Help on a Ni-Cd Charger

[Ravi]

Hi Firends,

I am building a Ni-Cd charger to be given to Tsunami victims for their pocket FM radios. I have designed the ckt but would like to know whether it is ok before I actually attempt to build it. Please see the below description and the circuit and give me your views/ideas. Sorry it is little long....

The center tap transformer (T1), Diodes D2,D3 and IC7806 will supply the power to the circuit. Two Ni-Cd’s are been charged by a constant current source based on PNP transistor BD140 (Ic-max 1.5A, V max 80V, P max 12.5W) which is controlled by a voltage comparator LM393.

While the cells are being charged, out put of the comparator is low and the transistor is ‘on’. LED will now glow indicating that cells are being charged. Once the cells are charged to their 80% or more capacity, their terminal voltage rises to about 1.45volts, the trigger threshold of the comparator which is set by R3 POT at around 2.9 volts (1.45 X 2), exceeds the output of comparator goes high. Now the transistor is ‘off’ and charging ceases.

The capacitor C3, controls the comparator toggling, when the battery voltage is too low. Once the charging ceases, the e.m.f of the battery drops almost instantly. This might cause the charging circuit to be switched on again, resulting in a charging voltage rising to the level set with R3. To prevent this oscillatory condition, the capacitor across the comparator enables the battery to stabilize. If after a short delay, the battery voltage proves to be too low, the current is switched on again. The capacitor then assures that the current will flow for a while, irrespective of the battery e.m.f. Can a feed back resistor be used to get the same results?

Switch S1 can set between 15 ohms and 6.8 ohms resistors to have 14 hours and 5 hours (fast charge) respectively. The reference voltage for constant current source is obtained from the forward voltage drop of the LED, which is around 1.5 volts for a red LED. V(BE) for a Si transistor is around 0.65 volts. Then voltage drop across R2 or R7 is V (LED) – V (BE). This is 0.85 volts. Thus formula for charging current becomes 0.85/R2 or R7. With R2=15 ohms, the current is around 57mA or 125mA if R7=6.8 ohms. Is this sufficient for a AA size cell of 600 mA to charge 14 hours or 5 hours?

In order to have a audio indication, the high logic state of the comparator is fed to the CMOS IC 4011 B to make a pulsed-tone alarm.

 

[audioguru]

Hi Ravi,
You can't use the negative feedback provided by C3 with a LM393 comparator. When the comparator's output becomes linear, it will oscillate like crazy!
Instead of using C3 or using a compensated opamp instead of the comparator, try using hysteresis, which is positive feedback that forces the comparator's output to snap quickly through its linear region. You can design the hysteresis voltage gap so the pot can be set for the transistor to turn off when the battery voltage is 1.45V, and the hyteresis will turn on the transistor when the battery voltage is, say 1.3V.
How to do hysteresis is described in the LM393's datasheet. You add a very high-value resistor from the comarator's output to its positive input. Then you add a series resistor from its positive input to the battery. You probably also must add a pull-up resistor to the output, which will also help the oscillator's gate.

The wiring around R6 looks strange. R6 and the ground aren't needed.

 

[chemelec]

Without Building it and testing it, it looks fairly reasonable.
However, I wouldn't trust it on Older Batteries.

their terminal voltage rises to about 1.45volts

This is True for a New Battery, but can be Less on older Used Batteries.

Take care...........Gary

 

[Ravi]

Thanks very much friends for the advise.

Yes audioguru, I could use a 741 op-amp or positive feed back. I will look for the data sheet. As you suggested I will remove the ground from R6. Please comment on R2 & R7. Are the values OK to charge 2 AA size cells for 14 hrs and 5 hrs fast charge?

Chemelec's view on the subject, so you mean to say that with older batteries the charger will not cease charging since they will not trigger the threshold of the comparator? Any alternative?

The reason for building this charger is that I am a member of a social club and we have given 50 FM radios to people in tsunami affected areas in my country. We decided to give Ni-Cds and chargers. I have a local manufacture who has very kindly agreed to built these free provided we give a suitable circuit. So the circuit has to be cost effective also.

 

[audioguru]

Hi Ravi,
An old 741 opamp won't work on only a 6V supply, but a dual LM358 opamp will, and it has the same input spec's and pin numbers as a LM393. Like a 741 which is much worse, the output of an LM358 doesn't swing completely to ground like an LM393.

Charger manufacturers don't use a fixed voltage threshold to terminate charging. Each battery cell will have a slightly different voltage and some might not reach 1.45V. Therefore they will keep on charging in your circuit and maybe vent.
They use a method to detect a peak in the charging voltage and use a timer and/or a temperature probe as a backup in case the peak-detect method fails. Ni-Cads are obsolete now and have been replaced by similar Ni-MH cells but I found a manufacturer's Ni-Cad charging recommendations. Their Ni-MH charging recommendations are also there.

Energizer's website is packed with batteries and chargers. Select Datasheets. https://data.energizer.com/