This is only a preview of the December 2020 issue of Practical Electronics. You can view 0 of the 72 pages in the full issue. Articles in this series:
|
Clever Controller for a
Dumb Battery Charger
BY JOHN CLARKE
Most cheap battery chargers – the type you might buy at a hardware
store or auto retailer – are pretty dumb. As many people have discovered
(because these chargers are so dumb) they can actually destroy the
battery under charge! If you have one of these chargers, you can upgrade
it to one with a clever controller, suitable for flooded lead-acid, sealed
lead-acid (SLA) or even LiFePO4 rechargeable batteries.
M
any manufacturers’ idea of a battery charger
is a transformer, a diode or two and a pair of clip
leads... and not much else. You may even have one
of these sitting on a shelf in the garage. They’re everywhere!
Sure, it will charge a flat battery but the chances are if you
don’t unclip it, it will keep on charging and charging and
charging... until the battery electrolyte is boiled dry, the
plates are buckled or, worst case, you have a fire on your
hands that may be very difficult to control!
Our new Charge Controller is used in conjunction with one
of these basic, low-cost lead-acid battery chargers. It transforms this ‘dumb’ charger into a more advanced device that
can still charge at the same maximum rate, but also offers
proper charge termination, float charging and temperature
compensation. Since it’s fully adjustable, it caters for the
lithium-iron-phosphate (LiFePO4) batteries that are starting
to become available as a replacement for lead-acid types.
Compared to lead-acid, LiFePO4 offers faster charging
and discharging, more charge cycles, smaller volume and
lighter weight, albeit at a higher cost.
A
12V
A
2 3 0V
AC
0V
DIODE 1
330Ω
12V
A
A
N
12V
TRANSFORMER
+
K
K
DIODE 2
THERMAL
CUTOUT
K
K
Basic charger flaws
The configuration of a typical low cost lead-acid battery
charger is shown in Fig.1. It comprises a mains transformer
with a centre-tapped secondary output. The output is rectified using two power diodes to provide raw DC for charging the battery. A thermal cutout opens if the transformer
is delivering too much current.
Charge indication – if it is present at all – may be as
simple as a zener diode, LED and resistor. The LED lights
when the battery voltage exceeds the breakdown voltage of the zener diode (12V) and the forward voltage of
the green LED (at around 1.8V). Thus the LED begins to
glow at 13.8V and increases in brightness as the voltage
17V PEAK
12V RMS
TO
BATTERY
12V
ZENER
A
16
GREEN
LED
Adding a fully automatic Charge Controller to a basic
charger will also prolong the life of your batteries, and
you can leave a battery on a float charge as long as you
want, ready for use when required. LiFePO4 batteries usually are not float charged, so you can disable that step for
these batteries.
–
0V
Fig.1: the basic arrangement of a typical low-cost lead-acid battery
charger. It consists of a centre-tapped mains transformer and a
full-wave rectifier (D1 and D2). There’s usually a thermal cutout
and perhaps an LED indicator to show when the battery is charged.
The output voltage of this simple arrangement is shown above.
Practical Electronics | December | 2020
VOLTS
UNLOADED
CHARGER
OUTPUT
Features
BATTERY
VOLTAGE
0
10ms
20ms
30ms
TIME
• Charges 6V, 12V or 24V flooded lead-acid, SLA or
LiFePO4 batteries at up to 10A (with a suitable charger)
• Charge rate: adjustable from 1-100% of charger
capability in 1% steps
• One, two or three charging phases: bulk, absorption
and float
• Adjustable or pre-set charge termination and float voltages
• Adjustable temperature compensation for lead-acid
batteries with an internal or external thermistor
CURRENT
• Automatic slow charge mode for batteries that are
heavily discharged
TIME
A
CHARGING VOLTAGE AND CURRENT
BATTERY
VOLTAGE
• Battery discharge protection
• Cold battery charge protection (won’t charge below 1°C)
UNLOADED
CHARGER OUTPUT
• Thermistor fault protection (won’t charge lead-acid
batteries if the thermistor is open or short circuit)
• Six status indicator LEDs with error indication
REQUIRED
BATTERY VOLTAGE
CHARGING TIME
B
CHARGING CHARACTERISTIC
Fig.2. in more detail, the charging current from the circuit
shown in Fig.1 consists of a series of high-current pulses at
100Hz. As shown in part (b), the relatively high peak voltage
can result in the battery being over-charged if the charger is
left on long enough.
rises. Some chargers may also have an ammeter to show
the charging current.
The charging current to the battery is a series of highcurrent pulses at 100Hz, as shown in Fig.2(a). The nominal
17V peak output from the charger will eventually charge
a battery to over 16V if left connected long enough, which
will damage the battery. As shown in Fig.2(b), the maximum
battery voltage for a full charge (called the cut-off voltage)
is exceeded when left on charge for too long.
The solution
By adding in the Charge Controller to that simple charger,
we can do much better. Fig.3 shows how the Charge Controller is connected in between the charger and the battery.
The Charge Controller is housed in a compact diecast aluminium case. In effect, the Charge Controller is a switching
device that can connect and disconnect the charger to the
battery. This allows it to take control over charging and to
cease charging when the correct voltage is reached.
The various charging phases for lead-acid batteries are
shown in Fig.4. The Charge Controller can switch the current on or off and apply it in a series of bursts, ranging from
20ms every two seconds through to a continuous current.
During the first phase, called ‘bulk charge’, current is
typically applied continuously to charge as fast as possible.
After the bulk charge phase, the Charge Controller switches
to the ‘absorption phase’. This maintains the cut-off voltage
for an hour by adjusting the burst width while it brings the
battery up to an almost full charge. After that, the Charge Controller switches to ‘float charge’. This uses a lower cut-off voltage and a low charge rate, to keep the battery fully charged.
Practical Electronics | December | 2020
• Low-cost, easy to build and easy to use
• Microprocessor controlled
The switch from absorption to float occurs when the
charging current drops to 3% of the original bulk charge
rate or after an hour, whichever comes first. The absorption
phase is optional; you can choose to skip this phase and
go straight from bulk charging to float charging.
When absorption is enabled, this phase will be bypassed
if the bulk charge takes less than an hour. This prevents
excessive absorption phase charging with an already fully
charged battery.
While the bulk phase is usually done at the full rate, for
lower capacity batteries where this charging current would
be too high, the burst width can be reduced to limit the
average current.
For example, if you have a 4A battery charger, the current can be reduced from 4A anywhere down to 40mA in
1% steps, using the charge rate control.
Lithium-iron-phosphate battery charging
Typically, LiFePO4 batteries are charged to 3.47V per
cell, although 3.6V per cell is also used. A nominally 12V
LiFePO4 battery therefore has four cells, and the cut-off
voltage is either 13.88V or 14.4V, depending on which percell figure you use.
The Charge Controller can cease charging once the cut-off
voltage is reached, or you can opt for an absorption phase.
