A Low Cost Thermal Peak Detection NiCd Charger

The electric model aircraft and car industries have produced a bewildering array of field chargers for NiCd motor battery packs. These range from simple 6 or 7 cell chargers consisting of a resistor and mechanical timer, to more complex chargers with peak detection, cycling, and the ability to handle 36 cell packs. The resistor/timer type of charger is cheap, but it is has two drawbacks: it might not fully charge the pack in the allotted time, or it might overcharge the pack. The more complex chargers have none of these drawbacks, but they are very expensive.

The charger described in this article can charge packs of 4 to 7 cells with capacities ranging from 600mAh to 2Ah. The charger automatically begins charging when a pack is connected. It charges at a (nearly) constant current (adjustable), and terminates the charge when the pack begins to get warm (a NiCd pack begins to warm up when it has reached full charge). An LED indicates that charging is in progress.

The Circuit

The circuit for the charger (Figure 1) is simple. Z1A is a comparator which compares the voltage on its two inputs and produces a high output when the "+" input (pin 5) is higher than the "-" input (pin 6), and a low output otherwise. R1 and R2 form a voltage divider, presenting a fixed voltage (about 7V) to pin 5. A pair of thermistors (TR1 between the PAK+ and TEMP terminals, and TR2 between the TEMP and GND terminals; see Figure 2) form another voltage divider which presents a voltage to pin 6. This voltage is proportional to the temperature difference between TR1 and TR2. When TR1 is within 10°C of TR2, this voltage is below 7V, and the output of Z1A turns on Q1. This causes current to flow through Q2, R5, and R4, turning Q2 on. This in turn causes Q3 to conduct, resulting in current flow through the NiCd pack connected between the PAK+ and PAK- terminals. The amount of current flowing through Q3 (and thus the pack) is determined by the current flowing through Q2, which in turn is determined by the setting of R5. The current can range from about 2A to 5A. While Q1 is on, current will also flow through LED1 and R6, thus illuminating LED1 to indicate that charging is in progress.

Figure 1. Charger schematic. Click to enlarge.

As the temperature of TR1 rises, the voltage at pin 6 rises. When the temperature of TR1 exceeds that of TR2 by 10°C or more, pin 6 will exceed 7V, and Z1A will turn off Q1, Q2, and Q3, terminating the charge. Z1B performs the same comparison as Z1A, but it's output is used to provide hysteresis (the current flowing into the base of Q1 pulls Z1A's output too low for it to perform this function). When Z1B goes low, current flows through R7, lowering the voltage at pin 5 to about 1.4V. This ensures that the charger will not switch back on as the pack cools off (unless it cools to about 50°C below ambient). The only way to restart the charge is to disconnect the pack being charged. This will disconnect TR1, causing pin 6 to go to 0V, which will turn the charger back on. Since there is now no pack connected, no current will flow through Q3, even though the CHARGE LED is lit. When a pack with a sufficiently cool TR1 is plugged in, charging will recommence.

TR1 and TR2 are identical thermistors. Their resistance is 10kΩ at 25°C, and the resistance increases or decreases by about 4% for each 1°C fall or rise in temperature (the actual rate of decrease and increase varies with temperature). TR2 is installed in the charging cable near the charging plug to measure ambient temperature. TR1 is installed in the battery pack to measure pack temperature. By using two thermistors, the charger will shut off based on the temperature rise instead of the absolute temperature (otherwise the pack will be overcharged on a cold day or undercharged on a hot day).

Connectors and Cables

When I first built this charger, I used 4-pin computer power supply connectors for charging my packs. My packs were all wired permanently into my planes, so this connector does not need to handle the motor current. Even if you connect your packs to your planes with connectors (eg. Sermos), each pack could have a separate charge connector, thus reducing wear and tear on the more critical power connectors. I use the connector with the male housing (and female pins) in the plane, and the female/male connector on the charger. Computer power supply splitter cables are a good source of these connectors. I use the red and yellow leads for the battery + and - connections, and the two black leads for the thermistor (TR1).

Figure 2 illustrates how the cable, connectors, and battery pack should be wired.

Figure 2. Cable schematic. Click to enlarge.

Note that the order of the wires on the left side of Figure 2 does not correspond to the order of the outputs on the right side of Figure 1. Refer to the output and wire names when making up the cable! The PAK+ and PAK- conductors should be 18ga or 16ga, since they must handle up to 5A. The GND and TEMP conductors can be thinner since they handle less than 3mA.

For the 12V+ and 12V- inputs to the charger, use a two conductor 18ga or 16ga cable terminated with large clips suitable for connection to a car battery. Lamp cord is good for this.

As the diagram implies, I had a separate thermistor in each of my packs. Later, I moved away from having a charge connector on each pack, and I made a single charge connector that plugged onto the end of my charger cable. This connector had a pair of Sermos connectors for charging, and a short lead with the thermistor on it for temperature sensing. This required that I install a normally-open START push-button between the TEMP and GND terminals, since the charger will not reset automatically with TR1 permanently connected. To charge using this method, you have to remember to insert the temperature probe into the pack.

