There are two major configurations for this board: you can make it so that it only trickle-charges your cells, or you can add the ability to fast-charge the cells as well. Fast-charging requires more parts, is more complex, and your cells won't last quite as long. For many, the benefit of getting your cells charged in just an hour or two is worth this cost. (Trickle-only charging takes at least 10 hours.)
Because these two configurations are so different, the information below will refer to fast-charge and trickle-charge circuits. Although fast-charging circuits also do trickle-charging, assume that when I talk of trickle-charging that I mean a trickle-only setup.
See the schematic, which shows both circuit configurations.
U1 is an adjustable linear regulator. It's set up as a constant current source to force charge back into the batteries.
The standard part here is an LM317T. You can buy these everywhere, they're well understood, and they perform just fine for the purpose. The only problem with 317s is that you must use a heat sink with them if you're going to set up a fast charger.
There are several low-dropout (LDO) regulators which are more or less compatible with the LM317, such as the LM1086. Because LDOs will work with a smaller voltage drop across them, they can be set up so that they dissipate less power. This means the heat sink area on the circuit board is sufficient for most purposes. If you add an external heat sink anyway, that will lower the regulator's temperature further, which may help it to last longer. The downside is that LDOs are more expensive, and you need to add C1 to stabilize the regulator.
LDOs are only valuable when you can set the power supply voltage precisely. If there's more than a couple of volts dropped across the regulator, you might as well use a standard regulator.
Optional? No. It's used in both fast- and trickle-charging circuits.
If the supply voltage to the board is over 20V and you're using a charge controller (U3), you need to put a TLE2426CLP here. See the Theory of Operation page for the explanation of why U2 is necessary.
Optional? Yes. If you don't need it, put a jumper from the IN to OUT pins.
This is the MC3334x charge controller chip. Its job is to monitor the charging process and to control the fast-charging regulator to ensure that the battery pack is charged safely.
There are two main variants of it, the '340 and the '342. The '342 is a bit cheaper, but I recommend you get the '340 instead. The difference in the '342 looks like it only makes sense for cold-weather charging or other odd situations. For room temperature charging, the '340 is a better chip.
Optional? Required if you want a fast-charging circuit. Otherwise, leave it out.
This serves the same sort of purpose as U2.
Optional? Yes. If you don't need it, put a jumper from the IN to OUT pins.
This cap is there to stabilize the regulator if you're using an LDO type like the LM1086. See your regulator's datasheet for the specifics, but 10 µF works for most LDOs. It's not needed for standard linear regulators like the LM317.
You definitely don't want an electrolytic cap here. The ESR of electrolytics is too high for the purpose, and you don't want to put an electrolytic near a heat source because that will shorten their life dramatically. Tantalum caps are expensive, true, but if you want to save the cost of this cap, use a regular regulator instead of an LDO so you don't need C1 at all.
Optional? Required only if you use an LDO in U1.
Largest Part Size: 0.100" pin pitch dipped cap
These are bypass capacitors for the charge controller chip. Ceramics are fine here, but you can use polyester film caps instead if you happen to have some on hand. The circuit won't perform materially better with film caps here, though.
Optional? If you use the charge controller chip, no.
Largest Part Size: 0.300" × 0.100"
These are standard silicon diodes. The 1N400x series works just fine for the purpose.
Optional? Fast-charging requires all diodes. See schematic for trickle-charge configuration.
This LED indicates the charge state.
For fast-charge circuits, it's off when there is no power to the circuit, blinking when fast-charging, and solidly on when trickle-charging.
For trickle-only circuits, it turns on when there's external power applied.
Optional? Yes. Leave it out if you don't want a charge indicator.
This is the trickle charge current limiting resistor in fast-charge circuits:
R1 = (Vs - 1.4) / It
where:
Vs = Supply voltage to the battery board
It = Desired trickle charge current
The "1.4" represents the two diode drops from D2 and D3.
The trickle charge current shouldn't be more than 0.1C for NiMH cells. (E.g. 70mA for 700mAh cells.) Lower is okay, especially if you mainly just need trickle charging to maintain the charge level in the cells.
Optional? Needed in fast-charge configuration only.
This is the regulator's current setting resistor:
R2 = 1.25 / If
where:
If = Desired fast charge current
"1.25" is the voltage drop from the regulator's OUT to ADJ pins. Although the voltage drop across R2 isn't precisely 1.25V when you use R3 (the two make a voltage divider) it's close enough to the real value when R3 is much larger than the value of R2.
