Using a single 7812 IC voltage regulator and multiple outboard pass transistors, this power supply can deliver output load
currents of up to 30 amps. The design is shown below:
The input transformer is likely to be the most expensive part of the entire project. As an alternative, a couple of 12 Volt car
batteries could be used. The input voltage to the regulator must be at least several volts higher than the output voltage (12V)
so that the regulator can maintain its output. If a transformer is used, then the rectifier diodes must be capable of passing a
very high peak forward current, typically 100amps or more. The 7812 IC will only pass 1 amp or less of the output current, the
remainder being supplied by the outboard pass transistors. As the circuit is designed to handle loads of up to 30 amps, then six
TIP2955 are wired in parallel to meet this demand. The dissipation in each power transistor is one sixth of the total load, but
adequate heat sinking is still required. Maximum load current will generate maximum dissipation, so a very large heat sink is
required. In considering a heat sink, it may be a good idea to look for either a fan or water cooled heat sink. In the event
that the power transistors should fail, then the regulator would have to supply full load current and would fail with catastrophic
results. A 1 amp fuse in the regulators output acts as a safeguard.
Simulation with LTspice XVII
The circuit above is a simulation in LTspice XVII. In the simulation the output load is stepped
in 0.3 ohm increments, starting at 0.4 ohm increasing to a 3 ohm load. The plot of load current
versus load resistance is shown below.
The regulation performance is simulated below. With a 30A load the output voltage is 12.0016V
and with a full 30A load the voltage drops to 11.9986V. This is a change of just 0.003V or 3mV.
For anyone wanting to experiment further the LTspice download circuit is available at the bottom of
This circuit is a fine example of Kirchhoff's current and voltage laws. To summarize, the sum of the currents entering a junction,
must equal the current leaving the junction, and the voltages around a loop must equal zero. For example, in the diagram above,
the input voltage is 24 volts. 4 volts is dropped across R7 and 20 volts across the regulator input, 24 -4 -20 =0. At the output
:- the total load current is 30 amps, the regulator supplies 0.866 A and the 6 transistors 4.855 Amp each , 30 = 6 * 4.855 +
0.866. Each power transistor contributes around 4.86 A to the load. The base current is about 138 mA per transistor. A DC
current gain of 35 at a collector current of 6 amp is required. This is well within the limits of the TIP2955. Resistors R1 to R6
are included for stability and prevent current swamping as the manufacturing tolerances of dc current gain will be different for
each transistor. Resistor R7 is 100 ohms and develops 4 Volts with maximun load. Power dissipation is hence (4^2)/200 or about
160 mW. I recommend using a 0.5 Watt resistor for R7. The input current to the regulator is fed via the emitter resistor and base
emitter junctions of the power transistors. Once again using Kirchhoff's current laws, the 871 mA regulator input current is
derived from the base chain and the 40.3 mA flowing through the 100 Ohm resistor. 871.18 = 40.3 + 830. 88. The current from the
regulator itself cannot be greater than the input current. As can be seen the regulator only draws about 5 mA and should run cold.
Initial Testing and Faulting
For the initial test, do not connect a load. First use a voltmeter across the output terminals, you should measure 12 Volts, or
very close to it. Then connect a 100 ohm, 3 Watt resistor or other small load. The reading on the voltmeter should not change. If
you do not see 12 Volt, power off and check all connections.
I have heard from one reader whose supply was 35 Volt, not the regulated 12 Volts. This was caused by a short circuited power
transistor. Should a short in any of the output transistors, occur, all 6 need to be un-soldered.
Check with a multimeter set to resistance and measure between collector and emitter terminals. Power transistors usually fail
short circuit so should be easy to find the faulty one.
Finished PSU by Ryan Laurencia
I've recently heard from Ryan Laurenciana in the Philippines who has built himself a 12V 30A power supply. Below are images from
Ryans power supply.
Finished PSU by Alejandro from Venezuela
Below are some pictures from Alejandro in Caracas, Venezuela. Alejandro's power supply is used to drive his amplifier.
Finished PSU by Aamer Reza from Pakistan
Below are some pictures from Aamer Reza who built his 12V 30Amp power supply and added water cooling.
Some notes from Aamer about his setup:
For better understanding, I have sketched the inside of the heatsink.
The body I got casted in Aluminium from a casting shop. I smoothed the sides by myself.
I used a 2mm thick Aluminium plate for the cover.
Drilled 6 holes for the transistors and 4 for fixing.
Drilled 2 holes for INLET and OUTLET.
Used 2 brass nozzles (IN/OUT) with Epoxy adhesive to ensure leak-proof bonding.
Used Silicon RTV caulk between the body and cover to form a gasket. Kept it overnight in a vice.
One thing, I would like to share -------- The 12V output should be collected right from the connecting point of all the COLLECTORS
and the OUTPUT of LM7812. PCB traces are not reliable for 30A load. Even the trace for the GROUND should be wide enough to
withstand the high current. I had to reinforce the traces by soldering copper wire.
LTSpice Download Circuit
This simulation runs under LTSpice IV and LTSpice XVII. Just unzip the file into its own
directory and open 12v_30a_sim.asc from within LTSpice.