TPS563240: Weird bode plot (DCAP3)

Hi Jay

Would you please take a picture of your set up and also the screenshot of the bode set? Many thanks.

BR

Ruby

Hi Ruby,

Here's the pictures and screenshot.

Thanks,

Jay

Hi, Jay

Would you please take a picture of the location of the ground point in the PCB when testing the bode plot and your SW node operating waveform in the time field? Thanks.

BRs

Wenhua

Hi Jay

1. Regarding the set up of the bode, would you please change the BW to 10Hz, this can help you avoid some noise.

2. As Wenhua mentioned, would you please do me a favor to capture the waveform of SW and Vout so that I could see whether there is any noise here.

3. And also the test ground is important, please let the test ground be close to the device to avoid additional noise.

Hope this helps.

BR

Ruby

Hi Ruby and Wenhua.

Thanks for your suggestions.

1. BW to 10Hz seems to help avoid some noise. But the result seems similar.

2. Here's SW waveform. Unfortunately I couldn't get a scopeshot of Vout. 

3. The white cable is ground cable which seems close enough to IC. I took it when measuring bode plot.

Do you have any suggestions?

Thanks,

Jay

Hi Jay,      

Here is a file of the bode plot test, where you can refer to the test principles and compare your test wiring to see if there is anything different.

 Considerations for Measuring Loop Gain in Power Supplies.pdf

If you still don't find the different and if you're in a hurry to use the bode plot, you can send me your schematic (including the device parameters), and I can test the bode plot on the EVM with your device parameters to test if there's a problem and send it to you for comparison.

BRs

Wenhua

Hi Wenhua,

Is there an EVM for TPS563240? I couldn't find it but TPS563249EVM-031.

Wenhua Wang said:

I can test the bode plot on the EVM with your device parameters to test if there's a problem and send it to you for comparison.

I'll be my honor if you do this for me. I tried several times but didn't work. 

You can see some parameters as above SCH. Would you draw 1.2A for the Iout?

Thanks, 

Jay

Hi Jay,

I'll send it to you after I've tested the bode plot.

BRs

Wenhua

Hi Wenhua,

Thanks for sharing your result.

Wenhua Wang said:

Would you please tell me whether your application can get a steady state output voltage of 3.3V under 5.1V input voltage and 0.3a, 0.8, 1.2A loads?

I got 3.2V for maximum Vout as well. Also Vout is reduced to 2.8~2.9V in 0.8A~1A load.

Wenhua Wang said:

I tested the bode plot at 0.2A load current at 5.1V and 6V input voltage respectively, as shown below, the bode plot are really messy at 5.1V input voltage. The bode plot at the 6V input voltage is normal, and the output voltage is 3.3V.

So You mean, TPS563240 is not suitable for 5.1V input and 3.3V out, is it? Is there any ways to use this device on current customer's power rail?

Thanks,

Jay

Hi Jay

TPS563240 cannot support this power rail because of max duty limitation.

You can try to replace it with other device, such as TPS563212/TPS562212, which can support your application with more good feature and better performance.

TPS562212: 18V 2A, 1.2MHz

TPS563212: 18V 3A ,1.2MHz

BRs

Wenhua

Hi Wenhua,

Thank you so much for your suggestion. 

One last question here. Do you think increasing BST cap value will help this case?

I read a thread as below.

https://e2e.ti.com/support/power-management-group/power-management/f/power-management-forum/1093981/tps54233-how-much-minimum-input-voltage-is-required-for-output-3-3v-1a?tisearch=e2e-sitesearch&keymatch=1.4Mhz%2525252525205V%2525252525203.3V#

---

"I once met a case in which the input voltage is 5V and output voltage is 3.3V, load current is hundreds of mA and the boot cap is 100nF. In this case, the customer met the BOOT UVLO issue (defective rate 3%). They used TPS54332 which is also a 28V device. But they did not meet the issue before because they changed some devices in the BOM such as the diode. I tested 100nF+47nF for the boot cap in the lab and it still had the issue, and I used the 220nF for this application, it could work. To avoid the issue, finally, we suggested 470nF for the application."

---

BTW, It was Current mode Buck not DCAP3.

Regards,

Jay 

Well noted. Thanks Wenhua.