During this phase, the cut-off voltage is maintained for an
hour, or until charging pulses drop to 3% of the original
bulk charge setting.
+
+
+
–
–
–
LEAD-ACID
BATTERY CHARGER
+
–
CHARGE
CONTROLLER
BATTERY
Fig.3. the Charge Controller is connected
between the charger and battery. It takes
control over charging and ceases charging the battery at
the correct voltage; ie, when it is fully charged but before it
becomes over-charged and starts out-gassing (or worse).
17
Specifications
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
Charging pulse width: 20ms-1980ms in 20ms steps, or continuous
Charging cut-off voltage: 0-30.5V in 29.8mV steps. Independent LiFePO4, SLA and lead-acid battery settings (presets are
also available, see Table 1)
Temperature compensation: 0-50mV/°C in 256 steps (separate SLA and lead-acid battery adjustments)
Minimum battery charging temperature: 1°C
Maximum compensation temperature: 60°C
Under-voltage burst charge: 5.25V for a 6V battery, 10.5V for a 12V battery, 21V for a 24V battery
Under-voltage burst rate: 200ms burst every 2s at maximum charge rate. The burst width is reduced with a lower charge
rate (10% of the normal rate).
Battery discharge protection: if charger power is lost, it switches off after two hours with battery voltage below 6.25V (for
a 6V battery), 12.5V (for a 12V battery) or 25V (for a 24V battery)
Power on: LED1 lights
Thermistor error: LED2 lights
Temperature too low: LED2 flashes at 1Hz
Bulk charging: LED3 lights
Absorption charging: LED4 lights; optionally, LED3 flashes to indicate charge rate
Float charging: LED5 lights; optionally, LED3 flashes to indicate charge rate
Battery detected: LED6 lights
Battery voltage low, charging slowly: LED3 flashes; if charging a lead-acid battery, LED4 and LED5 also flash
Lead-acid cut-off and float voltages
The actual cut-off and float voltages for lead-acid batteries
are dependent on the particular battery, its construction and
the operating temperature. Typical cut-off and float voltages at 20°C are 14.4V and 13.8V, respectively. For sealed
lead-acid (SLA) batteries, the voltages are lower at 14.1V
and 13.5V respectively.
These values, plus 13.88V for the LiFePO4 battery, are
pre-set within the Charge Controller and selected using the
Lead-Acid/SLA/Lithium jumper shunts, but only when the
‘default’ shunt is inserted (not ‘adjustable’). See Table 1.
Other settings are possible, and can be set manually from
0-30.5V in 29.8mV steps – see Table 2.
These voltage settings can also be compensated for temperature changes; as the temperature rises, the charge voltages for a lead-acid battery are normally reduced. A typical
temperature compensation value is –20mV/°C for flooded
CUTOFF
VOLTAGE
CUTOFF
POINT
BATTERY
VOLTAGE
cells and –25mV/°C for SLA batteries. LiFePO4 batteries
do not require temperature compensation.
Temperature compensation values can be set from between 0 to –50mV/°C in 256 steps. Temperature compensation is applied for temperatures between 0°C and 60°C.
No charging is allowed at temperatures at or below 0°C,
to protect the battery.
A negative temperature coefficient (NTC) thermistor is
used for temperature measurement, and the Charge Controller will use the internal thermistor if an external one
is not connected via its jack socket. The external thermistor provides for a more accurate measurement when it is
placed against the battery.
Four trimpots are used to make the settings. One sets the
charge rate, as a percentage of the full charge current available from the charger. The remaining three are for setting
the cut-off voltage, float voltage and temperature compensation adjustments.
When charging the battery, the microcontroller adjusts
the pulse duty cycle to reach the desired battery terminal
voltage using negative feedback.
FLOAT
VOLTAGE
BULK
CHARGE
ABSORPTION
FLOAT
CHARGE
CURRENT
Setting
SLA
Flooded
lead-acid
14.4V
13.8V
LiFePO4
Cut-off voltage
Float voltage
Temperature
compensation
14.1V
13.5V
-25mV/°C -20mV/°C
None
13.88V
None
Table 1 – default settings
TIME
Fig.4: the three typical charging phases for a lead-acid
battery. It starts with the bulk charge phase, then switches
to the absorption phase (optional, selected using JP2) for
an hour or so, and then finally switches to float charging to
finish charging and keep the battery charged. For LiFePO4
batteries, there is no float phase. The charger switches off
when the battery is fully charged and switches back on again
later if it becomes discharged.
18
Setting
Set by
VR2
VR3
SLA and
Flooded lead-acid
0-30.5V*
0-30.5V*
Cut-off voltage
Float voltage
Temperature
compensation
0-30.5V*
None
VR4
0 to -50mV/°C
None
Table 2 – adjustable settings
LiFePO4
*in 29.8mV steps
Practical Electronics | December | 2020
TO
CHARGER
F1
–
10A
100
1W
Q1 IRF1405N
K
D1
1N4004
A
A
G
TP5V
REG1 LM317T
K
K
ZD1
18V
220 F
ADJ
8
120
1
+5V
1k
7
A
50V
100 F
330
K
ZD2
18V
16V
A
3
POWER
D2
1N4004
RLY1
(5V)
2k
15
Q3
BC337
VR1
10k
3.3k
18
Vdd
RA3 /AN3
AN1/RA1
10nF
E
RA4
TP2
VR2
10k
OUT:
DEFAULT
12V
IN:
ADJUST.
17
AN0/RA0
10nF
VR3
10k
+5V
13
12
4
16
AN5/RB6
RA2/AN2
RB4
A
7
100nF
A
RLY1b
RB5
1k
1k
1k
1k
1k
LED2
K
A
A
LED3
K
LED4
K
A
100nF
BATTERY
LED6
K
LED2: THERMISTOR
LED3: CHARGE
LED4: ABSORPTION
LED5: FLOAT
K
C
Q2
BC337
B
E
10k
1
11
STORE
RA7/OSC1
EXTERNAL
THERMISTOR
T
S2
R
THERMISTOR
100nF
TH1
5
S
CON1
BC 33 7
LEDS
K
A
10k
A
10k
LED5
A
Vss
ZD1, ZD2
K
8
RA5/MCLR
10nF
K
6
RB1
AN6/RB7
10nF
1k
1: SLA
1
2: FLOODED
LEAD-ACID 2
OPEN:
LITHIUM
3
RB2
10
JP3
1N4004
100k
51k
RB0
TP4
VR4
10k
2
2:
ABSORPTION
4
2
10nF
+5V
1:
STANDARD 1
2
TTC
9
IC1
PIC1 6F8 8
PIC16F88
TP3
JP2
RB3
RA6/OSC2
B
JP1
INPUT
100nF
14
TP1
C
IC2
+5V
A
56
SOURCE
6
Mcap2
Mcap1 Si87 51
100nF
K
+5V
K
5
GATE
+5V
GND
LED1
VR5
100
A
–
10pF
+5V
OUT
IN
S1
POWER
+
D3 1N4004
RLY1a
+
TO
BATTERY
S
D
B
E
C
8
Universal
Battery
Charge
Controller
SC UNIVERSAL
BATTERY
CHARGE
CONTROLLER
20 1 9
IRF1405N
IC2
4
1
G
LM317T
D
D
S
OUT
ADJ
OUT
IN
Fig.5: the Charge Controller circuit is based around a PIC16F88 microcontroller (IC1). This monitors the battery
voltage at its AN3 input and switches MOSFET Q1 on and off via isolated driver IC2, to control the charging.
