# Charging circuit for bulk 1.2v 1200mah ni-cad batteries

I am seeking advice about how to construct a charging circuit for 1.2v 1200mah ni-cad batteries. I have around 60-70 of these batteries and I want to create a bulk charging circuit so I can charge 5 or more (preferably 10 or so) at once while still being able to charge only 1 at a time if that is all that needs to be charged.

I found this schematic:

about the math behind the charging for the original posters circumstances, but I am wondering if I can simply parallel the load to multiple batteries to charge many at once. I am sure the LM317 can handle 10 batteries at 120ma charge rate.

Overall question: Is it possible to use that schematic and parallel the load to multiple batteries? Also is this math correct as per the following details:

Battery         -> 1.2v 1200mah
Charge rate     -> 0.1C*1200mah = 120ma
Charge time     -> (1.4 * 1200mah) / 120ma = 14 hours
R1              -> 1.25v / 0.120ma = 10.416Ω
DC supply input -> (1.2v * 1cell) + 2v = 3.2v or higher


How many watts should R1 be in this situation? Is it also safe to increase the charge rate from 0.1C to 0.2C or 0.3C? I don't have any way to monitor the batteries temperature. These batteries were manufactured in 2011 so they are pretty new (if it makes a difference).

• useful background information: Charging Nickel-cadmium. – Nick Alexeev Feb 5 '16 at 0:48
• @NickAlexeev Thanks for the post it was insightful although I won't be implementing any of those features for this bulk battery charger. I will simply trickle charge them at a low rate for a predetermined time in which a timer will cut the mains to the charging circuit. – randy newfield Feb 5 '16 at 1:25

How many watts should R1 be in this situation?

Watts = V2/R or I2*R, which works out to 0.15W. In order to keep surface temperature down and increase reliability you should double that to get the resistor rating, ie. 0.3W or higher.

Is it also safe to increase the charge rate from 0.1C to 0.2C or 0.3C? I don't have any way to monitor the batteries temperature.

Charging at much higher than 0.1C may cause the batteries to overheat once they reach full charge. However it should be safe at 0.2C if you time the charge and stop when a little over 1200mAh has been put in. This is recommended even at lower charge rates, because Nicads suffer voltage depression when overcharged.

Is it possible to use that schematic and parallel the load to multiple batteries?

The cells must be wired in series, not parallel. Each cell will reach ~1.5V at full charge, and the regulator needs about 3V of headroom, so with 10 cells in series your power supply will have to deliver at least 18V.

Now what happens when you try to charge a single cell with the same power supply? The regulator then has to drop 15.25V @ 0.12A, so it will dissipate ~18W. This is doable, but a very large heatsink will be required to stop it from overheating. It would be better to lower the supply voltage when charging fewer cells.

• Perfect thank you. I will keep that in mind when charging fewer cells to swap out the power supply with a lower voltage one (I will also have a fairly decent heatsink on it anyways). How much over 1200mah should I charge the batteries. With the link I provided they say 40% over charge should be used. – randy newfield Feb 5 '16 at 1:54
• 40% is too much IMO. I usually only put in the rated capacity, or perhaps 10% more. A new Nicad needs 3~5 charge/discharge cycles to reach full capacity, so you probably won't get 1200mAh out of them on the first charge (and if they are cheap Chinese cells you may never get 1200mAh out of them!). – Bruce Abbott Feb 5 '16 at 2:19

If you're going to charge a string of cells with a constant current, they must be wired in series and you must provide a bulk supply with enough output voltage to supply the headroom and the reference for the regulator, as shown in this excerpt from TI's LM317 data sheet:

For charging current limits check the NiCd manufacturer's data sheet.

The power dissipated by the series resistor will be$1.25V \times Ichg$, and the power dissipated by the LM317 will be $(Vin - (Vbat + Vref)) \times (Ichg + Iref)$.

Note that the LM317 will dissipate more and more power as fewer and fewer cells are connected.

• So if I wanted to charge 10 batteries at once, I would need to connect them in series and provide a Vin for the LM317 equal to or greater than 12v (1.2v x 10) + 4.25v (headroom) = 16.25v? Could you also verify if the rest of my math in the original post is correct and/or answer a few of my other questions? Thanks for the reply. – randy newfield Feb 5 '16 at 1:19
• @randynewfield: Something like that, but I think fully charged NiCads wind up with something like 1.3 volts across them. Best to check the manufacturer's data sheet, or check out Battery University – EM Fields Feb 5 '16 at 1:25

You're in luck. NiCads are extremely easy to charge if you're not in a hurry. Simply charge for 16 hours at a nominal C/10 rate, and don't worry about overcharge. A 1C rate will provide the nominal capacity in 1 hour, so a C/10 rate is 1/10th that. In your case it's 120 mA.

Furthermore, you don't have to be picky about precision current control. If you you have a 12 volt supply, you can use a circuit like

simulate this circuit – Schematic created using CircuitLab

to charge as many sets of 4 as you like. Each set draws 0.12 amps (more or less) so you can charge about 8 sets with a 1 amp power supply. The 62-ohm resistors can in theory be 1 watt resistors, but I'd go with 2-watt units, just in case you've got a shorted battery or two in any given chain.

I recommend checking the each string of 4 for voltage after about 1 hour. If the voltage is less than about 4.5 volts, you've probably got a dead cell in the string, and you want to pull it out. With a shorted cell the limiting resistor will see more than the nominal 7.2 volts, and the charge current will be larger than you expect. This is not a problem in the short run, especially if all the cells are pretty well discharged, but should not be allowed to persist for long.

I recommend starting with 4 cells you know are fully discharged, and then take voltage measurements every couple of hours or so (write them down somewhere) so as to get a rough idea of how recharge proceeds. If you then try to recharge some cells which are only partly discharged, you can get an idea of how the charge process is going by looking at the voltages, and you can stop recharging at a reasonable point. In general, though, NiCads won't be damaged by overcharge at a C/10 rate.