The fast charge current shouldn't be more than 1.0C for NiMH cells. (E.g. 700mA for 700mAh cells.) You may well want to go lower, either to keep heat down or to make the cells last longer or because your power supply has a lower current limit. I find that a charge current of about 500mA is a good balance between speed and power for 750mAh cells.
Keep in mind, the lower the resistor value, the higher the current and thus the higher the power dissipation in the resistor. For resistors down in the 2 Ω range, you should use at least a 2W resistor. Higher-wattage resistors usually have more surface area, so they don't get as hot at any one point. I like to use 5W cement wirewound resistors here. They're cheap, available everywhere, accurate enough, and come in the right range of values. Good alternatives are silicone wirewound and metal oxide.
For trickle-charge circuits, the value of this resistor will be up in the tens of Ohms, so the power dissipation will be much lower than in fast-charge circuits. 1/4W or 1/2W resistors will be sufficient, depending on the resistor value and the charge current.
Optional? No.
This resistor limits current from the ADJ pin to just high enough to keep the regulator operating. I always use 120 Ω here. If you don't have that value, 100 Ω will also work. If you want to use a different value, study the datasheet of your regulator.
Optional? Needed in fast-charge configuration only.
These divide the battery voltage down to between 1 and 2V. The charge controller uses this divided-down voltage to sense the charge state of the batteries. Whenever the voltage at the divider is outside this range, the charge controller will leave fast-charging mode.
The simplest way to set this divider up is to use the Configuration Calculator.
If you want to calculate your own resistor values, figure that your cells are at about 1.55V each while fast-charging, and can be about 0.9V each when fully used up. Pick a division factor such that the full voltage of the battery pack is divided down to just under 2V, and the minimum voltage when divided by the same factor is still above 1V.
Beware that the battery pack will discharge through this divider. NiMH cells self-discharge in about 30 days, so if the current through the divider is about 10× lower than would be required to discharge ideal cells in 30 days, the self-discharge rate will dominate the pack's discharge time. Therefore, I suggest a maximum current of about 0.1mA for AAA NiMH cells, and 0.3mA for AA NiMH cells. If you use the configuration calculator, it gives divider values that obey these rules.
Optional? Needed in fast-charge configuration only.
These parts are for setting the backup fast-charge termination mode. Normally the charge controller can terminate fast-charging correctly by sensing the pack's voltage, but there needs to be a backup method in case something goes wrong with the cells. For instance, when a cell starts getting weak, the stress of fast-charging can make it overheat.
If you don't want to think too much about this, I recommend that you use the Configuration Calculator and select time-based backup charge termination. The calculator will give you a reasonable R6-R8 configuration.
Temperature based fast charge termination is safer because a fault in the battery pack is likely to cause overheating. If that happens, you want the charger to turn off. With time-based termination, the charger will eventually turn off, but only after the pack has been too hot for too long. It's worth the effort to figure out how to set up temperature-based fast-charge termination. Read through the MC3334x datasheet to find out how to set these part values.
Optional? Needed in fast-charge configuration only.
This is the power indicator LED's current limiting resistor. It can be a 5% carbon type; the exact value isn't at all critical.
RLED = (V+ - Vf) / If
where:
V+ is the power supply voltage, rail to rail
Vf is the LED's forward voltage drop
If is the desired current through the LED
1 mA gives enough brightness for a power indicator with most LEDs, but some may require a bit more. You don't want it to be too bright, or it's annoying. Typical values for RLED are 1 kΩ to 10 kΩ, depending on the power supply voltage and the LED being used.
Optional? Leave this out if you aren't using a charge indicator LED.
This was intended to set up a constant current through the LED regardless of power supply voltage. Sadly, it interferes with the charge controller, so you can't make use of it. Just use RLED alone.
Arguably the single most important part you will buy for this project. Although you can use NiCd cells with this charge circuit, NiMH cells are a much better idea: they are better for the environment, they have much higher energy density, and their disadvantages relative to NiCd don't matter in this application. You can readily find NiMH AAAs in the 700 mAh range and AAs in the 1800 mAh range. You can find higher capacity types, but you'll save a lot of money by choosing the ones a step or two back from the highest capacity types available.
All NiMH cells are not created equal. From the links page, visit The Great Battery Shootout for info on the best NiMH cells currently on the market. The author recommends that you get your cells from Thomas Distributing, and I second that recommendation. Yes, their web designer is apparently on acid, but their service and prices are great.
