current plus radiant charger apears to be OU

[Bradley Malone]

what I have running right now came from a desire to build something that uses normal current charging along with radiant charging. I wanted a system that would pulse normal current then radiant in alternating pulses. I have used two AA batteries to charge 4 AA batteries a few times now so without hard data (working on an arduino batt logger for hard data) it apears to be a %200 system witch is also backed up by simulations.

In the simulation i used a self triggered relay as the switch and the inductor but in reality I am using a reed switch and a coil running a low friction flywheel with magnets. the way I am using it is this...two run batteries are in series pushing current through an inductor and then through one inverse charge battery (like a tesla switch). because I am using two to one for instance (24v run and 12 volt charge) it leaves one battery of potential accross the coil. the current simultaneously charges the coils inductance and the inverse battery with forward "normal" current. once the switch disconnects the diode allows the coil current to continue to discharge through the inverse battery "radiant" style. in the ciruit simulation you can see that the source ramps upward to the peak then shuts off while the current continues on and ramps down due to the coil inductance...giving the seeming 2 to 1. I have not tried a transistor version or anything other than the reed switch setup I have now but as a stated I have ran it a few times (3-4) and it appears to do just as the simulations show...I charge about twice as many batteries as I am using to run it...sometimes only 3 because of the losses in the motor section. I do know that a voltage on a battery does not prove anything so as I stated I am currently working on an arduino data logger for the batteries so I can track the charge/discharge and also be able to drain the batteries through a known resistance and log that to figure the capacity and get all the hard data...but at the moment I am just able to run it and watch batteries charge...but so far it looks promising and I have never seen a system that used both normal current and the inductive kickback from a coil in one system so i decided to share so all you smart people can make it better or try it out.

[Bradley Malone]

it would also be really simple to change an sg into this type of configuration to see what differences it has.

[Aaron Murakami]

Here is a scope shot for quite a few weeks ago of an almost identical schematic I've been testing at Graham Gunderson's lab.



The bottom green you can see the charging battery receiving the current pulse and then the spike.

It makes one hell of a charger, that's for sure.

I don't know if the circuit I tested is better than yours but it is the same idea with a 3 battery setup but slightly different arrangement.

[Bradley Malone]

i just changed a forced oscillation bedini sg (oscillator) over to this and damn! this thing charges like a banchy. as I stated before i have not got hard figures but just from observations it charges way!! faster than anything i have built so far. I am jumping around my house ....my current meter shows miliamp draw and the voltage is climbing....quickly. i will do my best to try and get some hard data soon but this looks promising!

thanks for the input aaron....I have read alot of your stuff and appreciate all the work you have done.


I am trying my hardest to not get to excited...but i have built sg's and all kinds of bedini stuff (to specs) and seen the benefits and I truly appreciate all the information giving by all involved. but I know they left stuff out...they even call it a learning tool. but even my best tuned machines did not charge like this. given that i did not have any large machines the fact that the little coil i have running this thing is charging the way it is is making me quite excited.

[Bradley Malone]

it cracks me up sometimes that the people working on this stuff end up coming up with the same stuff...can't count the times i had an idea about a circuit and come to find someone else on this forum or something like it "just came up with it" also. if anyone with better equipment tries this please upload any data you have so we can compare and see if the benefits are real...again aaron thank you I really appreciate it.

[Retep]

Hey Bradley,

The circuit Aaron is talking about is one I developed last December for an advanced SG project. Here's a drawing I made:



It's remarkably close to your drawing. This circuit is designed to run with an automatic 3 battery swapper, to keep an SG wheel spinning for a long time. Also, the extra diode on the positive line is necessary to make sure the Radiant goes 100% to battery #3.

Enjoy,
Peter

[Aaron Murakami]

Bradley,

For the data, that will come later at least for my experiment - been working on something related to this.

Peter - thanks for posting that. Super simple variation that does charge the battery like mad.

Bradley - I'd recommend looking at the differences between the 1N4007 diode, which is commonly used with the SG and the UF4007, which is ultra fast.

With the 1N4007 with the typical SSG, there is kickback to the front because that diode is really slow so you get some extended running time.
With the UF4007, its so fast that virtually everything goes to the back battery.

I guess it depends on if you want extended running time or you want the output battery to charge quicker.

The difference is plain as day when you try both and you scope the diode or even just the batteries.

Diodes make a huge difference depending on your goals.

