Adaptive solar capacitive pulse charger

[jonnaprabhu]

Hi,

This is Prabhakar from Hyderabad, India. This is my first thread on this forum.

Wanted to get your thoughts on a design that I have been working to get the circuit in Bedini patent #6677730 modified to a Solar so that it gives a constant current pulse irrespective or solar irradiation.
I have been able to get a 10A pulse from a 2A source so far with my experiments.
Wanting this to make it adaptive so that we get a constant current pulse of say 5A in low light conditions also.

Thanks,
Prabhakar

[Tom C]

JB has said often, charge in series, discharge in parallel.... so in low current situations you want all the caps to charge to 2x battery voltage, then dump in parallel so the current is higher with the dump pulse. so if you had low esr caps say 4 at 25000 MF that is 100,000 uf of potential when dumped as a single capacitor. they will charge at whatever rate the panel output is, and then discharge in parallel with a larger pulse. hi potential output. your pulse current is dependant on cap size, battery size, and its internal resistance. pulse frequency is dependant on how fast the caps charge. you can vary the pulse current by changing how fast the circuit dumps to the battery, the higher the dump threshold voltage the higher the current pulse will be. it will be very hard to measure the actual current unless you have a current clamp for your oscilloscope.

Tom C

[jonnaprabhu]

Thanks Tom for the quick reply.
Giving some specifics, I had a bunch of 4700uf which I wired in parallel effecting to around 0.1F
As you said, the charging rate is dependent on the panel output which peaks during midday.
My lab tests have shown a charging time of under a second with a bench power supply at around 25V with the current limit set at about 2.4A
With this, I was able to achieve a pulse of 10A consistently.

The next thing is about the battery internal resistance. What I found is that if we go higher on the cap voltage, better is the charge transfer.
For eg., for a 12V battery, the pulse current was limited to about 1A when the cap voltage was 15V. The pulse current kept increasing with the increasing cap voltage.
I was able to get a pulse of 10A being transferred to the battery after the cap voltage crossed 23V. This testing was done on a 12V 45Ah SMF battery.

I will have to test with normal flooded batteries to confirm on the relation with the internal resistance.
Another point worth mentioning is the use of just two MOSFET's for charge and discharge. For a duration of 30min, they hardly got warm.

Prabhakar

[Tom C]

double the battery voltage is a good place to start for the dump voltage. 1 Farad!! that is a lot of soldering
Tom C

[jonnaprabhu]

Yeah, that ended up almost the size of a brick
But it was worth the effort. I picked those as they were available cheap and were of a brand that I knew (sanwa).
I recently got some supercaps from aliexpress. Those came in cheap and I think they were 10F/2.7V.
I am yet to use them for this purpose.

For now, I am busy with this microcontroller based design. Am trying to get it to a working prototype.