The duty cycle is reduced by 15% every two seconds
if the battery voltage is above the required value by more
than 0.25V, or reduced by 1% every two seconds if the battery voltage is above the required value by less than 0.25V.
Conversely, the charge duty cycle is increased at
a fast rate (3% per two seconds) if the battery voltage is more than 0.25V below the required value and
increased at a slow rate (1% per two seconds) if the
battery voltage is low by less than 0.25V.
LED indicators
The Charge Controller has six LED indicators. LED1 (green)
shows power is applied, while LED2 (orange) flashes when
the thermistor temperature is below 0°C but otherwise does
not light unless the thermistor connection is broken or
shorted. LED3 (red) indicates the bulk charge phase, while
LED4 (orange) and LED5 (green) indicate the absorption and
float phases. LED6 (green) indicates that a battery is connected, but is not an indication that charging is occurring.
Practical Electronics | December | 2020
There is an option for LED3 to indicate when current
is being fed to the battery during the absorption and float
phases. This is useful, as it flashes whenever current is being fed to the battery.
So it indicates the duty cycle of power bursts. Brief bursts
indicate that the battery is close to the required voltage,
while longer bursts indicate that the battery requires further charging.
If this is not required, it can be disabled so that LED3
only lights during the bulk phase.
The absorption LED (LED4) will never light if you set
up the charger to skip this phase. Similarly, the float LED
(LED5) does not light when charging LiFePO4 batteries,
since that phase is not used for lithium batteries.
Isolated MOSFET drive
The circuit of the Charge Controller is shown in Fig.5. It
uses a PIC16F88-I/P microcontroller (IC1) to monitor the
battery voltage and adjust the switching of an N-channel
19
Transmitter
Receiver
MODULATOR
A
SemiconductorBased Isolation
Barrier
Input Signal
DEMODULATOR
Modulation Signal
B
RF OSCILLATOR
Output Signal
Fig. 6(a): Simplified Channel Diagram
Fig 6(b): Modulation Scheme
Fig.6: an excerpt from the Si8751 data sheet, showing its internal arrangement. It comprises an RF transmitter and RF
receiver to transmit gate drive power and control from the input side to the output. The receiver is isolated from the
transmitter by a semiconductor isolation barrier, rated at 2.5kV. When the RF transmitter is producing an RF signal, a
gate-drive voltage appears at the output. When there is no RF transmission, there is no gate-drive voltage.
MOSFET (Q1) to control the charging rate. Q1’s channel is
connected between the incoming positive supply (drain)
and the battery positive terminal (source).
To switch Q1 on, its gate needs to be brought several volts
higher than its source. Since the source is at the battery voltage, we need a way to generate a voltage above this. This
needs to be controlled by a 0-5V control signal from microcontroller IC1. To accomplish this, we use an Si8751 isolated FET driver (IC2). It provides up to 2.5kV of isolation
between its input and output, but here, 45V is sufficient.
IC2 runs from the same 5V supply as microcontroller IC1,
and Q1’s gate is driven from pin 8. The MOSFET source is
connected to pin 5. The gate drive output at pin 8 typically
charges the gate to 10.8V with respect to the source when
the input at pin 3 is high (5V). The gate output is pulled
down to the source voltage with a 0V input.
The 10pF capacitor between drain and MCAP1 (pin 7)
enables a feature of the chip to prevent a fast voltage rise
at the MOSFET drain from coupling into its gate and spuriously switching it on.
Internally, IC2 comprises an RF transmitter and RF receiver to send gate drive power from the input side to the
isolated output. Isolation is provided by a semiconductor
oxide barrier. When the transmitter is producing an RF
signal, this is detected in the receiver to produce the gate
drive voltage. When there is no RF transmission, there is
no gate drive. See Fig.6 for details of its internal operation.
The gate drive current is set by the resistor at pin 2. In
combination with the MOSFET’s gate capacitance, this
determines the MOSFET switch-on time. With the 100kΩ
resistor we’ve used, the switch-on time is around 5ms
to a gate voltage of 5V. It continues to rise to about 10V,
but the MOSFET is already mostly in conduction by 5V.
The 100kΩ resistance we have chosen reduces the supply
current for IC2 from 13.8mA down to 1.8mA, compared to
the fastest option of connecting pin 2 directly to ground,
which would give a 1ms switch-on time. The 100nF capacitor across the 100kΩ resistor speeds up switch-on without increasing current consumption. The switch-off time
is typically 15µs, regardless of the resistor value at pin 2.
Fast switching isn’t required in this application, as we’re
only switching the MOSFET on/off once every two seconds.
Low current consumption is important, so that REG1’s
dissipation is below 1W when charging a 24V battery. Otherwise, the regulator will run very hot and need heatsinking beyond that provided by the PCB.
Switching losses increase when the switching is slow
because the MOSFET’s dissipation is at a maximum when
it is in partial conduction. The instantaneous losses can be
high (hundreds of watts at many amps), but as they are infrequent, the average is low. Switching losses are: (switchon loss + switch-off loss) × switching frequency. So losses
are directly proportional to frequency.
Fig.7 is an oscilloscope screen grab showing the gate
drive waveform for MOSFET Q1. The period for the gate
to rise from 0V, with the MOSFET off, to fully conducting (4.5V) is 5ms. The switch-off time is relatively fast at
around 35µs for the full gate-voltage excursion.
Scope1: scope grab of the Charge Controller with a 2A
charger and a lead-acid car battery. The yellow trace
shows the charger output, the green trace the battery
voltage and the blue trace the charge current. Note how
the battery voltage varies with the charging current. The
difference in voltage between the charger and the battery is
due to the current shunt and cable losses.
Scope2: the same charging scenario as Scope1 but at a
much longer timebase, showing the many pulses that make
up two seconds of charging.
20
Practical Electronics | December | 2020
The overall energy loss in the MOSFET (and therefore
heating) is the switching losses plus the static losses. We’ve
already explained that the switching losses are reasonably
low. The static losses are simply the average current times
the MOSFET’s on-resistance. Its on-resistance is low enough
that even at 10A, the static losses are within reason.
Circuit description
Power for the circuit is usually obtained from the ‘dumb’
charger via reverse-polarity protection diode D1, although
it can also flow from the battery via the body diode within Q1. However, the latter has no useful function and can
eventually discharge the battery. We have a solution for
that, which is described below.
