Hacking a buck converter - TL494

In this article we will be hacking this ALiExpress buck converter. Made around the TL494 PWM controller IC from Texas Instruments.

Intro

Have a look at cheap buck and or boost converters available on popular marketplaces like eBay, Temu, AliExpress and Amazon.

They all are controlled by those familiar blue potentiometers. That on its own is not a complaint, that is great actually, we can set a hard limit to the maximum voltage and or current they will let through.

But I want to do more with them, change the duty cycle while still within those bounds. I want to dim them, with a signal from another device.

But how do we do that? Well… we need to hack the feedback pin of the converters main controller IC.

Let's do that for this specific converter.

The device

The seller lists it as follows:

"Step-down Module 12V-75V to 5V 19V 21V 24V 48V 600W 25A High-power Converter"

600 Watts and 25 Amps. I doubt those capabilities. Mainly because of the PCB to be fairy thin, so there is not that much copper in it.

If we look at the ICs on the board there is one that sticks out. And it is the TL494C (G4?). That is the one we need, it's the PWM controller IC.

Datasheet

Alright, lets have a look at the datasheet to figure out what we have to work with and how we can take control over it.

Datasheet https://www.ti.com/lit/ds/symlink/tl494.pdf

Relevant pages in the datasheet for this project: 1,3,8,10,13,15,16

The datasheets reference design comes from this application note, going into more detail.

Application note https://www.ti.com/lit/an/slva001e/slva001e.pdf

Datasheet page 1 - intro

If we were to create a brand-new buck converter, this IC would be the thing we need. It covers most if not all the challenges we would encounter.

It can do the PWM generation, obviously, but also dead-time control and has comparators for creating a feedback signal for voltage and current control.

At the top there are 2 op-amp's. These are for feedback amplification.

We can tie the negative terminals to a fixed voltage, that would become our limit. In reality those limit voltage would be set by those blue potentiometers.

Then the positive terminals are connected to real world voltages, from the output voltage or current (voltage over a shunt resistor)

If then either of those go over the input terminals voltage, the op-amp would go all out rising voltage in the direction of the upper rail (vcc/5v?) until there is balance.

These outputs are connected internally to the feedback pin.

Please note: There are diodes inside so these can only source voltage/current to the feedback pin.

Datasheet page 3 - pinout

These are the descriptions for the pins on the IC.

The most important one is the feedback-pin (3), it is an input-pin, for us to do as we please.

And a ground pin, we need that if we want to send signals from outside.

Datasheet page 8 - test circuit

This schematic shows us a test setup, so we can see what the device does when given control signals to DTC and FEEDBACK.

• C1 is the regular signal the buck converter output for controlling the ON time of the switching circuit.
• C2 is an 180-degrees out-of-phase version of signal C1.
• CT's voltage at CT comes from an oscillator charging and discharging a capacitor, creating a triangle wave signal.
• DTC is a manual signal in this example. On the actual board it is fixed. When that voltage rises we can see to OFF-time getting smaller.
• FEEDBACK also is a manual signal, it controls the duty cycle. That becomes larger the higher that FEEDBACK voltage is.

NOTE 1: I suspect this buck converter to invert the output signal C1 or C2. During testing the output voltage of the converter dropped when increasing voltage to FEEDBACK.

This is the basis for the PWM output signal.

--> traingel to square image <--

If we have a variable voltage and compare that with a comparator to the triangle signal, we can use that to create a PWM signal with a variable duty cycle.

The DTC voltage at pin 4 does that to force a minimal OFF time and the FEEDBACK voltage at pin 3 does that for duty cycle.

Datasheet page 10 - error amplifiers

There are 2 error amplifiers, these are to limit the circuits output. For this buck converter there is only a maximum voltage (blue potentiometer).

The amplifier compares a set voltage IN- to the actual voltage IN+. And it is in a negative feedback configuration.

Also, FEEDBACK is (slightly) pulled down to ground. So the converter starts at largest duty cycle. Remember the output PWM is inverted.

So as soon as the set voltage-limit is reached it starts to output voltage (through the diode) into feedback, tuning down PWM until the bucks' output voltage is stabilized.

The device uses a resistor+capacitor pair to set the frequency of the PWM. I needed, we could modify that a well with different values.

Datasheet page 13 - reference schematic

The next image will be a reference schematic for this IC, you can find more details about it in another PDF. The link for that is in the first part of this article next to the link to this PDF.

This is a reference schematic. That is a know good/working circuit. Our buck converter is most likely of modification of this circuit.

Datasheet page 15 - error amplifier voltage control

And here we have a bit more detail on the error amplification part of the circuit.

The VREF is 5v on this IC, and goes through an equal valued resistor-divider (R3+R4), so the negative input sees 2.5v. R5 and R7 do some dampening, but the voltage stays 2.5v.

The VO, the bucks' output voltage also goes through a resistor-divider (R8 andR9), and also equally sized. And then send to the positive input of the amplifier.

So that means, if the VO is 5V, the positive input sees 2.5v. The same as the other input. That means this circuit is targeting an output of 5V!

Datasheet page 16 - PWM generation

The graph here shows that when voltage at pin4 (DTC) changes in relation to the triangle wave, the duty cycle changes.

This also goes for pin3 (FEEDBACK).

In short, if one voltage crossed the other, the output of the comparator changes, which eventually becomes the output PWM signal.

The plan

So now we know a couple of things.

• The FEEDBACK pin is a valid input.
• Error amplifiers limit voltage and or current by putting voltage on the FEEDBACK pin.
• Voltage on FEEDBACK is in direct relation to PWM output.

So by default the circuit starts at max PWM, until voltage/current limiting kicks in by dimming the PWM.

And we can also dim the PWM by adding our own voltage to the pin.

NOTE 2: We also need to add a diode so we can only add voltage. Otherwise, we could accidentally draw down the voltage and override the voltage/current limit.

The interface wires

So I soldered a wire to pin 3 (FEEDBACK) and another to pin 7 (ground).

Controlling the output

This is my very first attempt to influence the output voltage. It works really well.

Now I have to build an Arduino with a DAC so I can control it digitally.

-- To be continued --