# circuit to draw a constant current from a lead acid battery

Discussion in 'General Electronics Chat' started by gkasun, May 2, 2015.

1. ### gkasun Thread Starter New Member

May 2, 2015
6
0
I supposed to draw a constant current from a lead acid 12v car battery.
i need a help to implement a circuit for that.
is there any proper circuit ?

File size:
66.6 KB
Views:
50
2. ### Dodgydave AAC Fanatic!

Jun 22, 2012
6,494
1,028
That wont work, use a constant current regulator like a Lm317, or jfet, what current are you wanting to use?

gkasun likes this.
3. ### dl324 AAC Fanatic!

Mar 30, 2015
5,010
1,081
For some reason, that circuit looks to me to be very dangerous to the transistor.

How much current do you want to draw? How low do you expect to run the 12V battery?

gkasun likes this.
4. ### crutschow Expert

Mar 14, 2008
16,573
4,477
Here's a simple constant-current circuit that should work for you.
Note that the top transistor must be a power type rated to carry at least twice the maximum current you want, and must be on an appropriate heat sink for that current and the voltage across it from the battery.
The base resistor carries the required base current for the collector current from the battery based upon the minimum transistor current gain (β or hFE).

gkasun likes this.
5. ### gkasun Thread Starter New Member

May 2, 2015
6
0
wow thank u for quick replies
i didn't think so.
actually i want to draw constant current from a lead battery and graph lead acid battery voltage variation with time.

6. ### Dodgydave AAC Fanatic!

Jun 22, 2012
6,494
1,028
What value of current do you want to draw, 0.5amp , 1amp, 2amps ???

gkasun likes this.
7. ### wayneh Expert

Sep 9, 2010
13,638
4,433
A quick and dirty approach is to just use a headlight. A fully charged battery might start at 13.5V and should not be run down past 12V, maybe 11V if you're pushing it. So the voltage fades 2.5V/13.5V or nearly 20%. The current in the headlight will drop a bit less because the resistance of the filament falls as it cools.

So yes, the current will fade as the voltage drops, but that small error can be estimated and might be insignificant compared to whatever you're testing for, for instance good versus bad battery.

The headlight is designed to shed heat and gives visual feedback.

gkasun likes this.
8. ### crutschow Expert

Mar 14, 2008
16,573
4,477
You don't think so, what?

gkasun likes this.
9. ### ian field AAC Fanatic!

Oct 27, 2012
5,563
985
There are 2 classic text book current regulators that you can find easily online, the first is a basic emitter follower that you include a current sensing resistor in series with the emitter - a second transistor is connected with its B/E junction across the sense resistor, when the current is enough to develop 0.6V across the sense resistor it biasses on the second transistor - its collector is tied to the emitter follower's base and shunts its bias to control the current.

The single transistor variety is a little like the full automatic bias common emitter stage; the base is fed by a voltage from a potential divider, from the emitter resistor's point of view its an emitter follower delivering Vref minus 0.7V - so a fixed emitter resistor will have a fixed voltage and pass a fixed current. Ie=Ib + Ic, so the collector current will be only a tiny bit less than the current in the emitter resistor.

A fixed Vref for the base is better than a potential devider - among the favourites for a fixed volt drop are a series pair of silicon diodes or a green LED.

10. ### upand_at_them Active Member

May 15, 2010
248
29
The EEVBlog had a constant current load project a couple years ago. I think his was even amenable to data logging, as it had an LCD on it.

gkasun likes this.
11. ### WBahn Moderator

Mar 31, 2012
20,267
5,762
This sounds like homework. Just what is the assignment?

gkasun likes this.

May 2, 2015
6
0
0.5A

May 2, 2015
6
0

14. ### gkasun Thread Starter New Member

May 2, 2015
6
0
This is not an assignment.i supposed to compaire different batteries under how drain there current with time.
To do that i need to build a constant current draining circut.
So i need to find a most suitable circuit to do that.
Finaly i hope to draw a graph of each battery voltages with time.

15. ### ian field AAC Fanatic!

Oct 27, 2012
5,563
985
If its a car/caravan battery - you'll need a *HUGE* heatsink.

Usually, I just gang as many H4 headlamp bulbs as I think I'll need. Its not accurate, but gives me a reasonable idea.

The dipped beam filament gets the most wear, so its easy to scrounge scrap bulbs with the HB filament still good.

Don't lay the bulbs on wooden floorboards - they'll scorch!

gkasun likes this.
16. ### wayneh Expert

Sep 9, 2010
13,638
4,433
Yeah, for ~0.5A he'll need a smaller bulb. Headlights are ~4A. A test at 0.5A will be a very long test.

gkasun likes this.
17. ### gkasun Thread Starter New Member

May 2, 2015
6
0
K .its fine.actualy i hav not decided yet that what is the drain current.

18. ### WBahn Moderator

Mar 31, 2012
20,267
5,762
If it's not an assignment, then why are you supposed to do this?

gkasun likes this.
19. ### WBahn Moderator

Mar 31, 2012
20,267
5,762
Well, that really matters.

A circuit that will deliver a 0.5A constant current is very different from one that will deliver 50A of constant current.

If you want one that is scalable, then that isn't too hard.

But you need to decide just what the problem is that you are trying to solve before you proceed to solve it.

20. ### ian field AAC Fanatic!

Oct 27, 2012
5,563
985
AFAICR: battery manufacturers base their Ah capacity on a 20 hour discharge rate - if you discharge a 60Ah battery at 60A, it won't last 1 hour.

IMO: a simple resistor is good enough for the TS requirements - simply calculate the resistance for 1/20 of the Ah rating and check the voltage at 20 hours.

I usually use old headlamp bulbs of known rating - its not accurate, but it gives a fair idea whether the battery does what it says on the tin.

gkasun likes this.