The incoming supply also passes through a 100Ω
dropper resistor and either power switch (pushbutton) S1 or the contacts of RLY1, and is then filtered by a
220µF electrolytic capacitor and fed to an LM317T adjustable regulator (REG1), set to deliver a precise 5.0V.
For REG1, the voltage between the OUT and ADJ terminals
is a fixed reference value of typically 1.25V, but it could
be between 1.2 and 1.3V. Assuming it is 1.2V, the 120Ω resistor between these pins has 10mA (1.2V ÷ 120Ω) flowing through it, which also passes through the 330Ω resistor and trimpot VR5.
We need 3.8V at the ADJ terminal for a 5V output (3.8V
+ 1.2V), so the total resistance of VR5 and the 330Ω resistor needs to be 380Ω for the 10mA current to produce this
voltage. VR5 is therefore adjusted to give 50Ω. This adjustment is provided to allow for variations in REG1’s reference
voltage and the resistor values.
The 5V supply feeds both IC1 and IC2. The accuracy of
the 5V setting adjustment determines the precision of the
battery-charge voltage settings. That is because IC1 uses the
5V supply as a voltage reference to compare the measured
battery voltage against.
Preventing battery discharge
To switch the Charge Controller on, momentary pushbutton S1 is pressed, allowing current to flow into REG1. IC1
then switches on RLY1, shorting out S1 so that the circuit
remains powered after it is released. RLY1 is controlled
by digital output RA6 of IC1 (pin 15), which goes high to
Scope3: we have now reduced the charging duty cycle
to around 75% and the average current delivered to the
battery has dropped (the reading is unrealistically low due
to the timing of the pulses). Note how the battery voltage
rises during the bursts, then falls a little between them,
averaging lower than before. The charger output voltage
rises substantially when it is not delivering current.
Practical Electronics | December | 2020
Using the Charge Controller
with 6V batteries
The circuit as presented is suitable for use with 12V or 24V
batteries and chargers, but it can easily be modified for 6V
batteries and chargers with a few changes. Note that if you make
these changes, you can only use the unit with a 6V charger.
The changes required are: replace D1 with a 1N5819 Schottky
diode, change the 100Ω 1W resistor to 10Ω 1W and change
REG1 to the low-dropout version, LD1117V. ZD1 should be
changed to a 15V 1W type and ZD2 replaced with a wire link.
The default position for JP1 cannot be used with 6V batteries;
set the adjustable cut-off voltage, float voltage and temperature
compensation values to suit your particular 6V battery.
drive the base of NPN transistor Q3, energising the relay
coil via a 56Ω resistor.
This resistor reduces the current through the relay coil,
as the relay will operate down to 3.75V and so we save a
little power this way. Without the resistor, the relay coil
current is 28mA, and with it, it is 21mA.
The other set of contacts in RLY1 make the connection
between the battery and the 51kΩ and 10kΩ battery-voltage-measuring resistors.
If the charger is switched off or a blackout occurs with
the battery still connected, the battery powers the Charge
Controller and it could become over-discharged and damaged if this continues long enough. With the charger power
off, the circuit draws around 50mA from the battery.
To prevent this, IC1 monitors battery voltage and when
the battery voltage falls below 12.5V for a 12V battery or 25V
for a 24V battery for at least two hours, the RLY1 switches
off. This totally removes the load from the battery, as current
can no longer flow from it into REG1 or the voltage divider.
Battery voltage measurements
When the Charge Controller is powered up, the 51kΩ and
10kΩ resistors allow IC1 to monitor the battery voltage at
its AN3 analogue input (pin 2). The resistors reduce the
battery voltage to be within its 0-5V measurement range.
So, for example, if you have a 24V battery at its maximum
standard charge voltage of 28.8V, the battery voltage is
Scope4: now the duty cycle has been reduced to 50%
and the battery voltage and average charge current have
dropped a little further.
21
Making a fully self-contained charger
While the emphasis in this project has been to make a dumb
battery charger clever, we can already hear the question: What
do you do if you don’t have a dumb battery charger? The answer to that is simple! There is absolutely nothing to stop you
making one, as per Fig.1 in this article, and add it to the project.
You won’t need the LED/zener indicator (the Charge Controller
tells you everything you need); the thermal cutout wouldn’t do
any harm, though.
In fact, you could place a 12V CT transformer and a pair of
diodes in a larger case and include this project to have a fully
self contained, clever battery charger. If you can’t lay your hands
on a 12V CT transformer, a single-ended 12V with a bridge rectifier will do the same job. Just remember that the transformer (in either case) must be a standard iron-core type (not an
electronic type) rated high enough – we’d suggest 4A or 50W
(did we hear someone say an old 12V downlight transformer?).
And the diodes or bridge need to be pretty beefy, too – a pair of
automotive diodes or a 30A bridge, for example.
Make sure the mains wiring side is exemplary – in fact, all
wiring must be workmanlike, properly anchored and so on. Any
metal case should be properly earthed (via the power cord).
So away you go . . .
divided down by a factor of 6.1, giving 4.72V at pin 2 of
IC1. The voltage is filtered with a 100nF capacitor to remove
noise from the measurement. IC1 converts the voltage to a 10bit digital value (0-1023), which gives a 29.8mV resolution
(5V × 6.1 ÷ 1023). Battery voltage measurements are made
when Q1 is switched off, so voltage fluctuations due to the
charging current in the leads to the battery don’t affect it.
Temperature measurement
An NTC thermistor is used to measure the battery temperature. One thermistor mounts on the PCB and connects to
pin 1 of micro IC1 via the switched tip contact of 3.5mm
jack socket CON1. When an external thermistor is connected via CON1, the internal thermistor is switched out and
the external thermistor connects to pin 1 of IC1 instead.
Note that the external thermistor is connected to ground
via the ring connection. The sleeve is left open. This allows
the metal enclosure of the Charge Controller to remain floating from the controller circuit.
Scope5: the duty cycle has now been reduced to 10%
but the battery is still charging (slowly), with an average
terminal voltage of 13.2V.
22
In either case, the thermistor is connected in series with
a 10kΩ resistor across the 5V supply. It therefore forms a
voltage divider and the resulting voltage, which is related to
the thermistor temperature, appears at the AN2 input (pin
1) of IC1 and is converted to an 8-bit digital value. IC1 then
uses a look-up table to convert the voltage to a temperature
value, as the relationship is non-linear.
IC1 can sense whether the thermistor is disconnected;
eg, if the wire to the external thermistor is broken. Pin 1
would then be at +5V. Similarly, if the resistor is shorted
to ground, IC1 can detect this as pin 1 will be at 0V. The
thermistor LED lights in either case, and charging ceases.
The thermistor LED flashes when the measured temperature is 0°C or below. Charging also ceases in this case.
Set-up adjustments
Analogue inputs AN5, AN6, AN0 and AN1 (pins 12, 13,
17 and 18) are used to monitor the settings for charge rate
percentage, cut-off voltage, float voltage and temperature
compensation, as set with trimpots VR1 to VR4.
