Maker Pro
Maker Pro

Lead Acid Battery Temperature Coefficient??

M

mike

Jan 1, 1970
0
Temperature Coefficient of Lead-Acid Battery??

My friend asked me to design a diversion controller
for a solar/wind powered mountain top radio repeater site.
There's a lot of conflicting info published on the web,
but the consensus from alt.homepower appears to be
~-3mv/cell/degree C.

I don't have access to the actual GelCell batteries involved. And
there's no vendor name that can be contacted. So, I did the
next best (worst) thing and stuck a 12V sealed lead acid
battery of the type used in alarm systems and UPSs in the
fridge. For a delta of 30F/16.66C I saw 40mV of change
in open circuit volts. 18mV/degreeC predicts 300mV.

This discrepancy is too much even for a seatofthepants
guy like me.

What am I doin' wrong?? Anybody got any first-hand experience
with battery tempco?

Thanks, mike
--
Return address is VALID.
Bunch of stuff For Sale and Wanted at the link below.
Toshiba & Compaq LiIon Batteries, Test Equipment
Honda CB-125S $800 in PDX
Yaesu FTV901R Transverter, 30pS pulser
Tektronix Concept Books, spot welding head...
http://www.geocities.com/SiliconValley/Monitor/4710/
 
A

Adrian Jansen

Jan 1, 1970
0
Your measured 40mV for 16 degC equates to 2.5 mV/Deg according to my math.
Seems about right for the 3 mV/Deg quoted.

My Faure manufacturer info for normal ( non gell cell ) batteries says 6
mV/DegC.



--
Regards,

Adrian Jansen
J & K MicroSystems
Microcomputer solutions for industrial control
 
M

mike

Jan 1, 1970
0
Adrian said:
Your measured 40mV for 16 degC equates to 2.5 mV/Deg according to my math.
Seems about right for the 3 mV/Deg quoted.

Your math is different from mine. 3mV/deg/cell * 6 cells in a 12V
battery = 18mV/degree/battery * 16.66 degrees C = 300mV. Yes? No?

I'd expect some difference between different cell chemistries, but
my experiment came up about 1/8th of the low end of the published numbers.
mike
My Faure manufacturer info for normal ( non gell cell ) batteries says 6
mV/DegC.



--
Return address is VALID.
Bunch of stuff For Sale and Wanted at the link below.
Toshiba & Compaq LiIon Batteries, Test Equipment
Honda CB-125S $800 in PDX
Yaesu FTV901R Transverter, 30pS pulser
Tektronix Concept Books, spot welding head...
http://www.geocities.com/SiliconValley/Monitor/4710/
 
T

Tom Bruhns

Jan 1, 1970
0
This is a bit of a round-about answer, but I hope it will be useful to
you.

I got involved in trying to make the backup batteries for a fire alarm
system last longer. They're gel-cell type, 2.5 and 5 A-Hr, 6V. These
alarm boxes are on the porches of buildings which span from about 100F
in the summer to -10F in the winter. We tried some simple
compensation schemes with shunt regulators, but nothing really helped
enough to make them reliable. Then we started using a charger circuit
based on TI/Unitrode UC3906 lead-acid sharge controllers, and it has
worked fine ever since. So I'd say to have a look at the data sheet
and corresponding ap note for that part for ideas about not only the
temperature compensation (which is built in) but charging in general.
FWIW, the internal reference for that part is 2.30V@25C, -3.9mV/C. --
You might also get different results if you test your battery under
moderate load, over a range of temperatures, like maybe C/100. Just
load it for a short time for each sample to avoid significant
discharge. And of course, be sure it really reaches thermal
equillibrium. (You have me just about curious enough to want to run
such a 'speriment myself now!)

Cheers,
Tom
 
B

budgie

Jan 1, 1970
0
This is a bit of a round-about answer, but I hope it will be useful to
you.

I got involved in trying to make the backup batteries for a fire alarm
system last longer. They're gel-cell type, 2.5 and 5 A-Hr, 6V. These
alarm boxes are on the porches of buildings which span from about 100F
in the summer to -10F in the winter. We tried some simple
compensation schemes with shunt regulators, but nothing really helped
enough to make them reliable. Then we started using a charger circuit
based on TI/Unitrode UC3906 lead-acid sharge controllers, and it has
worked fine ever since.

