help with solar system

Thread Starter

AE-Bound

Joined Jul 22, 2008
2
hi all,

I am assembling the componets for a solar system and I saw a news article on a diy system a guy had.

He had a 24x12v solar system and he had a unique charging system I would like to try as well. He was using a PWM system that scanned all the bats and any battery below a certain V ( I think he said 14v) got a pulsed charge. The PWM just cycled through all the bats continuously at some unknown (to me) freq and pulsed as needed to keep the 24 bats at that pre-set V.

I'm not possitive but I think he used a 24v pusing system for some reason, so I guess he set the 12v into sets but the stats he gave were great and I would like to try and duplicate it.

So during the day, he had I think 4-6 bats that received all the solar charge and then he pulsed the others from them. He also said when he grid charged he would charge the same 4-6 bats only and his PWM would pulse charge the others. This evidently kept all the bats at an even charge and they lasted longer and the whole system integrity was much higher.

Anyone have an idea how to build a PWm that would also scan 24 bats and pulse as needed. I think I would have to be able to adjust the PWM duty and freq to be able to fine tune it and of course be able to fine tune the pre-set V level for best operation.

There are a lot of PWMs available out there but not any that I can find that also will scan an inventory of batteries to pulse as well.

Any ideas?

I'm willing to exchange services or something with anyone who can come up with the best way to do the above.

Thanks

Horace
 

scubasteve_911

Joined Dec 27, 2007
1,203
Horace,

This project is easily accomplished with pretty much any available microcontroller. The microcontroller would monitor via switching an analog mux to an ADC, which can be internal or external to the microcontroller. You need to divide down the voltage so that it is within ADC range, and probably provide an input clamp in case of over-voltage.

I am not sure what sort of current you're working with here, but you may be able to drive a fet directly from the microcontroller's IO. You should pick the lowest RDSon FET, or at least one that will not dissipate too much power when on. You might need to drive the FET via a gate driver circuit / IC if the gate charge is too high.

The programming part will let you decide how often to poll the battery voltages and whether or not to transfer some charge or not. There may be special cases where all batteries are above this voltage, and the microcontroller needs to make the decision to charge the lowest read voltage. Things like this make the microcontroller an attractive solution.

I find the PSoC microcontroller really simple to use, or some people like to use PICs.

goodluck,


Steve
 

Thread Starter

AE-Bound

Joined Jul 22, 2008
2
Thanks John and steve for the fast responses.

John, I don't know much about electronics. I thought the guy said PWM or I got the idea somehow that's what he was using.

I do computer work and write. I have little experience with electronics, so I am just trying to figure it all out now.

Steve, that idea of the micro controllers sounds promising. If the lanquage is not complex I could get that to work maybe. I will go investigate these things now and write here again soon.

I just don't know how much I need to learn about circuitry and electronics to get what I want done. Not sure if I should go for it myself with some help and guidence by guys like you two or seek pro-help.

I can see I have a lot to learn.

thanks again

Any other ideas - pointers- help?

Horace
 

SgtWookie

Joined Jul 17, 2007
22,230
Actually, I've been considering something similar for maintaining batteries on a fleet of vehicles. Not solar, but it's still applicable.

A uC would control switching amongst a bank of batteries, the batteries being in vehicles, of course. Basically, where the PWM comes in is that the batteries at a low charge level would receive a porportionately longer charge time than the other batteries. However, if the charge time becomes overly disporportionate, it indicates a problem with the battery. You can spend a heck of a lot of time and energy trying to charge a battery with a shorted cell, and it will be wasted. Identifying such defective batteries and eliminating them from the charge cycle is problematic.

Somewhere I've read that rapid cycling of charge current application can help to "kick" sulphation from the plates of the lead-acid cells. I don't know of the validity of these claims. Snake-oil salesmen abound nowadays. :rolleyes:
 

scubasteve_911

Joined Dec 27, 2007
1,203
No problem, let us know if you need help. You'll find on this forum, you can get answers to a really wide-range of problems, but nobody will do the work for you. You seem like a really motivated person that is willing to learn, the perfect type we like to see on this forum!

I am really busy working (electrical eng. consulting) and doing my own hobby work on the side, so I rarely get the chance to visit the forum. You can email me at my secondary address if you need help right away mcfadyen_s at hotmail dot com

Good luck!

Stephen
 
Top