Switch S2 is pressed to store the settings in IC1’s Flash. S2
is normally open, and an internal pull-up resistor within IC1
holds the RB5 input (pin 11) at 5V. When S2 is pressed, the
pin 11 input is pulled low (to 0V) and this signals the program
within IC1 to store the settings for VR2, VR3 and VR4 as the
adjustable values for either SLA, lead-acid or lithium batteries.
These values are only stored if the jumper JP1 is in the
‘adjustable’ position. Where the values are stored depends
on the position of the battery chemistry selection jumper
JP3. This is monitored by IC1’s RA7 digital input (pin 16).
Jumper link JP1 sets whether the Charge Controller uses
the standard (or default) values or the adjustable settings
referred to above.
JP2 selects the absorption option. If this jumper is not in
the ‘absorption’ position, when charging lead-acid batteries, the charger switches to float charging as soon as bulk
charging is complete. For LiFePO4 batteries, in this position, charging ceases as soon as the bulk charge is complete.
If absorption charging is enabled by JP2, the absorption
phase will run after the bulk charge, provided that the charging process has been going for more than one hour. At the
end of the absorption phase, the unit either switches to float
charging (for lead-acid) or ceases (for LiFePO4).
Since the battery chemistry selection jumper (JP3) can
have three possible states, including ‘open’, there is a 10nF
Fig.7: this scope grab shows the voltage at the gate of Q1
for a single, short pulse. The vertical scale is 2V/div and
the horizontal scale is 2.5ms/div. The MOSFET switches
on at around 4-5V, so we can determine from this that the
switch-on time is around 5ms, while the switch-off time is
much shorter, les than 0.1ms (100µs).
Practical Electronics | December | 2020
10nF
TH1
A
Power
1k
LED1
10nF
10nF
IC1 PIC16F88-I/P
100nF
100nF
100nF 1
CON1
LED2
A
Thermistor
10k
10nF
S2
SILICON CHIP
Q2
BC337
STORE
LED3
A
Charge
LED4
A
Absorption
51k
100k
TP3 TP4
10k
JP1
10k
<at>25 °C
100nF
100nF
1k
2
VR4
3.3k
10k
VR3
JP3 10nF
1
LED5
1k
18V
ZD1 ZD2
S1
1
18V
5V ADJUST
JP2
Si8751
COMP.
FLOAT
1k
TP
GND
TP5V
330
VR2
TP1 TP2
10k
120
D3
VR1
1k
+ 100 F
CUTOFF
IC2
1k
CHARGE %
JP1 IN: ADJUSTABLE
OUT: DEFAULT (12V only)
JP2 1: STANDARD
2: ABSORPTION
JP3 1: SLA, 2: Flooded LA
OPEN : LiFePO4
2k
F1 10A
1k
+
220 F
50V Q3 BC337
10k
4004
COIL
56
14107191
REV.B
C 2019
TO BATTERY +
10pF
REG1
LM317T
LED6
A
A
Float
Battery
Fig.8: fit the parts to the PCB as shown above and the photo below. Watch the
orientation of the diodes, ICs, LEDs, trimpots and relay. Note that the LEDs
should be fitted at right-angles, as shown here, to project through the side
of the case. Q1 is fitted last as it’s attached to the bottom of the case before
soldering its leads on the top side of the board. Jumper JP1 selects between
default or adjustable charging parameters, JP2 enables or disables the
absorption phase, and JP3 selects the battery chemistry.
Construction
The Charge Controller is built on a PCB coded 14107191, measuring 111 × 81mm and
available from the PE PCB Service. This is
housed in a 118 × 93 × 35mm diecast aluminium box.
It’s best to start by preparing the box. This
way, you can use the blank PCB as a template. First, locate the PCB in the bottom of
the box with the edge closest to the LEDs
against that edge of the box. Mark out the
four corner mounting hole positions, then
drill these holes to 3mm and deburr them.
Copy the panel artwork (Fig.8) and use
it as a template to drill out the holes in the
front of the enclosure for the switch, 3.5mm
socket and LEDs. Make sure the template is
lined up with your PCB mounting location
before drilling the holes.
The power switch hole is 4.5mm in diameter (5mm is OK) and the thermistor socket is 6.5mm (7mm is OK). The other panel
holes are 3mm.
You can now start assembling the PCB.
Fig.8 shows the overlay diagram, which
you can use as a guide during construction.
Start by fitting IC2. This is an 8-pin surface mount device that’s relatively easy to
solder using a fine-tipped soldering iron.
The pin 1 location is marked with a small dot on the package. Line the IC up on the PCB pads and tack-solder one of
the corner pins. Check that the IC is still aligned correctly
on all the pads. If not, re-heat the solder and adjust again.
When aligned correctly, solder all the pins including the
original tack-soldered pin. If any pins are bridged together,
use flux paste and solder wick to clear the bridge.
Next, insert the three M4 screws from the underside of
the PCB at each of the eyelet mounting positions and secure using M4 nuts on the top of the PCB. Using a soldering iron, preheat each screw and solder it to the board.
Make sure the solder adheres to each screw head. When
cool, the nuts can be removed.
Note that you may be able to build the unit without having to solder the screw heads if you use M4 copper crinkle
Practical Electronics | December | 2020
Q1
IRF1405N
(UNDER)
TO CHARGER +
VR5 100
D1
1 9 1 7TO
0 1 4BATTERY
1
–
TO CHARGER –
4004
100 1W
12/24V Universal Battery
Charge Controller
Indicator LED driving
Power indicator LED1 runs from the 5V supply via a 1kΩ current-limiting resistor. LED2,
LED3, LED4 and LED5 are driven from the
RA4, RB0, RB1 and RB2 digital outputs of
IC1 (pins 3 and 6-8), via 1kΩ resistors.
LED6 is the battery detection indicator
and is driven via transistor Q2 via a 1kΩ resistor from the 5V supply. The base of this
transistor connects to the switched side of
RLY1’s second set of contacts via a 10kΩ
resistor. This transistor switches on when
battery voltage is present. This prevents the
LED brightness from varying significantly
between different battery types.
D2
RLY1
4004
capacitor connected from pin 16 of IC1 to
ground. IC1 can therefore briefly pull this
pin high or low, then cease driving it and
sample the voltage at it.
If no jumper is inserted, the voltage will
be as expected, but if a jumper is in place,
it will prevent the capacitor from charging
or discharging.
washers under each screw head instead, but they are not
that easy to find.
Construction can now continue by installing the fixed
resistors. Take care to place each resistor in its correct position. It’s best to use a multimeter to check each set of resistors before fitting them because the colour bands can
be hard to read.
Next, fit the optional PC stakes for the test points labelled
TP GND, TP5V and TP1-TP4. They make it easier to attach
clip leads during set-up. Then mount the 2-way header for
JP1 and the 3-way headers for JP2 and JP3. Now install the
diodes and zener diodes, with the orientations and positions shown in Fig.8.