With the old Ainslie mosfet circuits pulsing a battery into an inductive resistor, I bypassed the junk intrinsic diode with a high quality fast one and got a 100% increase in the kickback going to the front battery. That just means the intrinsic diode is slower than molasses.

I wouldn't get too caught up in the diode deal but if you want to throw everything at the back battery or in this case, the single opposed battery, then you might want to consider using UF4007 - they're cheap too.

http://www.amazon.com/gp/product/B00...SIN=B00NWDT61G

I got 400 delivered for under $8 on Amazon prime.

[Aaron Murakami]

I am trying my hardest to not get to excited...but i have built sg's and all kinds of bedini stuff (to specs) and seen the benefits and I truly appreciate all the information giving by all involved. but I know they left stuff out...they even call it a learning tool. but even my best tuned machines did not charge like this. given that i did not have any large machines the fact that the little coil i have running this thing is charging the way it is is making me quite excited.

Actually, there is way more in the books than people think. That is definitely the case in the Advanced SG book. Peter put some of the most important stuff in the back of that book, but doesn't seem that many people noticed

[Bradley Malone]

I know what your talking about I have read all the books and looked pretty closely for details...but theres always something ya miss . as for the diodes I am using sr 360 shottky diodes at the moment witch seem to do pretty good but i do not know how fast they are...data sheets don't say and i dont have my scope at the moment. right now I have the forced oscillation circuit running at about 200ma and its charging a 100 amp hour car battery that went bad. after only about 20 to 30 minutes it brought it up to 12.77 so things are looking good. will be doing a few "real" test sometime very soon to test between this setup and the original "radiant only" setup to lock in what differences happen between the two circuits.

thanks Peter and Aaron

[Dave Wing]

Hi All,

This thread reminds me of some of the circuits I was also working with back in 2014,which again are very similar to what you have all been working with, found here...

http://www.energyscienceforum.com/showthread.php?t=2054

I asked people to try it back then but no one posted their results... Here is my own quote from that thread... "I have a favour to ask... can someone try this circuit and tell me what they find out."

Dave Wing

[Aaron Murakami]

Thanks Patrick,

I have been drawing the circuit wrong for a while... Memory is going down hill I guess. Here is the correct drawing.

-Dave Wing

Thanks Dave,

I have tried that before but it separates the current punch on the inverted battery from the spike going to a separate battery.

From my brief testing, Peter's variation with the alternating current/spike/current/spike to the same battery seems to work best.

Bradley's probably should do the same for the most part, but just requires testing to have a real comparison.

Just starting off with a full charged battery in all 3 positions, the battery receiving the charge was pushed to 16.9 volts before I realized it was that high.
With any other variation of normal SG, common ground mode, etc..., none of those will push the battery to that voltage in a short period of time like that.
My SG was a 7 power winding 1 trigger SG - the old plastic bicycle wheel kits that were available for a short time.

Will do more thorough tests with a battery that is drained down - someone is working on a data logger and auto battery rotator.

[Bradley Malone]

I have a logger that works ok with AA batteries. it is only off by 0.01 volts so its good enough to make a graph. I am using a voltage divider made from two 100k resistors to the A1 pin on the arduino. I am using a ds3231 Real Time Clock and an sd shield to log everything. here is the code if anyone has this stuff laying around.

#include <SD.h>
#include <Wire.h>
#include <RTClib.h>

#define LOG_INTERVAL 2000 // time between readings
#define SYNC_INTERVAL 10000 // time between sd write
uint32_t syncTime = 0;

#define NUM_SAMPLES 5 // change to times analog is read(averaged)

#define redLEDpin 10 // write to sd status pin
#define greenLEDpin 11 // analog read status pin

#define battPin A1 // analog read pin

RTC_DS1307 RTC; // define clock object
const int chipSelect = 4; //set to chip select pin on shield

unsigned char sample_count = 0; // current sample number
float sum = 0; // sum of samples taken
float battVoltage = 0.0; // variable for analog read data
File logfile; // file to write to

void error(char *str) // error string
{
Serial.print("error: ");
Serial.println(str);
digitalWrite(redLEDpin, HIGH);
while(1);
}


void setup() // setup
{
Serial.begin(9600);
Serial.println();

pinMode(redLEDpin, OUTPUT);
pinMode(greenLEDpin, OUTPUT);
pinMode(4, OUTPUT);