(snip)

Ever since I first used a UC3906 charger on gel cells, I have used nothing else.
The description of the charge regime makes interesting reading, and I would
recommend the device - and the data sheet/app-note - to all who use gel cells.
 
J

James Beck

Jan 1, 1970
0
(snip)

Ever since I first used a UC3906 charger on gel cells, I have used nothing else.
The description of the charge regime makes interesting reading, and I would
recommend the device - and the data sheet/app-note - to all who use gel cells.
I'll third that. I think the UC3906 is great. Works well and has all
the bells and whistles built in.

Jim
 
J

Jim Thompson

Jan 1, 1970
0
Temperature Coefficient of Lead-Acid Battery??
[snip]

Several of my main claims to fame are patents in the area of TC'ing
the charging of lead-acid (car) batteries.

But I can't remember the best TC :-(

If I get a free moment I'll get out a ladder and get a patent down off
the gallery wall and get that information for you.

...Jim Thompson
 
A

Adrian Jansen

Jan 1, 1970
0
Yes, sorry I forgot you are talking about a multi-cell battery. Faure
figures are -6 mV/DegC per cell. Unitrode quote -3.9 mV/DegC per cell.

Maybe it takes longer to cool in the fridge than you allowed ?

--
Regards,

Adrian Jansen
J & K MicroSystems
Microcomputer solutions for industrial control
 
F

Fred Bloggs

Jan 1, 1970
0
mike said:
Temperature Coefficient of Lead-Acid Battery??


The float threshold at C/50 or less charging current is definitely
-2mV/oC. The cycle threshold at C/10 or more is -6mV/oC. This explains
why Unitrode uses -4mV/oC in the '3906. The cycle threshold is that cell
voltage at which charge lost due to a previous discharge is 100%
recovered when re-charging at C/10 or larger currents, it is a function
of the charging current magnitude. The manufacturer should supply the
equivalent of 20oC data. The 20oC float threshold is 2.3-2.35V and the
cycle threshold can be much higher as a function of charging current,
ranging from 2.4 to 2.7V per cell. The Unitrode chip will cut the
constant current re-charge stage short, and float at a slight overcharge
at very low temperatures, but this is okay because the battery capacity
has fallen at 1%/oC and there is no danger of premature failure due to
heat at those temperatures. You will have to use a very oversized solar
panel to make the '3906 work as envisaged by the designers- your charge
profile tapers, with its temperature dependence.
 
M

mike

Jan 1, 1970
0
Adrian said:
Yes, sorry I forgot you are talking about a multi-cell battery. Faure
figures are -6 mV/DegC per cell. Unitrode quote -3.9 mV/DegC per cell.

Maybe it takes longer to cool in the fridge than you allowed ?

Good point, I only left it over night. I can try longer.
mike

--
Return address is VALID.
Bunch of stuff For Sale and Wanted at the link below.
Toshiba & Compaq LiIon Batteries, Test Equipment
Honda CB-125S $800 in PDX
Yaesu FTV901R Transverter, 30pS pulser
Tektronix Concept Books, spot welding head...
http://www.geocities.com/SiliconValley/Monitor/4710/
 
A

Adrian Jansen

Jan 1, 1970
0
I think Fred Bloggs has the answer. There seem to be dramatic differences
between the no current terminal voltage coeff and the voltage coeff when
under charge, particularly at high charge rates.

Batteries, or at least the electrochemical processes in them, are complex
things !

--
Regards,

Adrian Jansen
J & K MicroSystems
Microcomputer solutions for industrial control
 
J

Jim Thompson

Jan 1, 1970
0
Temperature Coefficient of Lead-Acid Battery??

My friend asked me to design a diversion controller
for a solar/wind powered mountain top radio repeater site.
There's a lot of conflicting info published on the web,
but the consensus from alt.homepower appears to be
~-3mv/cell/degree C.

I don't have access to the actual GelCell batteries involved. And
there's no vendor name that can be contacted. So, I did the
next best (worst) thing and stuck a 12V sealed lead acid
battery of the type used in alarm systems and UPSs in the
fridge. For a delta of 30F/16.66C I saw 40mV of change
in open circuit volts. 18mV/degreeC predicts 300mV.

This discrepancy is too much even for a seatofthepants
guy like me.

What am I doin' wrong?? Anybody got any first-hand experience
with battery tempco?

Thanks, mike

See my patents on temperature-controlled alternator regulators:

3,496,447

3,505,590

3,522,482

3,546,563

...Jim Thompson
 
Top