Construction

The circuit is designed to be installed in a Radio Shack® project case (see Figure 5 and parts list). Any suitably sized enclosure with a metal lid (or an all-metal enclosure) will do. R5 is glued to the component side of the board, with short lengths of wire connecting it to the appropriate pads. The lid of the case is drilled for R5 and LED1. The potentiometer is then installed in the appropriate hole, and this holds the board in place inside the case.

Here is the printed circuit pattern for the charger:

Figure 3. Copper side. Actual size is 1.9" x 1.6"

The following diagram illustrates component placement on the board:

Figure 4. Component side.

Solder short lengths of wire to the appropriate terminals of R5. Glue R5 to the board, ensuring that R5's shaft is in line with the holes for LED1, and solder the leads. Install LED1, paying attention to polarity. The negative lead (usually indicated by a dot, flat spot, or shorter lead on the LED) is furthest from R5. The LED should be installed so it is high enough above the board to protrude through the corresponding hole that you'll make for it in the case.

Transistor Q2 should be laid flat on the board. A piece of aluminum channel, about 1.5" long and the width of Q2 should be placed on the copper side of the board, extending past the end of the board. Hold Q2, the board, and the aluminum channel together with an appropriate sized bolt. Ensure that the channel does not short circuit any traces.

Install the remaining components, ensuring that none of them stick up high enough to interfere with the case once the board is installed. Transistor Q1 should have it's rounded side facing R3. I suggest you use a socket for Z1, because it is easily damaged by soldering, and hard to remove if it is damaged. Install the socket, with pin 1 at the top left corner.

Transistor Q3 should be installed on a hefty heatsink on the outside of the case and connected to the rest of the circuit with wires. The emitter of Q3 connects to the point marked E and the base to the point marked B. If the leads are not marked on your transistor, refer to the bottom-view diagram at right for the pin-out (click on the diagram to enlarge it if your browser doesn't show it clearly). Use 16ga or 18ga wire for the emitter-to-E connection.

Connect the charging cable to the circuit. Connect the PAK+, TEMP, and GND leads as marked on the board. Connect the PAK- lead to the collector of Q1 (the case). Connect the supply leads to the circuit at the points marked 12V+ (red) and 12V- (black).

Install everything in the case. Figure 5 shows how the front panel will look. The circuit board occupies the left half of the case. The heat sink for Q3 is installed on the right half of the front panel. The heat sink for Q2 extends past the circuit board, underneath the top-right quarter of the case. LED1 protrudes through a vinyl grommet, providing more contrast.

Figure 5. Panel layout.

Testing and Calibration

Connect the power leads to a 12V source (eg. a car battery). The LED should light immediately. Connect a 50kΩ potentiometer between the PAK+ and TEMP leads, with the resistor set at the half way point. The LED should stay lit. Slowly decrease the resistance. When the resistance reaches approximately 10K (assuming 20°C room temperature), the LED should go out. The LED should stay off even as you increase the resistance again. Temporarily disconnecting and reconnecting the potentiometer should cause the LED to light once again.

To calibrate the charging current, use an ammeter in line with the PAK+ lead (an extra pair of 4-pin connectors is handy for this). Monitor the current when charging a depleted pack (the current will reduce towards the end of the charge, especially when charging 7 cell packs). Note the settings of R5 required for different currents and mark them on the case if you wish.

When using the charger for the first time, monitor it carefully. Feel the pack from time to time (don't touch the thermistor though or your body heat will terminate the charge). The pack should barely start to warm up before charging stops.

Parts List

The following table lists all the parts needed. Radio Shack® part numbers are provided for those parts available there. The thermistors used in the prototype are from Radio Shack, although almost any thermistor with a 3% to 5% resistance drop per 1°C temperature rise will do.

Part	Description	Radio Shack®
R1	33kΩ ¼W	271-1341
R2	47kΩ ¼W	271-1342
R3, R4	10kΩ ¼W	271-1335
R5	50kΩ small potentiometer*	NA
R6	470Ω ¼W	271-1317
R7	4.7kΩ ¼W	271-1330
C1	0.1µF 50V	272-1069
LED1	High-brightness LED	276-87
Q1	2N3904, 2N2222, or equiv.	276-2009
Q2	TIP42, MJEF34, or equiv.	276-2027
Q3	2N3055 or equiv.	276-2041
Z1	LM393 dual comparator	NA
TR1, TR2	Thermistor	271-110
Miscellaneous	4" x 2 1/8" x 1 5/8" case*	270-231
	2" x 2" TO-3 heat sink*	NA
	knob to fit R5	274-403
	vinyl grommet to fit LED	64-3025

*Notes: If you cannot obtain an appropriately sized potentiometer, Radio Shack 271-1716 will do, but you'll need a larger case. You can also use a 1MΩ potentiometer, which will provide a wider current range (about 1A to 5A). If you cannot find an appropriate heat sink, you can get away without it if you use an all-metal case such as Radio Shack 270-239.

