if the input1 shake,there output is low,but do not switch with the shake

Hi User,

I noticed that you have made two posts on E2E. I have closed the other thread (https://e2e.ti.com/support/power-management/f/196/p/930970/3439961#3439961) and we will support this issue here.

Can you please provide a schematic of your setup of TPS2121? This will help us troubleshoot your issue. Your waveform is a little hard to read so can you also provide the voltage levels of both inputs and the expected current for the load during normal operation?

During XREF switching without hysteresis, there is a chance that the device bounces between the inputs 

Hi，Shreyas

  we test further for the TPS2121  in the XREF mode.we use the Mechanical switch to switch the in1 on or off. The in 2 is always on.And output is 1.5A  electronic load.But we see ,when  Mechanical switch on,there is some shake on the in1.but the output  is low to 0V.and the device do not bounces between the inputs .Here the schematic .

the test wave is below.

Hi User,

Thanks for the schematic and detailed waveform. The delay in switching is occurring due to the mechanical switching on VIN1.

As you can see, the input rises in voltage which triggers the switchover incorrectly. The device tries to switch from IN2 to IN1 but IN1 is 0V (please look at the fourth division on the last waveform you shared). This triggers a UV fault in the device and the device soft starts into VIN2.

Then IN1 starts to shake again causing another trigger event which is only resolved when IN1 is stabilized.

Is it possible for you to add more input capacitance (1uF) to IN1 to avoid this voltage shake? Another workaround may be to add a 10nF capacitor from PR1 to GND. 

Can you retest the board with these changes?

Hi，Shreyas

Thank you for your reply.we will retest the board with your suggest  changes  quickly .But   "This triggers a UV fault in the device and the device soft starts into VIN2" I have question about this. This mean   when the output is 0v  ,the device soft starts into VIN2,but the VIN2 soft starts time is too long .The time is about 500us. It is normal?

Hi，Shreyas,

Thank you.we update the schematic below.And retest the board with your suggest  changes  quickly,the result is that there is no shake on the in1,so the output is normal .The waveform is also below.The output we  series a 10uh  inductance，for the  inductance can  release the energy to charge the two 100uf  capacitance. But we do not know if there is  negative impact of the 10uh  inductance.So I think the tps2121 input can not have shake on the input.And now we use the 22uf to filter the shake.But you know,the shake is random，every time the shake maybe different.So I'm worried if there is a big shake， the 22uf  can not  filter，the output maybe low again.So,i am right?

Hi Shreyas,

We take your suggestions to add 22uF capacitor at the input of main battery, and 10nF capacitor at PR1 on the PR1 node, and 2x100uf capacitor at output end. The main battery has no obvious "shake", but there is still  incorrect switchover behavior when the main battery is taken away at sometimes. I wonder why this abnormal switching phenomenon still occurs. In addition, we want to know whether there is "shake" or not What is the cause? Is it over-current, under voltage or other reasons? Is there any fundamental solution?

Test waveform as below