Battery Charger Basics.

[lord loh.]

I have been reading all previous threads on Battery Charging circuits and was wondering that in a basic charger circuit which requires current to charge, how do I maintain the current,

If I place a current limiting resistor, shall not form a potential divider with the cell and give a low voltage drop across the charging cell?

I am assuming that I can consider the resistance of the cell large as I am connecting it parallel to the charging source. The current form the target cell shall oppose the current from the source cell. And as the target cell charges, it's voltage level shall incerease and so shall it's current delivering capacity. And it's "resistance" shall further increase. Is my assumtion correct?

So do I need to decrease the voltage output across the terminals of the source as the target battery charges?

And how much current would a NiCd cell need for fast charge(1/4 rated?) and trickle charge(1/10 rated?)?

Please guide me...

Thank you.

[Oh! sorry, I wanted to post this in the general electronics Chat. Can a moderator move it there for me? Thank you.]

 

[Russlk]

Think of the target cell as a large capacitor whose internal series resistance is very low and which has an initial charge equal to the cell voltage.

 

[audioguru]

The rechargable cell's internal pressure, temperature and voltage rises as it is charged.
Energizer don't make Ni-Cads anymore but have their datasheets, a tutorial and the same for their Ni-MH cells. They say that many conditions will cause a cell to vent or explode if charging is not stopped when a cell is fully charged at a fast rate.

 

[k7elp60]

I have built many chargers for NI-CAD and NI-MH batterys. The easiest way to limit the current is with a 3 terminal regulator like the LM317.
Connect a resistor between the output terminal and the adjust terminal. The adjust terminal is connected to the + of the battery to be charged.
R=1.25/I charge current. I have found that if you set the charge current to C X .095 you can leave the charger on forever at room temperature and you will never overcharge the battery. C= the capacity of the cells.
The input to the LM317 needs to be about 5V above the full charge voltage of the battery. A heatsink may be required for the LM317. I have used the LM317LZ for charge currents less than 100Ma and used the LM317T for charge currents up to 1A.

 

[lord loh.]

Okay, I tried to charge the battery using a constant current source.

My target is two Ni-Cd 700mAh cells in a series, I used a 23.5ohm resistor in a series and the source voltage of 3v. The initial voltage was2.29v and my target is (1.35v x 2) = 2.7v

The initial charging current as measured was 26.7mA and reduced to 16mA within the first 10 minutes. This reduction included a lot of fluctuations. The current reduced to as low as 14mA and increased to 20mA during this period. After this the charging current decreased steadily. The battery temperature remained at room temperature.

After 40 minutes, the target battery voltage was 1.62v and current 14mA.

Can anyone explain me this phenomenon and enlighten me with their comments?

 

[lord loh.]

Okay, here is a graph of the experiment.

 

[Nigel Goodwin]

Can anyone explain me this phenomenon and enlighten me with their comments?

You should charge NiCd's with constant current, you were using a VERY un-constant current.

For simple chargers a resistor is all you need, BUT you need a much higher voltage feeding the resistor to make the current more near constant. A simple constant current source is JUST a resistor fed from a high voltage - the higher the voltage the more nearly constant the current is!.

 

[chemelec]

Okay, I tried to charge the battery using a constant current source.

My target is two Ni-Cd 700mAh cells in a series, I used a 23.5ohm resistor in a series and the source voltage of 3v. The initial voltage was2.29v and my target is (1.35v x 2) = 2.7v

The initial charging current as measured was 26.7mA and reduced to 16mA within the first 10 minutes. This reduction included a lot of fluctuations. The current reduced to as low as 14mA and increased to 20mA during this period. After this the charging current decreased steadily. The battery temperature remained at room temperature.

After 40 minutes, the target battery voltage was 1.62v and current 14mA.

Can anyone explain me this phenomenon and enlighten me with their comments?

The LM317 is fine, But Your Source Voltage is TOO LOW.
Use 6 to 12 volts.
With only 3 volts in, You have lost half that in the Regulator.

 

[audioguru]

An LM317 current source needs about 1.5V for the minimum voltage across the regulator at low current and an additional 1.25V across the current-setting resistor, for a total of 2.75V more than the output voltage.
Since two Ni-Cad or Ni-MH cells are nearly 3V when fully charged then an LM317 current regulator will need a min input of 5.75V.

