Solar Charging/Pulsing Circuit and Variants

[Farmhand]

Sucahyo, I didn't actually mean two caps in paralell would reach desired voltage quicker, rather that two in paralell charged to a certain voltage contain more energy than two in series and when the two in paralell are switched to series the voltage would be double the series charged one's.

I don't have enough time just now to study your drawing but I like the look of it. I'll watch the video now though. I can't say for sure What the benifits would be. But charging two in paralell should be quicker than charging two separately to the same voltage one after the other. Because of the lower resistance of the paralelled one's.

Limiting the current kind of defeats the purpose, I want to remove any current limiting effects. Maybe I miss understand, I think that relay array might be good I can't follow all the wires but I trust your judgement, i'll look at it better later. The current limiter definately works.

FRC, I'm not sure how I will do it yet, no idea, but it must be possible. I was just thinking on it and figured it was worth a try. Make no mistake if I could already do it I would post a drawing and a video.

I like to share even my idea's. Maybe someone can do more with my idea's than i can. I will need to make some drawings as I think. Or adapt a method for a similar thing from somewhere else.

Cheers

[sucahyo]

Sucahyo, I didn't actually mean two caps in paralell would reach desired voltage quicker, rather that two in paralell charged to a certain voltage contain more energy than two in series and when the two in paralell are switched to series the voltage would be double the series charged one's.

But charging two in paralell should be quicker than charging two separately to the same voltage one after the other. Because of the lower resistance of the paralelled one's.

Thanks, but I still confuse....

What happen if you put a volt meter only on one capacitor and put this capacitor along the rest for parallel and series charging test. I think the test should only be done with solar panel because using battery and power supply may change results. It is a test of what kind of load is better for the solar panel. Low impedance or high impedance? Or doesn't matter?

Because we don't know if dumping in series would be better than dumping in parallel or not.

Maybe discharging in series allow longer charging to the battery, but it will also lowering the capacitor voltage. So:

discharging in parallel: short discharging time but higher left over voltage, causing faster regaining dump voltage.

discharging in series: long discharging time with lower left over voltage, causing slower regaining dump voltage.


I think this must be tried for real since battery may have different result from being charged with high voltage differential (series) than low voltage differential (parallel).

Charging battery with cap in series may have worse result. So knowing what kind of impedance the solar panel like is important.

Limiting the current kind of defeats the purpose, I want to remove any current limiting effects.

Current limiting is used to drive the relay at required voltage. Not to limit current from the solar panel.

[FRC]

Thanks Farmhand and sucahyo

Here is an of the shelf timer circuit. Wish I had one myself.

Multi Mode Timer Circuit

George

[Farmhand]

Well I agree sucahyo I need to test it to be sure, I have a small panel 5
watts so I can test it with some big caps i'll do it tomorrow. Should be clear sky. I'll make a report on the tests when i'm done.

Current limiting is used to drive the relay at required voltage. Not to limit current from the solar panel.

Yes I see what your drawing means now. Looks good have you tried it ?

I agree series dumping may not be desireable anyway, now that I think about it the solar panel works at maximum power at 17.8 volts so just charging caps to 6 volts above the battery or a fixed 19 volts alternately might be best way after all. It would be a lot easier too. I can already do that with one cap bank but I need the oscillator to trigger the SCR to turn off. It turns on ok but not off thats an SCR thing if the power keeps flowing it stays on I think.

FRC I would't mind one of those myself. I have a PIC controller Dev board but I don't know how to use it yet.

Cheers

[sucahyo]

Here is an of the shelf timer circuit. Wish I had one myself.

I think we can not use timer because what is needed is something that detect the capacitor voltage

I'll make a report on the tests when i'm done.

Yes I see what your drawing means now. Looks good have you tried it ?

Thanks. I just draw it. I have the relay but I don't have the capacitor and panel.

I can already do that with one cap bank but I need the oscillator to trigger the SCR to turn off. It turns on ok but not off thats an SCR thing if the power keeps flowing it stays on I think.

