I've been running my system for a few weeks now, and its taking an amazingly long time to charge these batteries.

I'm using two 8AH batteries, and its taking it like 3 days to get them up .6 v.. Is there anything I'm doing wrong?

Ewhaz, with my 2000 turn trifilar on the bicycle wheel, charging caps with mechanical switch discharge, a few hours they would have a really good strong charge.

What are the coil and components you're using? Magnets, rotor size, caps you're charging. Your doing discharge with a 555 circuit? What voltage are caps before you discharge?

If the battery is not damaged and is really in good shape, I would take it off the SG, charge it to 14+ with a normal hot charger until good and full. Then drain it with a resistive load down to 12, do that a couple times. Then as soon as it is drained, then hook it to the sg charger.

The battery just may need to be reformed.

I'm using the 555/scr set up to discharge my caps into my batteries. The hard part about this is I'm still waiting for the proper caps to use for this circuit, as it's designed, but they are still on their way in from an oversea EBAY purchase. I went and cannibalized some disposable camera's to get the caps from them, and since they are so large they take longer to charge to any decent voltage, I'm only reading about 22 to 23 volts across them, where ass the other caps where getting into the 150v range. I've already hardwired my 555 circuit so I can't easily change the timing to let the caps charge higher.

I was also testing to see if dumping the coil directly into the battery (after the rectifier of course) would change the charging. It hasn't seemed to make any difference. I've got two more batteries on the way, I'll charge them up and discharge them like you suggested to start fresh when I finish getting all my componants.

I did want to ask about discharging the batteries. I have been using small 14v lamps wired in series to discharge the batteries, they were rated at .2 amps, which I figured would give me .8 ams overall for a C20 rating (with the two 8ah batteries in parallel) But I didn't get any decent discharge until I put two of the lamps in parallel. I can't seem to measure amps on my meter, it just cuts off anything behind it so I don't get any idea of the amp usage, only the amps available. How can I calculate the exact load for discharging these batteries at a C20 rating.

The best results I've had is with LARGE capacitance banks and only charging 2-3 volts above the battery voltage....around 14-16 or so. What is the largest capacitance you can come up with with the caps you have? You can probably find some 33,000 uf 60v's on ebay pretty cheap. I found some years ago on ebay for about $5 each.

With too high of voltage pulsing too quick, there isn't enough current to get the batts in real charging mode.

Right now, the caps I have on hand are some 450v 3.3 uf caps and two 330v 80uf caps.

So your saying that as long as I'm getting about 2 to 4 volts over my topmost battery charge then I'm Ok?

With high capacitance impulses that is. The capacitance of the caps you're using seem too small. A couple hundred uf's at minimum is really good for the capacitive discharges at a couple volts above the batteries. Even 15-30k uf would help improve your results.

Hi Aaron. Howdy' guys and girls.
I have a 132.000 uF/100V capacitor bank (6x22.000uF). Really BIG cans. 3$ a piece from a bargain sale.
It takes about 5 seconds to charge them to 50V in parallel configuration. The thing is that I cannot dump them onto the battery when they are over 43V using the 555 cap pulser. I blowed my 2N3055 Darlington in a split second trying. Three times in a row. Stupid me!
It looks that really big amperage develops in the capacitor bank after the 43V area and I think I'm gonna try a 2SC5244 (30A/1600V) in order to go safe beyond that range. I don't mind the 5-7 second time gap. I just want to experiment the way it charges the battery at higher voltage & amps.

Now the question is: is there any way to protect those expensive trannies from overcurrent? Thermistors? Diodes? And how do I prevent SCR's self-latching? I tried one in the circuit but it stays on forever.

My 555 cap pulser is fully adjustable in terms of frequency and duty cycle and I have found that going with a smaller duty cycle (10-90% or 20-80%) nothing happens. The battery won't charge at all. Even by using higher voltage.
I got best results by now at 800ms ON / 1s OFF using all 6x22.000uF in series. The battery (12V/12A) gains a tenth of a Volt every 5 minutes or so.

Mr. Bedini was right again (look at the patent). Don't bother to build your cap pulser having adjustable duty cycle unless you want to use the same circuit for the solid state version. It's a waste of time. 50-50% or 40-60% duty cycle works the best for now.

