TL2575-05I Heat, noise and wrong output

Hamze Hachem

Hi,

I'm troubleshooting a circuit built by a fellow electrician. I traced the problem back to TL2575. Already at a 12V input voltage, the IC is getting really noisy and extremely hot. On the other hand, it is only delivering 3.4V instead of the promised 5V. I checked the circuitry and it is exactly as recommended in the datasheet. Can anyone guess what could be wrong?

Thanks in advance

over 11 years ago

0 Ron Michallick over 11 years ago

TI__Guru*** 140136 points

Hamze,

A good place to start is measuring input current and output current.
Then capture waveform on pin 2 (output) with an oscilloscope.

The power efficiency can be calculated as (VOUT*IOUT)/(VIN*IIN).
The power lost (self heating) is (VIN*IIN)-(VOUT*IOUT)

The pin 2 waveform shows switch on time, ans inductor discharge time and shows voltage losses.

0 Hamze Hachem over 11 years ago in reply to Ron Michallick

Prodigy 140 points

Hi,

Thanks for the reply. Here's the waveform on the output pin:

According to the datasheet, the voltage is not supposed to drop below 0!

The input power is about 6W, which is way too high. I cannot measure the output current, as I do not have a suitable probe.

0 Hamze Hachem over 11 years ago in reply to Hamze Hachem

Prodigy 140 points

Here's the schematic:

0 Ron Michallick over 11 years ago in reply to Hamze Hachem

TI__Guru*** 140136 points

Hamze,

The output is supposed to go a little negative about 1V (or less) when TL2575 switch is off and the external catch diode is conducting. The 1N4001 is a good diode for a 60Hz sine wave input, but it is very poor diode for a 52kHz square wave (steep edges). The diode needs to be a Schottky power diode, a 1N5819 for example.

The schematic is labeled VDD/2.1A. The TL2575 is rated up to 1 Ampere, not 2.1 amp.

0 Hamze Hachem over 11 years ago in reply to Ron Michallick

Prodigy 140 points

Hi,

I believe you are right. I'll try to switch the diode tomorrow. As for the 2.1A, it has nothing to do with the current, it was added automatically by the layout software to distinguish between grounds I believe.

0 Hamze Hachem over 11 years ago in reply to Ron Michallick

Prodigy 140 points

Hey Ron,

I switched the diode with an PMEG4050EP from NXP and now it works. Thanks for the support.

Regards,

Hachem

0 Ron Michallick over 11 years ago in reply to Hamze Hachem

TI__Guru*** 140136 points

Hamze,

I am glad it is working now.
For anyone else who read who read this thread, I attached a good output pin waveform when VIN was 12V and output was 1 amp (5V).
There is a good high and flat high output voltage and a low (negative) and flat low output voltage.
Coil current is continuous (always greater than zero).

TL2575 12V 5V 1A.TIF

0 Hamze Hachem over 11 years ago in reply to Ron Michallick

Prodigy 140 points

Hello Ron,

I'm having problems again. After many months of running successfully, the step down converter stop outputting the right voltage.

Here's what I can see at the output pin:

However, a multimeter is indicating that all pins on the component are somehow connected together, which is abnormal. I replaced the diode, capacitors and coil but to no avail. I'm guessing the component died along with the MCU which it was powering.

0 Ron Michallick over 11 years ago in reply to Hamze Hachem

TI__Guru*** 140136 points

Hamze,

What is the output voltage and load for this waveform?

0 Hamze Hachem over 11 years ago in reply to Ron Michallick

Prodigy 140 points

Thanks for the reply. The output voltage for this wave is about 60mV . I have no load connected, as I am afraid, it might under or overpower my other components

0 Ron Michallick over 11 years ago in reply to Hamze Hachem

TI__Guru*** 140136 points

Hamze,

I do not know why the output is wrong. With a low feedback pin voltage, the output should be a very high pulse width.
I see a low duty cycle and it is skipping every other pulse (26kHz not 52kHz).

Power management

Power management forum

TL2575-05I Heat, noise and wrong output