You use SCR circuit for cap dumping method like mentioned on Inquorate thread?

[FRC]

Scr

You use SCR circuit for cap dumping method like mentioned on Inquorate thread?

Sucahyo, I am using the Bedini inverted SCR circuit for dumping from my
solid state SSG circuit to convert from negative to positive energy in the battery. I am interested in how this could be used with electrolytic caps of
smaller value, and how the values of the components would be changed.
The SCR is a 2N6504, the cap is NAM !0j250v, diode is IN5408. Also what
was the Inquorate thread called ?

FRC

[sucahyo]

Sucahyo, I am using the Bedini inverted SCR circuit for dumping from my solid state SSG circuit to convert from negative to positive energy in the battery. I am interested in how this could be used with electrolytic caps of smaller value, and how the values of the components would be changed.
The SCR is a 2N6504, the cap is NAM !0j250v, diode is IN5408. Also what
was the Inquorate thread called ?

FRC

Wrong name, should be nvisser:
http://www.energeticforum.com/renewa...battery-2.html

I don't realize that using SCR can not shut off problem still unsolved?

Bedini solution:

After reading your post 44n on capacitor discharge, I can see why it did not work, you can not switch that way, you must look at the potential across the whole battery system. Then invert the switch in the system and it will switch. Think about what I just said, a fet will work fine if you do that.......... Two black wires, one in and one out, look at the IRF260 fet easy to figure the difference in potential between the capacitor and the battery. And look at it for DC SOA.
John B


Brent,
I'm talking about making a solid state relay for the switching.
The Fet would get around the problems with the SCR.Mathew Jones shows the circuit a few pages back, but that solid state relay is costly so we could make that with the opto and the regulator for the gate voltage and the resistor to pull the gate off. The way around his circuit for timing is a SIDAC at 40 to 60 volts to fire the opto when the capacitor reaches the correct voltage. that is the switch I'm working on. NTE makes that part NTE6415. The higher the voltage the better, then the battery will charge fast that is what we want. Capacitor dump circuits must be double in voltage or very close to it.

John K,
I would use that coil for the side coils. and wind another by putting two of the bobbins together double the impedance in Ohms 14. And make sure you put the shield on the bottom then run the Bedini/Cole circuit. The capacitor should be about 18000Uf use a SCR and a 4.7volt Zener with two 2.8 volt super bright Leds in series with the diode circuit. Machine RPM should be not more then 25 Rpm's. just giving you a tip. Good Work

John B

[FRC]

Thanks sucahyo

Thanks, I still am somewhat confused. Looks like Farmhand was trying to solve the same problem before. Your circuit at the bottom does make some sense to me with the zenner diode in it. I am also intrigued by the mosfet method. I have lots of mosfets from old PC power supplies if I could utilize them for the cap dumping purpose it would be nice. But I have trouble understanding those circuits also.

FRC

[Farmhand]

OK I see, that method with the LED's i'll try tonight, It might work by burning off the power to the gate maybe worth a try. If I get time that is, i'm working on the bearings in my bike wheel fo my SSG, this quick release wheel doesn't use a ball race it use's a heap of single bearings. Grrrr.

I did some tests but I had to do them in low light "panel upside down", because I don't have enough capacitors to be able to tell in full sun they charge too quick.
I'll let the results speak for themselves. I used 15 000uf 35 volt Caps.

Two capacitors in paralell- 1 min = 11.1 volts. 2 Mins = 15 volts. Full 15.1 volts. In one minute they stored 1.848 Joules.

One Capacitor- 1 Min = 14.52 volts. Full 14.8 volts. In one minute it stored 1.581 Joules.

Two capacitors in Series 1 Min = 14.7 volts. Full 14.7 volts. In one minute they stored 810.338 Millijoules. Connecting in series halves the capacitance.

I got the "Full" measurement by just leaving it untill it didn't go any higher.
This was done with the panel upside down so the full 20 volts was not possible. I forgot to measure the open circuit voltage.