Cheers.
Valentin

Hi Valentin,

I would dump the caps when they're only a few volts over the battery so probably 15-16v is all you need then you'll be able to charge them faster to that and dump once a sec or once every other sec. You'll get better results anyway.

You can use a transistor instead of a scr if you can't get the scr to turn off.
2SC3281 worked great for me.

Aarons right, and besides, 130k+ uF dumping @ nearly 4 times the chargings capacity could be a little dangerous, especially if your battery isnt up to scratch.

Like Aaron said, try a good transistor in there, Ive used a few different ones including the MLJ21194.

Ive been playing around with my trifilar charging 18000uF cap. The 555 is set to dump every 1-1/1/2 seconds which builds the voltage up in the caps to around 20v. Its too much still I think, I need some bigger caps. Thats a bargain. $3.00 a cap. GOLD.

Thank you Aaron.
Already followed your previous post suggestion and my battery climbed from 12,37V to 12,58 within an hour. Steady, not fluffy charge. It works great this way. I dump it with 20V every second or so. To make a difference, for the next charge I will use 18V.

As for the transistors, I'm already tired of changing them so I stopped experimentig with higher voltage& current.

I got rid of the SCR the very first minute of running and made a quad Darlington from the 6N138 opto-Darlington, BD243C and 2N3055 + 870ohm base to emitter resistors. That's what I had around and it works pretty well. With only one condition. Not to go over 42-43V or the Darlington goes puffff (the 2N3055) because the current from the capacitor bank is way higher than this transistor can handle.

Right now, I'm working on a new coil(s) for the solid-state version because my actual SSG behaves erratically. I managed to make it self-oscillate but as soon as I connect the capacitor (any) it stops from oscillating for good.
And it won't start with the capacitor or battery connected. Any ideas why?
Probably because the core is made from a soft iron bar and not welding rods? Dunno'.

Thank you again!
Valentin

Thank you Ren. I've already learned my lesson with the cost of three 2N's.
As for the caps, they have a chinese mark engraved in gold on top of them. But they really kick bottoms! And I don't care that they're chinese made. They're up to the job.

Valentin

If you charge a battery with this method for an hour, you may be able to turn it off, watch the voltage in the charging batt and see it climb by itself for a while. It 'may' happen with this type of discharge but a mechanical switch might be necessary to see that effect.

Because of the spark. That's where the radiant efect appears. Right?

There may be something to the spark but a mechanical switch has the lowest loss of any switch to my understanding. I have always had the best results with mechanical switches.

Then I will try an automotive lights relay. With a little adaptation to the cap pulser to drive it. I'll see what difference comes from this. Thanks for the ideea. Can't do it directly with the SG because the wheel is too little and has no use for the timing.

As for the solid state, do you have any ideea why it stops oscillating every time I connect a cap or battery? Can't start the self-oscillation otherwise. As soon as I have a load, nothing happens. I tried a range from 1-100K for the base resistor. Nothing!
I have a 50K pot in series with a 1K pot and a 10 ohm pot. For fine tuning. Without a load, oscillation starts at 3 Kohm.

Hi guys.
Yesterday I tried the following:


A a simplified version of the Bedini scalar battery charger (first was a self latching configuration of the relay). Too lazy to hook it to the cap pulser at that time.

It does not work according to my schematic. The measured voltage on the load (small motor) terminals is around 6V, not to mention the battery running down...

Also driving the relay with the cap pulser made no difference except for changing the frequency. The caps (68uF/400V) are charging to 12V in parallel config, then drop to 6V in series. I don't understand why.
I saw this strange effect by slowing the cap pulser.

To test it, I tried the same thing by replacing the battery with an another 68uF/400v capacitor (charged at 12V), without a load. It runs down rapidly instead of charging up. I know I made THE mistake. Forgot about the diodes...

Here's an update to what I have in mind to test this week-end if I manage to find a 4PDT relay.


If everything goes right, the next step will be replacing the battery and capacitors with 22000uf/100V capacitors and bigger voltage load + zenner diodes to keep everything at a certain level.
Then,
And then: