TPS54494 stability

Balint Jozsef Janvari

Other Parts Discussed in Thread: TPS54494

Hello,

I am trying to use the TPS54494 IC as a dual 3.3V / 5V power supply - with little success. The output is mostly stable, but every 20-30ms the IC seems to "stop switching". I have attached an oscilliscope reading:

There was very little load on the power supply (around 5-10 mA) when these measurements were made, additional load will increase the voltage drop during these phases.

I am using a 4.7uH inductor (Eaton DRA124-4R7-R) and a 47uF capacitor (TDK C2012X5R1A476M125AC) on both sides.

Any idea what may cause this problem?

Thanks,
Bálint

over 10 years ago

0 JohnTucker over 10 years ago

TI__Guru**** 191445 points

What is the waveform you are showing? can you post your schematic? Did you follow the layout guidelines?

0 Balint Jozsef Janvari over 10 years ago in reply to JohnTucker

Prodigy 40 points

The waveform is the 3.3V output of the circuit. I have attached the schematic.

0 JohnTucker over 10 years ago in reply to Balint Jozsef Janvari

TI__Guru**** 191445 points

What is the VBAT voltage? Can you post waveforms of VBAT, SW2 and VREG5? What is teh 5 V output on the CH1 side doing? Is it operating correctly?

0 Balint Jozsef Janvari over 10 years ago in reply to JohnTucker

Prodigy 40 points

VBAT is a 3 cell LiPo battery, designed to operate between 9.6V - 12.6V. I am no longer near the lab, but I will post waveforms as soon as I am able.

0 Balint Jozsef Janvari over 10 years ago in reply to Balint Jozsef Janvari

Prodigy 40 points

This is what the SW and the V33 signals look like side by side:

A closeup on "normal" part of the two signals:

VREG5 and VBAT is stable with very little noise.

0 JohnTucker over 10 years ago in reply to Balint Jozsef Janvari

TI__Guru**** 191445 points

It looks like you are operating in eco-mode (light load condition) then suddenly the circuit stops switching. Usually that would be EN going low, VREG5 out of regulation or current limit causing under voltage protection. do you have any system event that occurs when the voltage rail first drops out? Since teh SW node waveform peak voltage drops a lot, it implies a heavy load during that time. It may also be an issue with your layout and the location of critical components and traces. Can you post your layout? Can you also show a waveform at say 5 usec/div triggering on the first falling edge of the 3.3 V rail? Show SW and if possible the inductor current.

Power management

Power management forum

TPS54494 stability