MPS 8771 5V Out

DC-DC Power Converters
1

Project Overview: I am using an MP8771 to step down a 14.4V battery pack to a stable 5V output. The design was initially created using the MPS DC-DC online tool, then transferred to Altium for PCB layout. This PCB is a 2-in-1 design with:

  1. An MP8771 buck converter for 5V output.
  2. A separate buck-boost converter for a 12V line.

Issues Observed with MP8771 Circuit:

  1. Output Voltage Incorrect: The MP8771 is only outputting 2.5V instead of the expected 5V.
  2. Feedback (FB) Voltage Low: The FB pin reads 0.3V, while it should be 0.6V according to the datasheet to achieve 5V output. This lower FB voltage aligns with the lower output voltage (2.5V) according to the feedback equation, but adjusting the resistor divider to compensate did not resolve the issue.

Troubleshooting Steps Taken:

  1. Resistor Divider Check:
  • Verified and adjusted the feedback resistors to achieve the target output voltage of 5V, but even with modifications, the output remained at 2.5V. For example, changing the top feedback resistor (R1) to 40kΩ had no effect on the output voltage.
  1. Component Checks:
  • Double-checked component values and connections for all key components (feedback resistors, bootstrap capacitor, input/output capacitors, inductor).
  • Ensured that R1 and R2 values match the recommended values for a 5V output (20kΩ and 2.7kΩ respectively) but saw no change in the output.
  1. No-Load Testing:
  • Tested the circuit with no load connected to the output, but the output voltage remained at 2.5V.
  1. IC Replacement:
  • Suspected a potential issue with the IC, so replaced the MP8771, but observed the same behavior.
  1. Power Supply and Grounding:
  • Verified that the input voltage (VIN) is stable at 14.4V and that the EN pin is correctly pulled up to enable the IC.
  • Ensured proper grounding on the PCB and verified PCB layout for any potential issues with high-current paths.
  1. Over-Current Protection (OCP) Hypothesis:
  • Considered that OCP or another protection feature (like Short-Circuit Protection) might be activating, which could cause the MP8771 to enter a hiccup mode. However, the IC doesn’t show typical symptoms of hiccup behavior (no cycling on/off).

Key Observations for Forum Readers:

  • FB pin voltage is consistently at 0.3V instead of 0.6V, even after adjusting the feedback resistors.
  • Output voltage remains at 2.5V, regardless of adjustments to the feedback network.
  • No external load on the output during testing, so load-induced voltage drop is unlikely.
  • Components have been checked and are correctly rated for the current and voltage requirements of the circuit.

Questions for the Forum:

  1. What could cause the FB pin voltage to remain at 0.3V despite having the correct resistor divider values?
  2. Could there be another factor (e.g., grounding, layout, or hidden internal fault) that would prevent the output from reaching 5V?
  3. Any specific tests or measurements that could help diagnose whether protection features (like OCP) are triggering unexpectedly?

Any insights or suggestions to help troubleshoot this issue would be greatly appreciated. Thank you!

MP2
MP21053×753 162 KB

2

Do you have an EVAL board? They can really save time

1 Like
3

dumb comment retracted

4

I do not. Our team’s budget is kind of tight (college Formula EV Racing team) so we opted in not getting one. Do you suggest getting one? Do you know if anything like this has happened in other types of buck converters?

Thank you :slight_smile: !

5

After initially reading up on this issue, it is odd that changing the feedback resistors would not yield any change in the output. I would double check the connections between your feedback network and the rest of the system. Also, do you have a schematic that has any component values at all? This may help further narrow down the debugging process.

Getting an EVB is often a good baseline to see how exactly your system deviates from a working ideal model and gives you a starting point for debugging your own system upon replicating your system on the EVB layout.

6

They should comp you one IMHO, well how about you study the EVAL board data and compare.
Buck converters are a tried and true technology this should work. It might be helpful to share your component values, Is Rbst reasonable? Like 5 ohms or less

7

Sorry about that, here is a schematic with the values for the components:

image
image1395×991 63 KB

I will try to get an EVB, I agree that it will be very helpful.
Thanks for your response!

8

Yeah, I will try to convince them for the EVB. In the meantime, I will look at the data.

For RBST, I am using a 0-ohm resistor (CRG0603ZR—Manufacturer #). Here is a schematic with values. I’m sorry I didn’t have this to begin with!

image
image1035×735 102 KB

Once again, thanks for the help!

9

What is the voltage at the SS cap, could it be leaky cap that is preventing the SS from getting out of the way? What is the voltage at PG? Is the internal system “satisfied” with the 2.5V output voltage?

10

Not sure what the SS cap voltage is but FB is supposed to be 0.6 (internal), yet it was reading 0.3, so Im assuming something was not ‘satisfied’ internally. Yet this led me to think, that this was the reason why 2.5 was being read (Vout = FB(1+R1/R2)). Unfortunately, I assumed maybe I did not have a large enough load on the circuit, so I placed a load too large and blew up the IC (yes very dumb of me, but now I know not to do this again) so I can’t tell you at the moment what the SS cap is reading. But before that the 2.5V output was a consistent behavior. I am ordering a new IC and once that gets here I can get back to you.

11

Well the error amp has two inputs, and regulates to the lower of the two so if the SS cap is leaking and only rising to 0.3V then that would explain what you see.