Serial.print("Initializing SD card...");

if (!SD.begin(chipSelect)) {
error("Card failed, or not present");
}

Serial.println("card initialized.");

char filename[] = "LOGGER00.CSV";
for (uint8_t i = 0; i < 100; i++) {
filename[6] = i/10 + '0';
filename[7] = i%10 + '0';
if (! SD.exists(filename)) {
// only open a new file if it doesn't exist
logfile = SD.open(filename, FILE_WRITE);
break; // leave the loop!
}
}

if (! logfile) {
error("couldnt create file");
}

Serial.print("Logging to: ");
Serial.println(filename);

Wire.begin();
if (!RTC.begin()) {
logfile.println("RTC failed");
Serial.println("RTC failed");
}

logfile.println("millis,stamp,datetime,battVoltage ");
Serial.println("millis,stamp,datetime,battVoltage" );
}

void loop() // loop start
{
while (sample_count < NUM_SAMPLES) {
sum += analogRead(A2);
sample_count++;
delay(10);
}
float battVoltage = ((float)sum / (float)NUM_SAMPLES * 0.004912) ;
// change 5.015 to measured actual refference voltage
delay(10);
DateTime now;
delay((LOG_INTERVAL -1) - (millis() % LOG_INTERVAL));
digitalWrite(greenLEDpin, HIGH);

uint32_t m = millis();
logfile.print(m); // milliseconds since start
logfile.print(", ");
Serial.print(m); // milliseconds since start
Serial.print(", ");

now = RTC.now(); // log time now
logfile.print(now.year(), DEC);
logfile.print("/");
logfile.print(now.month(), DEC);
logfile.print("/");
logfile.print(now.day(), DEC);
logfile.print(" ");
logfile.print(now.hour(), DEC);
logfile.print(":");
logfile.print(now.minute(), DEC);
logfile.print(":");
logfile.print(now.second(), DEC);
logfile.print('"');
logfile.print(", ");
logfile.print(battVoltage * 2.0129); // change 11.132 to measured voltage divider (batt voltage divided by voltage out)
logfile.print (" V");

Serial.print(now.year(), DEC);
Serial.print("/");
Serial.print(now.month(), DEC);
Serial.print("/");
Serial.print(now.day(), DEC);
Serial.print(" ");
Serial.print(now.hour(), DEC);
Serial.print(":");
Serial.print(now.minute(), DEC);
Serial.print(":");
Serial.print(now.second(), DEC);
Serial.print('"');
Serial.print(", ");
Serial.print(battVoltage * 2.0129); // change 11.132 to measured voltage divider (batt voltage divided by voltage out
Serial.print (" V");

logfile.println();
Serial.println();
sample_count = 0;
sum = 0;
digitalWrite(greenLEDpin, LOW);
if ((millis() - syncTime) < SYNC_INTERVAL) return;
syncTime = millis();

// blink LED to show we are syncing data to the card & updating FAT!
digitalWrite(redLEDpin, HIGH);
logfile.flush();
digitalWrite(redLEDpin, LOW);
}



the multiple after reading the battery voltage is found by measuring the Aref and dividing it by 1024. then when it prints you multiply it by the divider ratio witch for me was 2.0129. anyone doing this will want to measure their reference voltage and divider and replace these values in the code. I used the voltage divider in the hope of being able to read 12 volt batteries but when i try it they read off by 2 volts witch is not good enough...but as i stated it reads only 0.01 off when reading AA batteries so for a small scale operation it will work.

[Bradley Malone]

also forgot to add that it is coded to take 5 samples and average them for the reading to get rid of any erroneous readings. you can change the sample count and also how often it samples by changing the sample number or "log interval" at the top of the code.

[wheatbelt.radio]

OU into batteries is great, especially with Large wet PB cells, nice sharp pulses and offtimes - gearing the pulse size too the batter size is the tricky part too ovoid loosing NRG, i found a simple bridge frmo the mains with a series capacitor on one rail of the bridge input, too be excellent, bridge outpuot straight over the batteries, and my country mains is 240 v, excellent desulphate and charge speed / efficency, a real contendor !! but currently working on rosemary ansile circuit - GOT COP10 reliable design and trying too market it, please , a little supoprt would be awesome.

https://igg.me/at/free-heat

[John_Koorn]

Thanks for posting your circuit Peter, I've been wanting to experiment with it for a while.

Here's one version I quickly put together. I already had the tape drive motor SG, so it only took a few minutes to get this running.
It's charging the heck out of B3!

I'm thinking about how I might be able use this in a four battery Tesla Switch

John K.