It's very difficult to get a proper measurement because as the panel heats up the ouput drops, the sunshine is very hot here when there are no clouds.

FRC, i'll get around to a trying a mosfet sooner or later. Probably be fairly simple. Like a logic level fet (they only need 5 volts to switch fully on ) with a small cap and resistor from the cap side + to the gate and another resitor to take down the gate charge to ground or something.

I want it to be as simple to build and understand as possible. I like simple thing the simpler the better for me.

Back to the SSG wheel bearings.

Cheers

[FRC]

Agree

I agree with you the simpler the better. So with the Mosfet could you have some kind of cap at the end that would reach 5 volts and then switch the caps for the dump. Or some differential when the cap bank reached full potential, the
5 volts would go to the Mosfet to switch it. Actually, if it would only trigger at the full potential. I guess there would have to be some intermediary circuit in between the caps and the mosfet. Does this make any sense ?

FRC

[FRC]

Computer

I know I could do timing and switching using a Computer. I have programmed my own timer programs before. Also a crude Tesla switch could be done this way. But the computer is going to draw power.

FRC

[FRC]

Battery charger

A battery charger that has lights that go from red to green when batteries are
fully charged could be used. Hook up to the caps instead, when the light turns
green it would trip a switch to dump the caps then repeat. Small 12v battery
booster packs also have this function. I have some old ones that just might work. They have lost their usefulness, but when hooked up to caps the light
function could be exploited for cap dumping. Will have to try it. What do you
think Farmhand ?

FRC

[Farmhand]

Drawing

Hi FRC, i'm not sure about the battery tester, I think they can work several different ways.

The mosfet switch would work very much the same as the SCR, When the switch is off the caps charge up past the battery voltage, when they reach say 6 volts above the battery the voltage difference is enough to break through the zener diode to the gate which switch's the mosfet "on" to dump the caps. The resister from mosfet gate to ground pulls the gate charge down to turn the mosfet "off" then it repeats. The rest is just details. but thats just one dumper which may be all you want. I'll need two of them to alternate, controlled by a relay or something like sucahyo's drawing, maybe a combination. But I do need to work out a viable single dumper first, so thats whats i'll try to get done. My SSG bearings are fixed. Tonight i'll try to whip up a test circuit with a mosfet and another with the SCR and LED's.

I'll see if I can do a quick sketch to help me explain what I mean, i'm not a very good explainer. I'll post it as soon as I can. Very busy now the rain is gone, I lost two months work time because of constant heavy rain.

I'm just having a break from work at the moment. More to do yet. I've got a goat with a abscess to deal with, it's not very yummy to look at, the poor thing.

Cheers

[Farmhand]

Here's a drawing (attached) I was in a hurry and made some mistake's, the zener and the 1n5408 should be swapped positions (the zener could be 4.7 v and a different diode could br substituted for the 1n5408). The 6A10 diode can also be substituted with a suitible one and it's 0.5 or so drop (measure it) needs to be added to the total for the drop difference.

So that would be 17.8 v. See below. If the panel put out 20 v max it would stop pulsing at 14.2 volts. Well thats the idea anyway.

Battery 12.0 v
zener 4.7 v
6A10 0.5 v
1n5408 0.6 v

It all depends on the output voltage of the panel and what voltage you want to charge your battery too.

Cheers

[kcarring]

Thats really cool Farmhand thanks. A mosfet controlled capacitor voltage multiplier essentially?

Have you ever thought about having a controller to detect an optimum time for the battery bank to receive a radiant "conditioning charge"? I often thought if a board was smart enough, you could designate a battery or two as a master, and then at night, if the voltages on the bank were low enough, such that they'd take a charge...and ... there was no "load" on the bank (cause it was the middle of the night...) then it could essentially pass ALL power needs to a single, known charged battery. This would handly your clocks and timers and stuff in the off grid home that you did not want to "completely power down" ... then that master battery would deplete itself "pulsing the bank" cleaning up the bank, and charging it at a time when it was sort of needed anyway (after a couple or few days of no sun).