Hi team,
We are evaluating TPS71711DCK with different Cnr.
Only when Cnr is 10nF, overshoot occurs.
What is the root cause?
Best Regards,
This thread has been locked.
If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.
Hi team,
We are evaluating TPS71711DCK with different Cnr.
Only when Cnr is 10nF, overshoot occurs.
What is the root cause?
Best Regards,
Hi,
For this test is the only difference the CNR?
Can you post the same images at a higher resolution? Do you also have scope traces that show both the input and output of the LDO together?
Thanks,
Nick
Hi Nick-san,
Yes this test is only changes Cnr.
When Cout is set to 22uF, overshoot occurs when Cnr is 1nF.
The waveform when Cout is 22uF is attached.
Is it due to the Quick Start function?
Best Regards,
Hi Kobayashi-san,
That is quite odd because if the quick start function was to blame I would expect the output to also overshoot with CNR = 0.1nF.
Is COUT = 22uF for all 3 of those scope shots? Are you able to measure the input current to the device during startup? I wonder if the device is entering current limit during startup and that is causing the overshoot.
Regards,
Nick
Hi Nick-san,
Yes, Cout is the same 22uF, only Cnr changes.
It is difficult to measure the input current.
I think as follows.
When Cnr is small, the output comes up with Quick Start.
When Cnr is large, the output rises sharply with Quick Start and then slowly rises with a time constant.
It looks like it's happening at a certain Cnr capacity.
Best Regards,
Hi Kobayashi-San,
Nick is currently OoO and he will get back to you once he returns on Monday.
Best,
Edgar Acosta
Hi Kobayashi-san,
I don't see a good reason that this would happen as you have described. If it overshoots with CNR = 1nF because of the quick start up, I would also expect it to overshoot with CNR = 0.1nF. The datasheet recommends to use a 0.01nF CNR or smaller, so maybe this is one of the motivations for this requirement.
I have an EVM here in the office. If you allow me 1-2 business days I can go in and try to reproduce this on my end.
Regards,
Nick
Hi Nick-san,
I checked with EVM.
It was normal with the originally implemented TPS717128DCK.
When the TPS71711DCK was installed, overshoot was confirmed. Cnr is 10nF and the NR pin is also overshooting.
However, in case of EN control, neither Vout nor NR overshoot occurs.
Please try to reproduce.
Best Regards,
Hi Kobayashi-san,
I tested with the pre-populated TPS71728DCK and I did not see any overshoot for CNR = 100pF, 1nF, 10nF. I don't have any TPS71711DCK to test with as you mentioned that those were the devices which were having issues. I've placed an order for TPS71711DCK devices; they should be here sometime next week. I'll continue testing once I receive them.
It's not surprising that the device output does not overshoot when EN is toggled after VIN is stable because the device is already in a stable biasing point, as opposed to the case where VIN = VEN, where the internal biasing is settling at the same time that the device output is coming up. What is interesting is that it seems to be a particular value for CNR that is causing this behavior.
Regards,
Nick
Nick-san,
Thanks for sharing your test result.
I have not confirmed overshoot with the TPS71728DCK, so please check with the TPS71711DCK.
Best Regards,
Hi Kobayashi-san,
Will do. The parts have shipped and are on the way to me. I should get them around the middle of this week.
Regards,
Nick
Hi Kobayashi-san,
See my results below. I tested with COUT = 22uF for all tests, and for each C_NR I tested with VIN = VEN and another test with turning on the device with EN after VIN is already stable.
I found the same behavior as you are describing. I did see a small amount of overshoot with C_NR = 100pF, but the overshoot was small. I do not have a good explanation for why this is happening with what seems is only this certain output voltage with a specific C_NR capacitor. I would recommend using C_NR = 10nF to avoid this behavior.
Regards,
Nick
Hi Nick-san,
Thanks for reporting your test results.
Although COUT=100nF is written in the report, is it a mistake of 22uF?
Best Regards,
Hi Kobayashi-san,
Oops! Yes COUT = 22uF for all tests. I updated my comment with the fix.
Regards,
Nick