IC1’s socket can then be installed, and this must also be
oriented correctly. Follow with tactile pushbutton switch
23
Fifteen holes are required in the diecast box – eight on
the front panel (see below), two on the rear panel (for the
cable glands) and five in the base. Four of these are for PCB
mounting, with the 6.3mm pillars already shown fitted
here. The last hole, just visible in the top right corner, is for
mounting Q1 on its insulating washer and bush.
And here’s the PCB fitted inside the case with the six LEDs
just poking through. As yet, we haven’t fitted the front panel
artwork (Fig.9, below). And the wiring we used here was
just for testing – polarised 15A auto figure-8 should be used.
S2, then jack socket CON1. Push both all the way down
Then remove the board, drill the MOSFET mounting
onto the PCB before soldering their pins.
hole to 3mm and deburr. Also drill the cable gland holes
Fit the on-board NTC thermistor and capacitors next. and check that they fit securely.
Note that the electrolytic capacitors must be oriented with
The MOSFET is secured with a 10mm M3 machine screw
the polarity shown.
and nut. If you find it awkward to secure it, the screw can
In each case, the longer lead is positive, and the stripe be fed in from the top instead.
on the can indicates the negative lead. Install transistors
Q1’s tab must be isolated from the case by an insulating
Q2 and Q3, then trimpots VR1-VR5, taking care to fit the washer and mounting bush. For details, see Fig.10. Now
100Ω trimpot for VR5.
check that the tab of Q1 is insulated from the metal box by
Mount REG1 on the top side of the PCB, with its leads measuring the resistance between the two with a multimebent down to insert into its pads. Secure the regulator tab ter. The reading should be high, above 1MΩ.
to the PCB with a 10mm M3 screw and nut before solderThe box is isolated from the electrical connections so
ing and trimming the leads.
that accidental contact of the box to a battery terminal will
Follow by fitting RLY1, ensuring that its striped (pin 1) not cause a short circuit. The PCB can now be mounted
end faces to the right as shown.
inside the box using the remaining M3 screws in from the
Fuse F1 comprises the two fuse clips and the fuse. The base of the enclosure into the spacers.
fuse clips must be oriented with the end stops facing outFit the two cable glands and feed the figure-8 cable
wards so that the fuse can be clipped into place. Make sure through them, ready to attach the crimp eyelets. We used
they’re sitting flat on the PCB and then attach them using the striped side of the wire as the negative and the plain
a hot iron and plenty of solder.
wire as the positive, but some people prefer the opposite.
The LEDs are mounted at right angles to the PCB. Bend Just make sure you’re consistent.
the leads 11mm back from the front lens of each, taking
Attach the crimp eyelets to the wire using a suitable
care to have the anode (longer lead) to the right and then crimping tool and secure them to the PCB using the M4
bend the leads downward. Insert into the PCB and solder nuts and star washers. Make sure the eyelets are not shortthem so that the bottom of the lenses are 6mm above the ing to adjacent parts, especially the fuse holder.
top surface of the board.
Attach the large insulated clips to the end of the battery
Now mount pushbutton S1, ensuring it is pressed down leads; red for positive and black for negative. The Charge
Controller leads can be terminated in bare copper, for clampfirmly onto the board before soldering its pins.
Secure the tapped spacers to each corner of the PCB us- ing in your charger clips, or they can be permanently wired
ing 5mm M3 screws, then mount Q1. It’s fitted to the un- to the charger. Finally, push the button cap onto S1 and fit
the four stick-on rubber feet to the underside of the box.
derside of the PCB and bolted to the case for heatsinking.
Bend Q1’s leads up at right angles, as in Fig.10. It is placed
Preparing the external thermistor
so that the metal face will sit at the base of the enclosure.
Note that the tab of Q1 must be at least 1mm away from The NTC thermistor on the PCB gives acceptable results
the back edge of the case, to prevent the tab shorting to it. with the Charge Controller close to the battery, as the metTest that it is in the right position by temporarily mounting al box will not usually heat up too much above ambient
the PCB in position and mark out the
mounting hole for Q1. Also mark out
Hole sizes: 4.5
3
6.5
3
3
3
3
3mm
the two holes for the cable glands.
Fig.9: this front panel artwork can be
copied, laminated and glued to the front
panel. It could also be photocopied and
used as a template for drilling the front
panel holes, once you have established the
PCB position. You can also download the
panel artwork from the December 2020
page of the PE website.
24
SILICON CHIP 12/24V Battery Charge Controller
Float
Charge
+
+
Power
+
External
Thermistor
+
Thermistor
+
+
Absorption
+
+
Battery
Practical Electronics | December | 2020
LEDS
INSULATING
SLEEVE
M3 NUT
Q1
5mm LONG
M3 SCREWS
SILICONE
INSULATING
WASHER
PCB
6.3mm x M3
TAPPED SPACER
BOX
10mm LONG M3 SCREW
5mm LONG M3 SCREWS
Fig.10: this diagram clarifies how Q1, the LEDs and the PCB
itself are mounted in the case. Note the insulating washer
and bush (sleeve) under the M3 nut securing Q1, which are
critical, as Q1’s tab must be electrically isolated from the case.
temperature. As a consequence, its temperature should be
similar to the battery temperature. But a thermistor on the
battery is going to give more accurate results and therefore
a safer and more complete charge.
To make this external thermistor, a stereo 3.5mm jack
plug is soldered to one end of the twin-core cable, with
the thermistor soldered across the wires at the other end.
For the jack plug, connect the internal wires to the tip and
ring terminals, and the wire sheath to the jack plug sleeve.
The thermistor can be covered in heatshrink tubing and
attached to the side of the battery using adhesive-backed
hook-and-loop tape (eg, Velcro) or good quality doublesided tape for a more permanent installation.
Testing
Before applying power, it is vital to adjust VR5 to its lowest resistance by turning the adjusting screw 20 full turns
anti-clockwise. You can check that this has been done correctly by measuring the resistance between TP GND and
the 330Ω resistor at the end near the cathode of ZD1. The
resistance should be near to 0Ω. This prevents REG1 from
producing more than 5V when power is first applied.
Now connect a multimeter set to read DC voltage between
TP GND and TP5V. Connect a power supply to the charger
input (eg, a 12V DC plugpack or bench supply), press and
hold S1 and adjust VR5 for a 5.0V reading on the multimeter.
Check that the voltage between the pin 5 and pin 14 pin
on IC1’s socket is also 5V. If so, switch off power and insert IC1, taking care to orient it correctly and make sure
all its pins go into the socket and don’t fold up under the
IC body. Plug jumpers into JP1, JP2 and JP3 as required
for your battery.
Determine the maximum safe charging current
Most lead-acid batteries can accept up to 30% of the quoted
Ah capacity as charge current. For example, a 30Ah battery
can be charged at 9A. In this case, as long as your charger
is rated at no more than 9A, the 100% setting can be used.
