section gives you pulsing DC on the order of at least some khz.
0Fluxor0 10 months ago
@0Fluxor0 The capacitor will effect the frequency of the oscillator. Maybe good or bad. Worth a try.
jeriellsworth 10 months ago
@jeriellsworth Thanks for the feedback. I have a few of these "thieves", I'll try doing that with one of mine.I didn't have the sense to think that by adding the capacitor it's now a tank circuit. Thanks for the reply, Great work keep it up.
0Fluxor0 10 months ago
@0Fluxor0 put a diode and a cap on that output. but be careful using it on your devices and components cause it gets more than 100v off load..
culerz 10 months ago
Sweettttt Another great video, might have to try that project myself :)
XIXMayheMXIX 10 months ago
Don't last very long meaning.... the battery is overcharged and damaged? or the battery never gets a full charge because the circuit is wasteful and the the LED just doesn't light for long? sry, I has an electronics noob.
10mintwo 10 months ago
@10mintwo The charge doesn't last for long at night.
jeriellsworth 10 months ago