Battery charger IC - Help me

D

Davide Fabri

Guest
I have to design a lead-acid battery charger for my company. The source
is power coming from a solar panel array (up to 45V). The output must be
able to drive at least 10A at 30V (overcharge mode for a 24V battery). I
think the UC3909 from Unitrode could be a good choice, but the overall
system has to be controlled by a microcontroller, that should provide a
presettable current limit in trickle and bulk charge mode. Unfortunately
such IC requires a resistor network that defines a fixed current value.
I need help to design an interface that allows the current limit to be
set by the PWM output of the microcontroller.
Thank you very much for help.


--
Posted via Mailgate.ORG Server - http://www.Mailgate.ORG
 
In article <faf4eee2a830fa832f63292e2e3751fa.68471@mygate.mailgate.org>,
dfabri73@yahoo.it says...
I have to design a lead-acid battery charger for my company. The source
is power coming from a solar panel array (up to 45V). The output must be
able to drive at least 10A at 30V (overcharge mode for a 24V battery). I
think the UC3909 from Unitrode could be a good choice, but the overall
system has to be controlled by a microcontroller, that should provide a
presettable current limit in trickle and bulk charge mode. Unfortunately
such IC requires a resistor network that defines a fixed current value.
I need help to design an interface that allows the current limit to be
set by the PWM output of the microcontroller.
Thank you very much for help.



If you are going to go through all of the trouble to use a micro and a
Click to expand...
PWM output, why not go all the way and use a DAC to measure current and
voltage and write the algorithm to do the bulk, overcharge, and trickle
yourself. On the other hand, I have used that exact IC in stand alone
mode and was very satisfied with its operation. The only gotcha is you
can't use the trickle charge resistor in your application. It will be a
constant drain on the battery when the solar array in not producing
power and there is no way to get around that. At least with a micro you
can sense the array and shut everything down and go to sleep until the
sun is back up again.

Jim
 
I second that -- and get a copy of "Rechargable Batteries, Applications
Handbook", by the technical marketing staff of Gates Energy Products, ISBN
0-7506-9228-6. It was published by Butterworth-Heinemann in 1992.

In my experience it's an excellent book. It seems to still be up-to-date
(they even tell how to use NiMH batteries, which was pretty cutting edge for
1992). They'll lead you by the nose through the lead-acid batterie's
charging needs -- including temperature compensation, which may be important
for a solar cell application that's presumably going to happen where there
isn't reliable heat.

"James Beck" <jim@reallykillersystems_INVALID_.com> wrote in message
news:MPG.1a8b082ffcfb6f2498993a@news.west.earthlink.net...
In article <faf4eee2a830fa832f63292e2e3751fa.68471@mygate.mailgate.org>,
dfabri73@yahoo.it says...
I have to design a lead-acid battery charger for my company. The source
is power coming from a solar panel array (up to 45V). The output must be
able to drive at least 10A at 30V (overcharge mode for a 24V battery). I
think the UC3909 from Unitrode could be a good choice, but the overall
system has to be controlled by a microcontroller, that should provide a
presettable current limit in trickle and bulk charge mode. Unfortunately
such IC requires a resistor network that defines a fixed current value.
I need help to design an interface that allows the current limit to be
set by the PWM output of the microcontroller.
Thank you very much for help.



If you are going to go through all of the trouble to use a micro and a
PWM output, why not go all the way and use a DAC to measure current and
voltage and write the algorithm to do the bulk, overcharge, and trickle
yourself. On the other hand, I have used that exact IC in stand alone
mode and was very satisfied with its operation. The only gotcha is you
can't use the trickle charge resistor in your application. It will be a
constant drain on the battery when the solar array in not producing
power and there is no way to get around that. At least with a micro you
can sense the array and shut everything down and go to sleep until the
sun is back up again.

Jim
Click to expand...
 