If your battery is rated in RC (reserve capacity), you will
need to convert to Ah to calculate its maximum charge current. Reserve capacity indicates how many minutes a fullycharged battery can deliver 25A before the voltage drops
significantly. A battery with an RC of 90 will supply 25A
for 90 minutes.
The amp-hour specification (Ah) refers to the total current that can be supplied over a long period, usually 20
hours. So a 100Ah battery can supply 5A for 20 hours. To
convert from RC to Ah, multiply the RC value by 0.42,
which is the same as multiplying by 25A to get the capacity in amp-minutes, then dividing by 60 to convert from
minutes to hours.
In practice, because the RC capacity specification uses
25A, the conversion from RC to Ah often gives a lower Ah
value than the battery’s actual capacity. This is because the
Practical Electronics | December | 2020
Parts list – Clever Charger
1 double-sided PCB, code 14107191, 111 × 81mm
1 diecast aluminium box, 119 × 94 × 34mm [eg, Jaycar HB5067]
1 2A DPDT 5V coil telecom relay (RLY1) [eg, Altronics S4128B]
1 PCB-mount SPDT momentary pubutton switch (S1) [Jaycar
SP0380, Altronics S1498]
1 pushbutton switch cap for S1 [Altronics S1482, Jaycar SP0596]
1 SPST micro tactile switch with 0.7mm actuator (S2) [Jaycar
SP0600, Altronics S1122]
1 PCB-mount 3.5mm stereo switched socket (CON1) [Altronics
P0092, Jaycar PS0133]
2 PCB-mount M205 fuse clips (F1)
1 10A M205 fuse (F1)
2 NTC thermistors (10kΩ at 25°C) (TH1 and external thermistor)
1 2-way header with 2.54mm spacing (JP1)
2 3-way headers with 2.54mm spacing (JP2,JP3)
3 jumper plugs/shorting blocks (JP1-JP3)
1 18-pin DIL IC socket (for IC1)
1 3.5mm stereo jack plug
1 TO-220 silicone insulating washer and mounting bush (for Q1)
4 6.3mm-long M3 tapped spacers
3 M4 × 10mm machine screws
3 M4 star washers
3 M4 hex nuts
2 M3 × 10mm machine screws
8 M3 × 5mm machine screws
2 M3 hex nuts
4 insulated crimp eyelets (wire size 4mm, eyelet for M4 screw)
2 cable glands for 4-8mm diameter cable
1 2m length of 15A figure-8 automotive cable
1 1m length of twin-core shielded cable (for thermistor)
1 20mm length of 6mm diameter heatshrink tubing
2 large insulated battery terminal alligator clips (red and black)
6 PC stakes (optional)
4 small adhesive rubber feet
Semiconductors
1 PIC16F88-I/P micro programmed with 1410719A.HEX (IC1)
1 Si8751AB-IS isolated FET driver (IC2) (mouser.co.uk)
1 LM317T 1.5A adjustable positive regulator (REG1)
1 IRF1405N N-channel MOSFET (Q1)
2 BC337 NPN transistors (Q2,Q3)
3 green 3mm LEDs (LED1,LED5,LED6)
2 orange 3mm LEDs (LED2,LED4)
1 red 3mm LED (LED3)
2 18V 1W zener diodes (ZD1,ZD2)
3 1N4004 1A diodes (D1-D3)
Capacitors
1 220µF 50V PC electrolytic
1 100µF 16V PC electrolytic
5 100nF MKT polyester
5 10nF MKT polyester
1 10pF ceramic
Resistors (all 0.25W, 1% metal film unless otherwise stated)
1 100kΩ 1 51kΩ
3 10kΩ
1 3.3kΩ
1 2kΩ
7 1kΩ
1 330Ω
1 120Ω
1 100Ω 1W, 5% 1 56Ω
4 10kΩ multi-turn top adjust trimpots, 3296W style (VR1-VR4)
(code 103)
1 100Ω multi-turn top adjust trimpot, 3296W style (VR5) (code 101)
Small Capacitor Codes
Qty.
Value
µF value IEC code EIA code
5
5
1
100nF
10nF
10pF
0.1µF
0.01µF
n/a
100n
10n
10p
104
103
10
25
TO BATTERY
TO CHARGER
Fig.11: once the PCB is mounted in the case,
wire it up as shown here. Make sure that the
crimp eyelets are firmly secured to the board
using the specified washers and nuts.
CABLE GLANDS
Reproduced by arrangement with
SILICON CHIP magazine 2020.
www.siliconchip.com.au
18V
18V
4004
4004
+
Ah capacity usually requires much less current from the
battery, over a longer period.
Setting the charge current
For most large batteries, you would set the charge rate to
100%. To do this, adjust VR1 to get a reading of at least 1V
at TP1 relative to TP GND. You can use the 100% setting
for all batteries that can accept the full charge rate from
your charger.
If you need a lower current than your charger would normally supply, as explained above, adjust VR1 to reduce the
maximum charge rate.
This still applies the full current from the charger to the
battery but in bursts. For example, when the charge percentage is set at 50%, the charge will be bursts of full current for 50% of the time.
This would be suitable, for example, with a charger
that is rated at 4A and a battery that can only accept a 2A
charge current.
Divide the desired charge rate percentage by 100 and
adjust VR1 to get this voltage at TP1. So for our 50% example, you would adjust for 0.5V at TP1.
Note that when charging a 12V battery that initially
has less than 10.5V across its terminals, or a 24V battery
with less than 21V, the actual charge rate will be 1/10th
of that set. So for example, if you have set the charge rate
to 100%, it will be charged with a burst for 200ms every
two seconds. During this process, the Charge, Absorption
and Float LEDs flash.
Once the voltage comes back up into the normal range,
full-rate charging will start.
Current limiting
Very small batteries may not tolerate these high-current
bursts, even if they are limited in time. In this case, you
could add a series power resistor between the Charge Controller and your battery.
26
For example, when using a 12V battery and with a charger that typically
provides up to 17V peak, there will be
14107191
5V peak across the resistor. So the resisREV.B
C 2019
tor value required is 5V divided by the
peak current that the battery can tolerate. If the peak current is 1A, then the
resistance can be 5Ω (eg, one 4.7Ω re1
sistor or two 10Ω resistors in parallel).
Its wattage rating will need to be 5V
squared (25) divided by 5Ω. That gives
us a 5W dissipation, so to be safe, you
would use a 4.7Ω 10W resistor, or two
10Ω 5W resistors in parallel.
This is a conservative figure since 5W
is the peak power, not necessarily the
average power. The actual RMS voltage across the resistance will be around
30% lower than this, so the dissipation
will be around 50% lower. Therefore,
you could probably get away with a
5W resistor.
As mentioned, the charge LED can
be set to flash when current is applied
during the absorption and float phases. This indicates the
duty cycle used to charge the battery.
If the LED is off, then the battery is over the required
voltage for absorption or float. If the LED is not lit very often, then the battery is at the required voltage. If the LED
is lit continuously, then the battery voltage is still being
brought up.
