I have been trying to debug my TPS54331 circuit (3.7V to 3.3V) for a couple weeks now and have not come to any solutions. I am floating the EN pin and the voltage is 2.93. My BOOT to PHASE, Diode, Inductor, and Output voltage do not look like normal behavior at all. I am using a 30 ohm load also.
I am attaching the waveforms, schematic, and layout images. The inductor value is 6.8uH, I do not believe that is labeled on the schematic.
8664.3.3V Regulator TPS54331.zip
Any leads on what may be messing up are appreciated.
Thanks in advance,
Dan
With 3.7 Vin to 3.3 V output, you are probably operating in low headroom drop out. It is hard to tell from your screen shots becuse the time scale is too long. Can you try raising the input voltage to say 7 V? Does the output start to regulate? What is the load? Does it improve if you increase the load? Under certain conditions, with light load ind low input voltage, the BOOT UVLO may cause the output ripple to have 2.1 V p-p.
John Tucker
Consumer DC/DC Applications
Another issue is the layout. I recommend you follow the guidelines in the datasheet. Pay close attention to the position of the C8 input capacitor and the ground routing also the routing of the catch diode. You will neet to remove the signal rout from under the IC so that you can solder the exposed thermal pad. If you provide your design equations, I will look at your component values. One ting you could do to check your layout would be to build your circuit up on a EVM board. That EVM will have the proper layout.
Increasing the input voltage to 4 volts corrects the problem. Staying at 3.7 and increasing the load still gives an unregulated output.
However I was planning on using a single cell LIPO battery (3.7V.) With this project I am also using another boost converter to give me 5 volts. So I would like to keep the input voltage at 3.7, but if there is no other way I can make this work. Is there any other way to get out of this low headroom dropout operation?
Let me know what waveforms to send.
If that fixes the problem, then no need to send more waveforms yet. there is a work around that involves adding a resistor and diode. I'll up load the PPT that explains it tomorrow that explains it all. You may want to post in the am to remind me...
Could you upload the PPT with this work around.
Thanks
Thanks for the reminder. The document is based on TPS54160, but the concept is the same for TPS54331:
2671.TPS54160 low dropout.ppt