In sci.electronics.design James Beck <jim@reallykillersystems_invalid_.com> wrote:
In article <faf4eee2a830fa832f63292e2e3751fa.68471@mygate.mailgate.org>,
dfabri73@yahoo.it says...
I have to design a lead-acid battery charger for my company. The source
is power coming from a solar panel array (up to 45V). The output must be
able to drive at least 10A at 30V (overcharge mode for a 24V battery). I
think the UC3909 from Unitrode could be a good choice, but the overall
system has to be controlled by a microcontroller, that should provide a
snip
If you are going to go through all of the trouble to use a micro and a
PWM output, why not go all the way and use a DAC to measure current and
voltage and write the algorithm to do the bulk, overcharge, and trickle
yourself. On the other hand, I have used that exact IC in stand alone
Click to expand...
This would have the advantage that you could also do peak power tracking.
The peak-power voltage for the solar array depends quite strongly on
sunlight intensity (and less strongly on temperature).
Peak-power tracking can increase the amount of charge going into the
batteries.
 
Thanks everybody.
I will use a micro (an AVR) mainly to perform MPPT algorithm.
It will be no problem to implement the four charging modes
into software, and that was actually my first idea.
In particular I thought to use the micro to drive a current
mode synchronous buck controller without adding other ICs,
but I have not been able to find a suitable one. Maybe a
Linear LT1339 could do this job: it can easily handle voltages
up to 60V and directly drive the two mosfets. Nevertheless
it has some drawbacks: mainly it is not born to be
current-programmable, therefore I am not sure I will be able
to set the current limit from the micro; then ot needs 12V to
be supplied; and finally it seems to be quite old and very
expensive (about 7$ + vat !!).
Can anyone confirm this could be the right choice? ...or maybe
suggest me an effective alternative? Perhaps a lower voltage
IC connected to an high voltage mos-driver (like a IRF2110
from International Rectifier)?

I am going to post these latter questions as standalone
message. Thanks again for help.



If you are going to go through all of the trouble to use a micro and a
PWM output, why not go all the way and use a DAC to measure current and
voltage and write the algorithm to do the bulk, overcharge, and trickle
yourself. On the other hand, I have used that exact IC in stand alone
mode and was very satisfied with its operation. The only gotcha is you
can't use the trickle charge resistor in your application. It will be a
constant drain on the battery when the solar array in not producing
power and there is no way to get around that. At least with a micro you
can sense the array and shut everything down and go to sleep until the
sun is back up again.

Jim
Click to expand...



--
Posted via Mailgate.ORG Server - http://www.Mailgate.ORG
 
In article <862db8e6e7ab59b97378fab6bb87ccb1.68471@mygate.mailgate.org>,
dfabri73@yahoo.it says...
Thanks everybody.
I will use a micro (an AVR) mainly to perform MPPT algorithm.
It will be no problem to implement the four charging modes
into software, and that was actually my first idea.
In particular I thought to use the micro to drive a current
mode synchronous buck controller without adding other ICs,
but I have not been able to find a suitable one. Maybe a
Linear LT1339 could do this job: it can easily handle voltages
up to 60V and directly drive the two mosfets. Nevertheless
it has some drawbacks: mainly it is not born to be
current-programmable, therefore I am not sure I will be able
to set the current limit from the micro; then ot needs 12V to
be supplied; and finally it seems to be quite old and very
expensive (about 7$ + vat !!).
Can anyone confirm this could be the right choice? ...or maybe
suggest me an effective alternative? Perhaps a lower voltage
IC connected to an high voltage mos-driver (like a IRF2110
from International Rectifier)?

I am going to post these latter questions as standalone
message. Thanks again for help.
Click to expand...
There are tons of controllers out there. I would start out with a check
list of everything you need to keep track of and go from there. Just
off the top of my head there is array voltage, batt voltage, batt
current, ambient temp, maybe even time of day for prdictive purposes.
You might find that using the right micro you don't need anything but a
little signal conditioning on the analog inputs and a drive IC to fire
the FETs yourself. Remember, you pay for hardware everythime you make a
unit. You pay for software once.

Jim
 
In article <e364d9033f881b3c596f18a6ddde2ba3.68471@mygate.mailgate.org>,
dfabri73@yahoo.it says...
The micro is already choosen and will be hard to change it: it is an
AVR ATMega8 that has 2 PWM at 14kHz maximum frequency: I cannot drive
a 30V-10A buck converter at 14kHz because such a low frequency will
require an expensive bulky inductance. Then a drive IC will not be
much more cheaper than the whole LT1339.
Let me know if you do not agree with my position.
Thank you
Click to expand...
Yep, if you are locked into a particular MCU that could hobble you a
bit. You might want to take a look at some of the National controllers
too. Have fun.

