tps63000: Device failure on when current is absent

Can you be more precise in what you mean with failures? Is the device damaged or is just the output voltage not regulated?
It may be helpful if you can post a schematic of your test setup.

Ok. This schematic follows the datasheet recommendations, so I don't think there is a problem.

What is connected to the input (labeled +5VUSB)? Is this going directly to an USB port compliant to USB standards regarding supply current capabilities?

Hi Juergen,

By failure, the device is permanently damaged and fails to generate the output voltage.

The only way to fix this issue is by replacing the IC.

The +5VUSB input is connected to a USB 5V dedicated charging port, mains wallwart, 1A current and protected by a 5V6 zener and 1.3A fuse.

The IC needs approx. 0.8A (with no soft start implemented). to turn on properly. When the current is restricted on start up (using a bench power supply), e.g. 5V5 @ 0.7Amps, the device fails to start. This collapses the input supply voltage to 2V6 volts whilst current limiting at 0.7A using a bench power supply. The device is still enabled as the cut off threshold is 1V8.

In this state the IC continues to pull 0.7Amps indefinitely but no output is provided. The power is being dissipated within the IC and this can be seen using an infrared camera. This power dissipated of 1.8W well exceeds the datasheet spec of 820mW. I am unsure if the thermal cut off is working as the device does not shutdown and continues to draw 0.7A.

• Could you confirm if the soft start circuitry required for the IC to behave properly? A previous iteration of the datasheet provided an example where no soft start was implemented, tying VINA, EN, PS/SYNC tied to VIN
• Can the device be damaged is starved of current?
• Have there been any similar reported failures

Many thanks,

Laurence Steele

The R-C connected between VIN, VINA and GND has nothing to do with startup. It is just used to provide a less noisy supply to the internal control circuit. If you want to slow down the startup I recommend to use the circuit which is described in this application note.

When reading your description of the experiments you did I guess you have damaged the IC with overvoltage on VIN, L1 or L2. Did you monitor the voltages and currents with an oscilloscope during the experiment?

It also would be helpful for further debugging if you can post the layout of your testboard.

Hi Juergen,

Yes, I have been monitoring the current and voltages on VIN. The absolute max input voltage on VIN is 7V. I have not exceeded this voltage on VIN as the zener will clamp at 5V6. Please see below a copy of the layout.

The layout around the IC looks ok to me. It is not completely following the datasheet recommendations. The way you connect the feedback divider exposes the internal regulator to more noise.
Anyway this does not explain why the device gets damaged. If the device is soldered properly including the exposed thermal pad, I still consider it most likely that you have caused an overvoltage condition with an oscillation you get on the input either by plugging in the USB cable or by the way the current limiting of your power supply interacts with the control of the device. The inductance of the supply cables together with the input capacitor form a nice resonant circuit. Delaying the startup to a time when the supply voltage has settled or slowing it down significantly can help mitigating problems related to that.
Please keep in mind that the maximum recommended supply voltage for operating the device is 5.5V. 7V is a stress rating only, not recommended for operation. A 5.6V zener diode may not be sufficient for having a reliable input overvoltage protection. Using a higher voltage rated device, like TPS63070 may be an easy solution for this problem.

Hi Juergen,

How does the thermal cut out on the device operate?

When the device is restricted of current at startup using a bench power supply (to simulate a user using an underated power supply) the input supply collapses to around 2.6V.

The device is therefore still enabled as EN>1.8V.

With the current limit restricted to 0.75A and input supply at the collapsed 2.6V the power dissipated in the IC is 1.95W. This is more than double the maximum dissipated power stated in the datasheet of 820mW but the device never shuts down and constantly draws 0.75A.

Would you expect any damage or premature failure of the device due to the above?

Cheers, Laurence

Hi Laurence,

Sorry for the delayed answer.
Were you able to try out the soft start circuit described in this application note: www.ti.com/.../slva553.pdf ?
By reading through the post, it looks like there is an overvoltage at the input is causing the damage.
Could you monitor the input voltage during start up?

thanks
Sabrina