For those of you who want to place orders to as few places as possible, I have secured a deal with Mouser for their Sanyo 650mAh AAA cells for $1.89 each. (Part# 639-HR-4U) Sanyo makes Kodak's NiMH cells, and Thomas sells 750mAh Kodak AAAs for this same price. From my reading of these two cells' respective datasheets, I think these are actually the same cell, and are only rated differently because Kodak is using aggressive consumer marketing, while Sanyo is selling to the OEM market so they are giving conservative specs for engineers. Even if the Sanyo cell is somewhat inferior, it's still a good product. Sanyo cells score right at the top of the Shootout ratings I reference above. To get this price, put "QT# 24120630WY" in the Comments field on the web order form, and make the order attention to Sheila Williams.
These are the battery holders.
A full AAA configuration is six 3×AAA battery holders giving 18 cells. You can also use 2× holders on the board. Using various combinations of 3× and 2× holders, you can come up with many different cell count arrangements.
There are no silkscreen outlines showing where the AA holders go, to keep the silkscreen layer from being confusing. You just have to find the holes that don't go with the AAA holder positions. A full configuration is two 4× holders plus one 2× holder for 10 cells. The upside of using AA batteries is their capacity is almost three times as high as for AAA's. The downsides are that you can't get as many cells on the board so the output voltage is lower, and AA batteries are larger in diameter so this isn't the best setup if the battery board is in the case with the amp board. AAAs are still best for high voltage and compact configurations.
Remember to put jumpers across any unused battery holder positions!
This is a safety fuse. I have no specific reason to believe it's necessary, but it's a power circuit, so I put it in as a matter of course. What the heck, it's cheap and the space is there.
There are no surge currents in this circuit, so your fuse's value should be between your highest charge current and the maximum output current the regulator is capable of. This ensures that the fuse will blow even when there's not a short circuit, but just an overcurrent condition somewhere.
The board takes 5×20 metric fuses.
The resistor pads on the battery board are 400 mils apart, sufficient for up to 1/2W resistors in most resistor lines. However, 1/4W is sufficient for all resistors except R2.
If the power supply voltage is too low, you won't be able to fully charge your battery pack. If the voltage is too high, the regulator has to drop the excess voltage, so it runs hotter than it needs to. Ideally, you want to set things up so that you have just barely enough voltage to charge the cells fully while not allowing the regulator to drop out of regulation.
The Configuration Calculator asks you for your battery board configuration and then gives you a minimal power supply voltage level. If you don't care how the calculator works, just pick your power supply voltage to be as close to the value it recommends without going under. If you're curious, read on.
The key to picking the power supply voltage is to look at all the voltage drops in the charge path. Figure that each cell will require about 1.55V across it at peak during fast charging. Then add in all the diode drops along the charge path and the 1.25V across R2. Whatever is left between this and the power supply voltage is dropped across the regulator. Depending on the output current, an LDO will require about 1V across it to maintain regulation. A standard regulator requires 2 to 3V across it at minimum. See your regulator's datasheet for details.
If you have the luxury of an adjustable bench supply, you can use it to pick your power supply voltage experimentally. You simply keep tweaking the voltage up as the battery pack charges until the current draw stays steady. When the supply voltage is too low, the current from the regulator drops below where it should be, so increasing the voltage increases the current. When the voltage is high enough for the regulator to maintain regulation, the current stops at the level set by R2. You will have to keep tweaking it upward as the batteries charge. Whatever voltage you end up with when the circuit goes into trickle charge mode is the minimum voltage you should use to charge that battery board. You might make the power supply put out about 1V more to provide a little headroom.
This is a pretty conservative setup. The voltage you get is sufficient for many setups, the charge time is pretty good, and it uses very generic components and resistor values.
Supply: 24V
U1: LM317
R1: 1 kΩ
R2: 4.7 Ω
R4: 100 kΩ
R5: 10 kΩ
This is the same as the 12-cell configuration but with 16 cells, which is the highest number you can charge with a 30V supply and a standard regulator. It's useful if you need more voltage than the 12-cell configuration provides and you don't want to deal with the problems that result from maxing out the battery board's 18-cell capacity.
Supply: 30V
U1: LM317
R1: 1 kΩ
R2: 4.7 Ω
R4: 100 kΩ
R5: 7.5 kΩ
This is the fastest-charging, highest-voltage setup without getting ludicrous. Use this when you must have high voltage and fast charging. You'll have to take special care to keep heat under control.
Supply: 32V
U1: LM1086
R1: 1 kΩ
R2: 2 Ω
R4: 120 kΩ
R5: 8.2 kΩ
|
<< Checklist |
Configuration Calculator >> |
| Updated Fri Dec 16 2022 12:23 MST | Go back to The PPA Project | Go to my home page |