 

[lord loh.]

So what sould my terminating voltage be for a NiCd? I am currently thinking of 1.35v.

Or should I charge the battery for 14h without bothering about the terminating voltage.

My application requires a fast charge. I am supposed to be charging robots at a docking station. So Should I apply a 1C charging current for 2h (by timer) and then switch indefinitely in trickle mode thereafter?

What if the robot docked is not discharged fully? In this case is it not unwise to start charging using 1C for 2h?

On some site I read that NiCds give a constant 1.2v till they exhaust their power fully then the voltage drops abruptly. And charging is an endothermic process and the joule heating and the endothermic reaction cancle out each other and the battries are at room temperature once the battries are fully charged, they start joule heating. So it appears that temperature sensing is the best way of terminating the charging process. Have I got all this correct?

Please help.

Thank you.

 

[Nigel Goodwin]

So what sould my terminating voltage be for a NiCd? I am currently thinking of 1.35v.

I suggest you google for NiCd chargers (plenty of info out there), you can't use a terminating 'voltage', when the battery if fully charged, charging it further cause the voltage to DROP - it's this drop you need to detect.

 

[audioguru]

So what sould my terminating voltage be for a NiCd? I am currently thinking of 1.35v.

Look at Energizer's graph I posted. A cell is about half-charged with a very low current when its voltage reaches 1.35V. Over and over, Energizer recommends sensing the cell's temperature rise as it approaches full charge to reduce the charge rate.

My application requires a fast charge. I am supposed to be charging robots at a docking station. So Should I apply a 1C charging current for 2h (by timer) and then switch indefinitely in trickle mode thereafter?
What if the robot docked is not discharged fully? In this case is it not unwise to start charging using 1C for 2h?

Use a timer only if you are absolutely certain that the celss are discharged. If you fast-charge a charged cell then it will melt, vent or explode.

On some site I read that NiCds give a constant 1.2v till they exhaust their power fully then the voltage drops abruptly.

A cell has 1.2V only with a certain amount of load current. Energizer's datasheets show the voltage drop when they run down.

And charging is an endothermic process and the joule heating and the endothermic reaction cancle out each other and the battries are at room temperature once the battries are fully charged, they start joule heating. So it appears that temperature sensing is the best way of terminating the charging process. Have I got all this correct?

Just read Energizer's Applications Tutorials from their website.

 

[lord loh.]

-ve delta.

I suggest you Google for NiCd chargers (plenty of info out there), you can't use a terminating 'voltage', when the battery if fully charged, charging it further cause the voltage to DROP - it's this drop you need to detect.

I am doing that now. But the trouble is measuring the voltage... I am using a charge and rest method. I plan to implement it using a micro controller later... As of now I am charging and resting the circuit using a 85% duty cycle generated from a function generator.

Every 5 minutes I am switching off the function generator and measuring the battery voltage. When I do this, the battery voltage falls rapidly for the first 5 seconds and then stabilizes....

When I move on to a microcontroller, should I give a 5 second rest before I measure the battery voltage?

https://www.mpoweruk.com/chargers.htm suggests a 3 second charge and a 20 to 30 ms rest... And I have to do my voltage detection using a microcontroller in this period. If the voltage does not stabilize, will my readings not be erroneous? Can I still in this case depend on the 50ms/cell drop?

Please help...

 

[Reynard]

Hi

i know that it is better to use a constant DC voltage to charge a battery. However if i have a random voltage supply consisiting of impulses, am i able to charge up a battery?

Cheers

You should charge NiCd's with constant current, you were using a VERY un-constant current.

For simple chargers a resistor is all you need, BUT you need a much higher voltage feeding the resistor to make the current more near constant. A simple constant current source is JUST a resistor fed from a high voltage - the higher the voltage the more nearly constant the current is!.

 

[lord loh.]

Constant Current and not constant Voltage.

Hi

i know that it is better to use a constant DC voltage to charge a battery. However if i have a random voltage supply consisiting of impulses, am i able to charge up a battery?

Cheers

You require constant current and not constant voltage to charge NiCd and NiMH type of batteries.

There is a method called charge and rest to charge cells that way... Aslong as the current is constant... But a erratic current supply is not desirable.