Parts not available at Radio Shack can be ordered from electronic supply houses such as Sayal Electronics or Digikey.

Addendum After using the charger for some time, I've found that the TIP42 PNP transistor (Q2) gets rather hot. This can result in thermal runaway, wherein the transistor refuses to switch off. There are two possible solutions: One is to mount the transistor on a proper heat sink, external to the case (and insulated from the case). The other solution is to install a 47Ω or 50Ω 10W resistor in the connection between Q2 and Q3. This will reduce the power dissipated by the transistor to about 0.5W (with the physically much larger resistor dissipating the remainder). The resistor can be mounted on the front panel, with the leads going into the enclosure through notches filed in the sides of the panel (insulate the leads with some heat­shrink tubing so they don't short against the panel). I've used the latter approach, and Q2 now barely gets warm.

I'd Like to Hear from You

If you build this circuit (or not), let me know what you think. If you have problems, I may be able to help you, but be sure to supply a detailed description. I can be reached at stefan@capable.ca.

Other R/C Electronic Projects

• LED Bargraph Optical Tachometer
• Analog Bargraph Expanded Scale Voltmeter
• High Speed NiCd Charger for Electric R/C
• BattMan II: A Computer Controlled Battery Manager
• Versatile Miniature High-Rate ESC with BEC and Brake
• Miniature High-Rate Speed Control with BEC
• Miniature High-Rate Speed Control with Brake
• On/Off Motor Control with Brake
• Getting the Most from Your Radio

 

Frequently Asked Questions Q: Can this charger be modified to charge more than seven cells? A: The short answer is "no". In order to charge seven cells, you need about 11.9V input (about 1.6V per cell, plus 0.7V overhead for losses within the charger). To charge more cells, you need 1.6V more per additional cell. Chargers that charge more than 7 cells from a 12V lead-acid battery contain an internal voltage boosting circuit. This takes the 12V from the battery, and steps it up to a sufficiently high voltage for the pack being charged. It's not hard to design a voltage booster for low currents, but to design one that can put out 5 or 10 Amps is not easy, especially since the currents on the input side of such a circuit are even higher (for example, to charge 21 cells at 8 Amps requires drawing 24 Amps from the 12V source battery). I currently have no plans to design a charger for 8 or more cells. Q: Can I charge nickel metal hydride (NiMH) cells with this charger? A: Not very well. Unlike the endothermic chemical reaction involved in NiCd charging, NiMH charging is exothermic. This means that the reaction itself produces heat instead of absorbing heat. Instead of the reaction absorbing the heat produced by resistive losses, it will contribute its own heat. Therefore, the cells will begin to get warm almost immediately. At any significant charge rate, this charger will stop charging long before the pack is fully charged. Q: Can I connect this charger to my car's cigarette lighter socket? A: No! A cigarette lighter provides a very poor connection, which is not suitable for anything more than one or two Amps. If you try to draw more current out, you will probably melt your connector and/or blow a fuse in your car. If you want to use your car battery, connect directly to the battery using a pair of heavy duty alligator clips. Also, don't charge more than about 10 times from your car battery before recharging the car battery. Any more than that and you risk running the battery too low to start the car. If you do succeed in starting the car, you will put a heavy strain on the car's alternator or generator as it attempts to recharge the battery. It's better to have a dedicated deep-cycle lead acid battery for charging your packs from. Q: I built your circuit, and it doesn't work. Can you help me? A: Maybe. I'm providing the information to build this project because I like to share my work. I can't provide detailed troubleshooting, since I only do this as a hobby, and my hobby time is limited. However, if you've built this circuit, and you send me a detailed description of the way in which it doesn't work, I might have an idea or two to help you fix it. There are no guarantees though. Q: Can you send me the parts, or build me a completed circuit? A: Unfortunately not. As I mentioned in the previous answer, I only do this as a hobby, and I don't have the time to collect and mail parts or build circuits for others. Even if you were to pay me to build one, it would cost far more than just going out and buying an equivalent commercial product. Homemade electronics cost more than mass produced products if you factor in the time it takes for construction.

 
 

Buy Stefan a coffee! If you've found this article useful, consider leaving a donation to help support Stefan's Electric R/C Web Site.
Last updated Sunday June 3, 2007.	E-mail Stefan
Disclaimer: Although every effort has been made to ensure accuracy and reliability, the information on this web page is presented without warranty of any kind, and Stefan Vorkoetter assumes no liability for direct or consequential damages caused by its use. It is up to you, the reader, to determine the suitability of, and assume responsibility for, the use of this information. Copyright: All materials on this web site, including the text, images, and HTML mark-up, are Copyright © 2008 by Stefan Vorkoetter unless otherwise noted. All rights reserved. Unauthorized duplication prohibited. You may link to this site or pages within it, but you may not link directly to images on this site, and you may not copy any material from this site to another web site or other publication without express written permission. You may make copies for your own personal use.