Once again I am playing with an IC I dug out of an old piece of equipment. Today it is the Texas Instrument TL 494 Pulse-Width-Modulation Control Circuit. The Data Sheet:
which indicates that it was originally designed cerca 1983 but it remains relevant as Newark Electrons continues to supply it.
The TL 494 has everything needed on chip to generate and control a PWM signal. This in turn can provide the ability to drive heavier current carriers such as transistors. The chip has two error amplifiers that can be used to control the output and an internal voltage reference and oscillator. The flexible output stage has 2 uncommitted transistors that can be hooked up as either common-emitter or emitter-follower configurations. The relatively low cost per unit from Farnell ($0.69) makes this a very desirable circuit for design of bench power supplies as it simplifies and integrates most of the circuitry needed.
The Data sheet has a sample circuit along with an excellent section by section description of the formulas and parameters for designing the circuit. Here is a schematic of the sample circuit:
It is my indention in this exercise to breadboard this circuit and experiment with adjusting the output voltage and the current limiting circuit. As I proceed you will notice that I have strayed from using the exact components listed in the schematic. For example I might use a 47 K resistor where the schematic calls for a 51K or my inductor is a 100 uH instead of a 140 uH. These discrepancies will affect the final performance of the circuit. Other factors that will adversely affect the performance will be the resistance of the bread board contacts, long leads and jumpers, and the layout of the parts on the breadboard. While these things are all of concern their adverse effects are outweighed by the easy availability of the parts I will use and level of imprecision that I am willing to accept. This is an experiment and if there are serious problem then more will be learned as I search for a solution.
The above circuit, as designed, will supply 5 volts at 10 amps. I will begin with an attempt to get close to 5 volts with a limit of 2 amps. Since the contacts in the bread board and the gauge of the wires being used will not easily accommodate 10 amps. Here is a picture of the bread boarded circuit:
As you can see I have placed Q1 and Q2 on thier own heat sinks with jumper wires and bread board pins to tie into the circuit. I am using 4.7K resistors for R8 and R9 which are the output voltage sensing divider. As they have the same proportion to one another as the original 5.1 K we should not see any difference in the output voltage. The original design calls for R13 ( the current sense resistor) to be 0.1 Ohms. In my first test I will be using a 0.47 ohm resistor which should bring my current limitation down to about 2 amps. An electronic load will be used to test the circuit and I am monitoring the voltage output with the Fluke and the activity of the output of the TL 494 with the scope. The power supply will be set at approximately 27 volts for the input supply and I can also monitor the current draw on the supply with the meters of the power supply.
The load is reading 4 volts due to the voltage drop in the sense resistor and the wires and connections. The scope illustrates the wave form of the output of the TL 494 on pin 11. The regulation of the power supply was well within expectations based on the inherent problems of connections and wire gauge being used in the circuit.
My next test involved changing the values of R8 and R9 to produce an output closer to 12 volts. Here is a photo where I have labeled R8 and R9.
Here is what the scope looks like with no load on the circuit:
Now with R8 still at 4.7k and R9 at 1.2k the output of the circuit went to 12 volts as expected. Turning the load up the current increased until the voltage across the sense resistor R13 tripped the error amplifier in the TL 494 and the current dropped back to the limit imposed by the value of R13 compared to the voltage divider R1 and R2 supplied by the internal voltage reference. I also experimented powering resistive loads (auto brake lights). The duty cycle of the PWM on the scope would increase as the load was increased up until the current cut off and then then stabilize.
In case you want to look closer at my bread board here is another picture with some labels:
While the TL 494 is rather old I can see many uses for it in circuits that I may want to design in the future. It also provided a good evening of exploring and learning a little bit more about electronics.