Jim
 
The micro is already choosen and will be hard to change it: it is an
AVR ATMega8 that has 2 PWM at 14kHz maximum frequency: I cannot drive
a 30V-10A buck converter at 14kHz because such a low frequency will
require an expensive bulky inductance. Then a drive IC will not be
much more cheaper than the whole LT1339.
Let me know if you do not agree with my position.
Thank you



"James Beck" <jim@reallykillersystems_INVALID_.com> wrote in message
news:MPG.1a8c28d5c925e4b498993f@news.west.earthlink.net


There are tons of controllers out there. I would start out with a check
list of everything you need to keep track of and go from there. Just
off the top of my head there is array voltage, batt voltage, batt
current, ambient temp, maybe even time of day for prdictive purposes.
You might find that using the right micro you don't need anything but a
little signal conditioning on the analog inputs and a drive IC to fire
the FETs yourself. Remember, you pay for hardware everythime you make a
unit. You pay for software once.

Jim
Click to expand...



--
Posted via Mailgate.ORG Server - http://www.Mailgate.ORG
 
"Davide Fabri" <dfabri73@yahoo.it> wrote in message news:<faf4eee2a830fa832f63292e2e3751fa.68471@mygate.mailgate.org>...
I have to design a lead-acid battery charger for my company. The source
is power coming from a solar panel array (up to 45V). The output must be
able to drive at least 10A at 30V (overcharge mode for a 24V battery). I
think the UC3909 from Unitrode could be a good choice, but the overall
system has to be controlled by a microcontroller, that should provide a
presettable current limit in trickle and bulk charge mode. Unfortunately
such IC requires a resistor network that defines a fixed current value.
I need help to design an interface that allows the current limit to be
set by the PWM output of the microcontroller.
Thank you very much for help.
Click to expand...
I would look at the PIC ( MIcrochip ) controllers they are very low
cost and have good support. If you are not a prgrammer and wnat to
demonstrate something overnight look at the STAMP 11 . You can simply
prgm. this in pbasic ( very easy stuff ).
I would also make sure that I added a pulse conditiong circuit to the
charger, it will greatly improve battery life. You can also add a
simple micropower "sun-up" detector system to power the unit only when
it is getting solar power. I have built and use a solar tracking
system to charge lead-acid batteries along these lines and it works
well.
ron
 
In article <590a927d.0402051351.6e56912f@posting.google.com>,
rdoctors@cox.net says...
"Davide Fabri" <dfabri73@yahoo.it> wrote in message news:<faf4eee2a830fa832f63292e2e3751fa.68471@mygate.mailgate.org>...
I have to design a lead-acid battery charger for my company. The source
is power coming from a solar panel array (up to 45V). The output must be
able to drive at least 10A at 30V (overcharge mode for a 24V battery). I
think the UC3909 from Unitrode could be a good choice, but the overall
system has to be controlled by a microcontroller, that should provide a
presettable current limit in trickle and bulk charge mode. Unfortunately
such IC requires a resistor network that defines a fixed current value.
I need help to design an interface that allows the current limit to be
set by the PWM output of the microcontroller.
Thank you very much for help.

I would look at the PIC ( MIcrochip ) controllers they are very low
cost and have good support. If you are not a prgrammer and wnat to
demonstrate something overnight look at the STAMP 11 . You can simply
prgm. this in pbasic ( very easy stuff ).
I would also make sure that I added a pulse conditiong circuit to the
charger, it will greatly improve battery life. You can also add a
simple micropower "sun-up" detector system to power the unit only when
it is getting solar power. I have built and use a solar tracking
system to charge lead-acid batteries along these lines and it works
well.
ron

Don't start that "Pulse Conditioning" arguement again....... LOL.
Click to expand...

Last I checked the US Postal Service said those chargers were bunk.

Keeping a Lead Acid Battery fully float charged is the best defense
against premature failure.

Jim
 
You must log in or register to reply here.

Welcome to EDABoard.com

Sponsor

Back
Top