SILICON CHIP
+
4004
COIL
19170141
LED option setting
The flashing LED option is on initially. If you do not require the charge LED to show during these phases, you can
disable this. Switching off power and holding S2 while
the power is re-applied using S1 will disable this feature.
The change is acknowledged by a minimum of two fast
(two per second) flashes of the Charge LED. The acknowledgement flashing continues until S2 is released. You can
re-enable the feature by holding S2 again at power up.
Setting the parameters
Most battery manufacturers will specify the required cutoff voltage (also called the cyclic voltage) for a given battery. For lead-acid types, the manufacturer will typically
also specify the float voltage (also called the trickle voltage) and the temperature compensation coefficient. Note
that the cut-off and float voltages must be the values specified at 20°C.
The temperature compensation required by manufacturers
is usually shown as a graph of voltage versus temperature.
You can convert this to mV/°C by taking the difference
between the voltages at two different temperatures and divide by the temperature difference.
For example, a battery graph may show the cut-off or cyclic voltage at 0°C to be 14.9V. At 40°C, it may be 14.2V. So
(14.2V – 14.9V) ÷ 40°C = –700mV ÷ 40°C = –17.5mV/°C.
Where the float temperature compensation is different
from the cyclic temperature compensation, a compromise
between the two values will have to be made.
Practical Electronics | December | 2020
Note that you can do this calculation over a smaller
temperature range if that is consistent with the temperatures under which you expect to be charging the battery, eg, 10-35°C if you live in coastal Sydney, but different in the UK!
To set the adjustable parameters, apply power to the
Charge Controller via a battery or charger and select the
battery type with JP3.
Then connect a multimeter between TP2 and TP GND
and adjust for one-tenth of the required cut-off voltage using VR2. So 1V at TP2 represents a 10V cut-off, 1.44V sets
it to 14.4V, and so on.
Now monitor the voltage at TP3 and adjust VR3 for the
required float voltage with the same 10:1 ratio.
For the temperature compensation, monitor TP4 and
adjust VR4 for the required compensation, with 1V representing –10mV/°C. So 5V represents –50mV/°C and 2V
represents –20mV/°C, and so on.
Once you’ve adjusted all these, make sure JP1 is inserted
and then press S2 to store the values.
The Thermistor, Charge and Float LEDs will all flash
twice to acknowledge that these values have been stored
successfully for lead-acid batteries. If adjusting the thresholds for LiFePO4 batteries, just the charge LED and absorption LED will flash.
You can store the parameters for each battery type by
changing the settings for JP3 and readjusting the trimpots,
then store the values again using switch S2. Adjusting the
trimpots without pressing S2 has no effect.
The adjustment of VR1, for the charge rate, is different.
This has an immediate effect. You will have to re-adjust it
each time you switch to charging a different battery that
needs a different charge rate than the last one.
To round out our description of this project, we should also
mention its possible shortcomings. These do not matter in most
cases, but may be significant in specific charging applications.
(1) Pulsed operation
The pulsed charging current can cause extra heating within the
battery as losses are proportional to the square of the current.
For example, when charging at an average of 1A from a 4A
charger, a 25% duty cycle is used. This averages to 1A, however, the losses are equivalent to charging at 4A2 × 25% = 4
times that of charging at 1A continuously.
(2) Absorption and float charge
We pulse the charge current, therefore the battery voltage
fluctuates during charging. We measure the battery voltage just
after the charge pulse finishes. Compared to a charger that has
continuous charging at a lower current, the battery voltage may
be maintained at a different value.
(3) Charge indication
As the battery supplies the circuit power via Q1’s body diode,
it can appear that charging is taking place even when the
charger is not connected or powered. It is important to check
that the charger is connected and is switched on when you
start charging.
(4) Battery discharge
If the ‘dumb’ charger is switched off with the battery
connected, the battery will eventually discharge due to the
50mA load of the Charge Controller. This is prevented using
a relay to switch off the power to the charge controller if the
battery voltage drops too low, but if this happens, you will
have to recharge the battery.
Order direct from
Electron Publishing
GET T
LATES HE
T COP
Y
OF
TEACH OUR
-IN SE
RIES
A
VAILA
B
NOW! LE
Charge Controller limitations
PRICE £8.99
(includes P&P to UK if ordered direct from us)
EE
FR -ROM
CD
ELECTRONICS
TEACH-IN 9
£8.99
FROM THE PUBLISHERS OF
GET TESTING!
Electronic test equipment and measuring
techniques, plus eight projects to build
FREE
CD-ROM
TWO TEACH
-INs
FOR THE PRICE
OF ONE
• Multimeters and a multimeter checker
• Oscilloscopes plus a scope calibrator
• AC Millivoltmeters with a range extender
• Digital measurements plus a logic probe
• Frequency measurements and a signal generator
• Component measurements plus a semiconductor
junction tester
PIC n’ Mix
Including Practical Digital Signal Processing
PLUS...
YOUR GUIDE TO THE BBC MICROBIT
Teach-In 9
Teach-In 9 – Get Testing!
A LOW-COST ARM-BASED SINGLE-BOARD
COMPUTER
Get Testing
Three Microchip
PICkit 4 Debugger
Guides
Files for:
PIC n’ Mix
PLUS
Teach-In 2 -Using
PIC Microcontrollers.
In PDF format
This series of articles provides a broad-based introduction to choosing and using a wide range
of test gear, how to get the best out of each item and the pitfalls to avoid. It provides hints
and tips on using, and – just as importantly – interpreting the results that you get. The series
deals with familiar test gear as well as equipment designed for more specialised applications.
The articles have been designed to have the broadest possible appeal and are applicable to all branches of electronics.
The series crosses the boundaries of analogue and digital electronics with applications that span the full range of
electronics – from a single-stage transistor amplifier to the most sophisticated microcontroller system. There really is
something for everyone!
Each part includes a simple but useful practical test gear project that will build into a handy gadget that will either
extend the features, ranges and usability of an existing item of test equipment or that will serve as a stand-alone
instrument. We’ve kept the cost of these projects as low as possible, and most of them can be built for less than £10
(including components, enclosure and circuit board).
© 2018 Wimborne Publishing Ltd.
www.epemag.com
Teach In 9 Cover.indd 1
01/08/2018 19:56
FREE COVER-MOUNTED CD-ROM
On the free cover-mounted CD-ROM you will find the software for the PIC n’ Mix series of articles. Plus the full TeachIn 2 book – Using PIC Microcontrollers – A practical introduction – in PDF format. Also included are Microchip’s
MPLAB ICD 4 In-Circuit Debugger User’s Guide; MPLAB PICkit 4 In-Circuit Debugger Quick Start Guide; and MPLAB
PICkit4 Debugger User’s Guide.
ORDER YOUR COPY TODAY JUST CALL
01202 880299 OR VISIT www.electronpublishing.com
Practical Electronics | December | 2020
27
|