TLV62085EVM-764: Switching is not stable

Hello Kaji,

would you share the schematic and the layout of your application?
If you want to keep them confidential, please communicate your email and you will be contacted privately.

Regards,
Emmanuel

Hello Kaji,

I'm really sorry, I did not get that you were talking about our EVM. Your output capacitors seems to suffer of a bit derating (It could be due to their production lot, DC voltage or temperature). If you add 10uF in parallel, the situation is very likely to get better. Try this and let me know the result.

Best regards,
Emmanuel

Thank you for your reply.
I added two 10uF to EVM.
The result is changeable, but it is not stable. It is stable at 400 mA,
but it is not stable above that.
The changed point becomes stable from 1800 mA to 1 MHz in frequency.
Is your measurement the same situation?

(Added capacitor)
GRM21BR60J106KE19B
0805,X5R,10uF,6.3V

Best Regards,

Hello Kaji,

I'm working on it. I'll come back to you soon!

Regards,
Emmanuel

Did you find out why switching operation is not stable with EVM?
Or is it stable on your measurement?

Hello Kaji,

I found an EVM reporting the same problem you found out. I changed both input and output caps (C1=10uF and C2=22uF) with new ones of the same value and same 10V DC rating and now everything works good. I also removed C5 and used an external bulk capacitance. Try it from your side and let me know.

Best regards,
Emmanuel

Hi,

I confirmed the operation with the following contents.
1. Remove C5 and add bulk capacitor.
2. The output capacitor is the same as the previous measurement, adding 10 μF to C 3 and C 4. The model number of the capacitor is the same as last time.
Because it does not have 22uF / 10V capacitor.

Attach the measurement result. As shown in the attached waveform, switching was stable. However, removing C3 and C4 will not stabilize.
As a result, I will ask you again.
What is the reason for using a ceramic capacitor for C5? I think that this factor is great.
Is the output capacitor 22uF enough? For capacitors that were originally implemented, you told me that there was a derating effect before.

Best Regards,

Hello Kaji,

it seems we are getting to the point. First of all, yes that output capacitor must be at least 22uF when you have a 0.47uH inductor(as the table in the datasheet). Hence, if your output cap has a derating due to voltage, or even aging, then your get that instability.

About C5, I am still investigating. What is sure is that the effect of C5 is different from a canonical input capacitor, which is instead C1. Indeed that has been designed as a bulk capacitor in order to stabilise the EVM's input power supply.

I'll come back to you as soon as I find an explanation.

Best regards,
Emmanuel