TPS54231: TPS54231 working is not stable in the Light Loads condition.

i don't have the EVM now, i will find the rignt person to help this as soon as possible

Vental,
Thanks very much for your support and waiting for your reply.

BR
Kelly

Hi Kelly,

Do you have the waveform for PH node when you encounter this kind of situation? It will show us a more clear pic if we could check on PH. From my understanding, your boot to PH voltage will be low since the device is working under LDO mode and your Vin is low. The internal high side MOSFET could work in saturation region which will have a pretty high Vds. This will make the output voltage low.

Hi Andong:
This part has the boot charge,so it can work at low Vin.
Do you have an EVM on hand? I have no EVM now.

Hi Kelly,

Vout=3.0V is very close to Vin=3.5V, an external VIN UVLO is recommended to add hysteresis. please confirm it first, and Anthony gave a good suggestion to double check the boot voltage, if it like what he said. you could add some dummy load at the output side, very light load is recommended.

BRs,

Ryan

Kelly,
TPS54231 is a non-synchronous buck converter. At status VIN 3.5V VOUT 3V Iout 5uA, the voltage between BOOT and SW will decrease near (3.5v -3.3V), then high side mosfet can't turn on normally. so for this device, ensure VIN-VOUT >=2V which Ryan mentioned.
Suggest TPS54202 for this application. Note, TPS54202 input voltage range is 4.5V to 28V.
www.ti.com/.../tps54202.pdf

Hi Anthony

The Green is PH

The Yellow is Vout

  Many thanks

Hi Jason,

For longer working time, the customer hope Vin Min is 3.5V.
Any suggestion?

BR
Kelly

HI Kelly,

If green is PH, then it looks like the high side FET will be turn on for a pretty long period and the PH voltage will continuously drop due to boot cap voltage being low. The boot cap could only be recharged at boot UVLO and low side FET will be on for short time to charge up the boot cap. This will make your output voltage lower than expected since high side FET is not fully on. To let high side FET fully on, please make sure to have some top room between Vin and Vout to avoid LDO mode.

Hi Kelly,

please refer to the application note SLVA547A, you could chose the best suitable solution for your application to avoid the BOOT UVLO.

BRs,

Ryan

Hi Ryan,

    Thanks very much for your reply.

I am checking the SLVA547A now.

Many thanks.

BR

Kelly

Kelly,
the root cause is clear. for non-synchronous buck, when VIN-VOUT is about 2V-3V & light load, buck will enter "LDO mode" (no eough power to turn on high side mosfet).
For your application, you should change your design.

Dear Jason,

 Thanks very much for your support.

The customer modified the schematic; add the diode tied from input to BOOT.

Dear Jason,

Thanks very much for your support.

The customer modified the schematic; add the diode tied from input to BOOT.

When Vin more than 3.8V, Vout = 3V, Iout = 2mA. The system works stable.

When Vin between 3.5V and 3.8V, the system will not work stable until the Iout above 20mA.


If customers insist on using 3.5v input, any suggestions for improvement?

Many thanks.

BR

Kelly

Jason,

I'm so sorry for replying you late.

Finally, the customer decides to modify the maximum Vin, and choose TLV62130A.

Many thanks for your support.

BR

Kelly