NiCd battery charger

[DemonicSandwich]

Long story short, I don't have a good enough charger for my several 8 cell NiCd battery packs and as the result the batteries (and plane I used them for) went mostly unused for a couple years.

But with a renewed interest (and better knowledge of electronics) I decided built a battery charger for those NiCd packs(circuit attached) using only the components I have on hand and would rather to keep it as such.

From what I've read on NiCds, when charging they peak at a voltage of 1.6V per cell when they reach full charge then start to fall slightly so for an 8-cell pack that peak voltage would be 12.8V.
Another thing that I found was that NiCds respond to charging better if it's pulsed.

The circuit I built uses a constant current circuit found here but uses a power BJT rather than a FET.
I added extra transistors (Q3, Q4) so I can pulse it on and off.
R1 being 1Ω sets the current to about 500-600mA. The power supply is regulated supply rated at 16V, 4A.
The IC is a LM555 with a rated max supply of 16V so I added D3 and D4 to drop it into a safe area. I plan on replacing them with a 6V regulator.

The circuit works and the batteries respond very well to the pulsed charge and peak at ~12.8V as expected.
But my problem is is that it has to be watched. I need a circuit that shuts down the charging circuit(or oscillator) when the battery reaches 12.8V. I was thinking of using an LM1458N OpAmp I have from another project but I am not very familiar with opamps.

I just need something to change states when the battery reaches a set voltage using either that opamp or simple components.

(Image above is attachment)

 

[ronv]

You can use the op amp or you could buy an inexpensive comparitor to sense the voltage. Then you will need to either build a latch out of and gates or use a latch ic to remember it was fully charged after the charger turns off. Might need a switch to reset it or maybe we could reset it when the batteries were removed. Does something like that sound like a plan?

 

[bountyhunter]

Looking for the peak delta voltage to terminate on an 8 cell string is dangerous because they don't peak at the same time. You might never reach to maximum voltage. You are better off monitoring cell voltages and terminate when the cell temps start to rise.

The NI-CD chemical charge reaction is endothermic which means the cell gets slightly cooler (even at high charge rates) so the rising cell temp is a fairly reliable termination method. I used to cut the fast charger at cell temps of about 5C above ambient then shift to a low trickle charge rate of about c/20.

I'll try to find the article I had published on a fast nicad charger.

 

[bountyhunter]

Here is an article of mine about building a fast NI-CD charger with automatic shutoff the prevent overcharge. I think it was published late 90's.

I also included bench data on the voltage/temp profile of a typical ni-cd cell subjected to 1.2C charge rate. It shows where a good point to stop is. On mine, the peak cell voltage only got up to 1.736V.

 

[bountyhunter]

From what I've read on NiCds, when charging they peak at a voltage of 1.6V per cell when they reach full charge then start to fall slightly so for an 8-cell pack that peak voltage would be 12.8V.

The NI-CD cells I bench tested peaked at about 1.736V for a charge rate of about 1.2C and ambient temp of 23C.

 

[audioguru]

My old RC model airplane used a cheap and heavy 8.4V Ni-MH battery. It barely climbed and each charge lasted only a few minutes. I replaced the battery with two Lithium -ion cells in series (7.4V) from an old laptop computer and the lighter battery allowed the airplane to climb well and each charge lasted a long time.

My new RC model airplanes use a lithium-polymer battery that has a lot of power and each flight is about 15 minutes per charge.

 

[DemonicSandwich]

I also included bench data on the voltage/temp profile of a typical ni-cd cell subjected to 1.2C charge rate. It shows where a good point to stop is. On mine, the peak cell voltage only got up to 1.736V.

At one point I placed two 1Ω power resistors in parallel to get about 1A of charge current and got worries when it reached, and exceeded, 13V. It's good to know they naturally reach such high voltages under heavy charge.

My old RC model airplane used a cheap and heavy 8.4V Ni-MH battery. It barely climbed and each charge lasted only a few minutes. I replaced the battery with two Lithium -ion cells in series (7.4V) from an old laptop computer and the lighter battery allowed the airplane to climb well and each charge lasted a long time.

My new RC model airplanes use a lithium-polymer battery that has a lot of power and each flight is about 15 minutes per charge.

My plane is a Great Planes Fundango. It's relies on the battery weight to both set it's center of gravity and give it stabily.

I have 4 packs I use for it. Three of them are AA-size 8-cell NiCds rated 1000mAh. The forth is an 8-cell Sub-C pack that came out of a drill battery. I estimate it's capacity at about 1500mAh but is capable of a much higher drain current that the others.

 

[audioguru]

My newest RC airplanes are a 3-channels ParkZone Micro-J3 Piper Cub that has a wingspan of about 15 inches and it weighs less than 1 ounce including its battery, and a ParkZone Micro-P51 Mustang that has a wingspan of about 15 inches and it weighs slightly more than 1 ounce and has 4-channels including ailerons. The Piper Cub can fly as slow as my trot (it hovers when there is a small breeze) and the Mustang can go extremely fast and climb vertically while I barrel-roll it.
I fly them in a nearby park almost every morning and evening when the wind is fairly calm. I met other "pilots" who bought the same airplanes and we have dogfights.
The radios work at 2.4GHz and automatically select channels that have no interference.

 

[DemonicSandwich]

Thank you bountyhunter for the PDFs. They shall prove useful.

---------

I looked into Opamps again and found it easier to understand than the last time I tried.

It's apparent that I won't get unwatchable charging by relying only on voltage unless it's setup to to detect a fall in voltage so I think I'll go with using that opamp to detect a rise in heat.

But until I get a thermistor, I'll just have to monitor them while they charge.
When I do get a thermistor, this will likely be my circuit.

Edit: Come to think of it I don't think the inputs should be pulled that close to supply.

 

[audioguru]

Come to think of it I don't think the inputs should be pulled that close to supply.

No.
The inputs are over 10V from the supply which is plenty of extra voltage. The supply is 16.6V and the voltage regulator will be 6V if it has an output capacitor and an output resistor to ground.

 

[DemonicSandwich]

I was actually referring to the Inverting input. I set it up to lock itself high by using Q4 to pull the inverting input directly to ground.

 

[audioguru]

The allowed input common-mode voltage range for the very old LM1458 dual opamp is 3V above the negative voltage which is 0V in your circuit. Q4 forces it lower than 3V so the inverting input will stop working and the opamp will not work properly.
An LM358 dual opamp has inputs that work at 0V in your circuit. Its pins are the same as the LM1458.

 

[DemonicSandwich]

The LM1458 actually does work as intended in this circuit despite being out of it's operational parameters though I will pick up an LM358 to ensure it's functionality and reliability.

Another unfortunate thing I discovered is that my AA packs are far too to gone to be airworthy. They deliver less than half their rated capacity at 1C discharge and have developed moderate internal resistance so even freshly charged they don't deliver like they used to.
Though this was not unexpected, they did go mostly unused for nearly 4 years so they will now be used for low demand projects instead.

On the other hand I did discover some very tantalizing prices on LiPoly packs and chargers so I think I shall switch to LiPoly though I will get a dedicated charger for them. No sense taking a chance there.