switching circuit

[mphotis]

Hi there...

I really need help. I am busy designing a battery charger using a microcontroller.

i need help with the attached circuit. When the battery is full i would like to be able to charge it using a far lower current. Im doing this using a simple circuit using two resistors in parralel.

When the battery is charging the switch is open and the current is running through the one resistor...this gives me a large current.

When the battery is full the microcontroller sends a voltage to a switch which causes the current to reduce(please see the circuit).

I want to know what switching device i must use and how i am going to use it...please help.

Thanx

View attachment 35774

 

[audioguru2]

Hi Mphotis,
You have Ohm's Law backwards. A single resistor passes a current depending on the resistor's value (ohms) and the voltage across it. A resistor that is in parallel will create a lower resistance (since each one passes its own current), and therefore together they will pass a higher total current.
A simple electronic switch for your circuit is a PNP transistor with a current-limiting resistor in series with its base. Your microcontroller will drive that resistor to near ground, for a high charging current. The microcontroller will drive that resistor to the positive supply, for a low charging current.

When the battery (what type?) is being charged with a high charging current, how can you sense that it is fully charged?

 

[mphotis]

Thanx alot...sorry im a bit slow with this sort of stuff.

Im using a 12V NiCd battery. IM going to have a voltage divider network accross it to provide a reference range of 0-5volts. This will then go into an a/d converter which will convert the voltage to a 8-bit digital value(0-255). This will then go to a microcontroller which will compare this value with a full charge reference value. If the values are within range of eachother then the micro will switch the charge to a trickle charge of small current. Thats why i want the switching circuit.

 

[audioguru2]

Hi again,
I'm glad to help you.
Your 12V NiCd will reach a voltage of nearly 15V just before it is fully charged, then the voltage will drop a bit when it actually is fully charged and begins getting hot. Is your micro going to detect that small voltage drop? Most commercial "fast" chargers use that method, and in case they miss the voltage drop, they incorporate a timer and/or a thermal cutoff as backup.

 

[ante1]

Just a thought about timer cut-off: If a cell in a pack goes bad or the pack is not completely empty the timer is not good enough. Peak detect and then thermal cut-off if anything goes wrong is by experience the safest way 8).

 

[MP1]

...then the voltage will drop a bit when it actually is fully charged and begins getting hot. Is your micro going to detect that small voltage drop?

Now there is a confusing 'bit' of information, excuse the pun.
What 'bit' will be dropped? LSB or MSB or one in between? Or is this a different kind of technical term? How much voltage is in a bit for you? That would be about 5V on my micros... ;D

MP

 

[audioguru2]

Hi MP,
Actually, I used the pun on the word,"bit" on purpose.
With the micro using only an 8-bit ADC, the LSB is 19.5mV. If the 12V battery (up to 15V) is scaled to a 5V range, then when its charging voltage drops one bit (LSB), the battery is already overcharging. That is why some charger circuits miss the voltage peak.
Another method to detect a Ni-Cd battery's amount of charge is to measure its voltage "rate of rise", which is maximnum when the battery is 90% charged.
A good article from our Articles section is here:
http://www1.electusdistribution.com.au/images_uploaded/nicads.pdf

 

[ante1]

Hi Guys.

Why not build an analogue peak detector and let the microP get the signal when reached?

 

[audioguru2]

Hi Ante,
I think that it would be difficult to make an analogue peak detector since the peak value will change depending on the temperature of the battery and its internal resistance, etc. It needs to be done dynamically, using sample-and-hold. Even a sample-and-hold is difficult to make for a hold time of about 5 minutes or more. A FET-input opamp leaks current, especially when warm. A large-value capacitor also leaks.
It is probably best to offset-voltage shift the peak voltage and amplify it, so that the peak voltage and its drop fits within the range of the micro's ADC. Since the voltage is amplified instead of simply attenuated to fit the ADC's 5V window, the LSB becomes neglegible.

 

[ante1]

Hi Audioguru,

I see no problem with an analogue peak detector, I am not sure what you mean? I have built several chargers with peak detection and never had any problem not even when charging different number of cells with the same charger. Her is one I have built a long time ago as an example.

View attachment 35801

 

[hosiufai]

Hi Ante & everybody,
I am new & looking for analogue peak detection charger for simple & low cost. Can you post more information about the cct you have tried here (or any other good cct). It looks not a complete cct.
Thanks

 

[audioguru2]

Hi Hos,
Welcome to our forum.
Go to www.maxim-ic.com They have some nice battery charger ICs.

 

[ante1]

Hi hosiufai,

Which part of the circuit are you referring to as missing?

 

[hosiufai]

Hi Ante,
For example, where is R1 going? Is it V+. RV1 & RV2 seems for current adjust. what value is suitable & the current range. The supply volt. No of cells possible to charge. Do you mind a briegf description of the cct.
Thanksssssss.

 

[ante1]

Ok I see what you mean I forgot to mark the rails and the optional input on R1. Well R1 is an optional input if you want to use a temperature switch to protect the pack. The temp switch (if used) goes between the positive rail (top) and the resistor R1 and should be of the NO (normal open) kind. I the temperature exceed the trip temp D1 is shorted and Q2 will shut down. If you don

 

[hosiufai]

Thanks a lot, Ante. I will try.

 

[Kevin Weddle]

I would suspect the reason for the microcontroller in the application of a battery charger would be to control the voltage applied to the battery. Neglecting the current, there is a voltage that must be applied to the battery depending on the energy already stored. The predestined goal of the charger is of course to make sure the battery delivers the steady voltage you want with the load attached. So for my better lack of understanding when it comes to the operation of the battery, I do know that you need a steady voltage from it.

 

[audioguru2]

No, Kevin:

I would suspect the reason for the microcontroller in the application of a battery charger would be to control the voltage applied to the battery. Neglecting the current, there is a voltage that must be applied to the battery depending on the energy already stored.

Batteries charge with current, not voltage. You could charge a battery from 1 million volts if you limit the current to the recommended value, and stop the current at full-charge.

The predestined goal of the charger is of course to make sure the battery delivers the steady voltage you want with the load attached.

1) A battery doesn't deliver a steady voltage. Its voltage drops as the battery is discharged.
2) Also, rechargeable batteries have a higher voltage when nearing full-charge.
3) The voltage of a battery also depends on its load current because it has its internal resistance in series with the load.