How to debug power supply dropping to 0V in multi-supply inverter setup?

Thread Starter

Quang1231

Joined Aug 29, 2025
24
Hi everyone,

I'm currently working on a 3-phase inverter and ran into an issue that I’m not sure how to systematically debug.

Setup:
  • Main DC power supply for the inverter stage
  • 12V adapter for gate driver supply
  • 5V from USB (laptop) for MCU
For the isolated supplies, I'm using an A1209 module.
In some parts of the design I used A1212 footprints (due to availability), but the actual module is A1209.
Also, the Zener diode used in the supply section is 3.3V.

Issue:
When all supplies are connected and powered at the same time, the main power supply drops to 0V.
If I remove any one of the supplies, the system behaves normally.

What I’m looking for:
I’d like to understand how to approach debugging this kind of issue.

  • What measurements should I take first?
  • Which parts of the system should I isolate or test step-by-step?
  • Any recommended method to identify the root cause in multi-supply systems like this?
I’d really appreciate guidance on a structured troubleshooting approach.

Thanks!
 

Attachments

wayneh

Joined Sep 9, 2010
18,085
Divide and conquer. You need to test each "module" or function by isolating it as best you can and then verifying that it's producing the output you expect from the expected input.

I can't pretend to have sorted out your diagram but something seems to be drawing too much current from DC supply, causing it to collapse under load. You may want to start by confirming that the supply is up to the load you intend.
 

schmitt trigger

Joined Jul 12, 2010
2,027
Agree with Wayneh; you are very likely overloading the supply.

This is the perfect time to start measuring and stop guessing.

Do you have a DMM which you could use to actually measure the supply’s output current loading? A device with a MAX function would be useful in capturing short term overloads.
 

MisterBill2

Joined Jan 23, 2018
27,159
Certainly Wayne has the correct description. So now you need to do as suggested and know how much current each load draws. AND you also need to know how much current your source is rated to deliver, AND how much current draw makes it shut down. THAT is because there might be an incorrect adjustment of the shut-down trippoint. THAT DOES HAPPEN.
I see a reference to "when all the supplies are connected", but I only see one supply mentioned.
Question: What is powering the "laptop", and how does it relate to the MCU??
Really, we need a more complete description of the system, such as voltages for the "main supply" and a greater description of that "three phase inverter" and a bit about the gate supply. AND something is deciding which gates get driven in what sequence. AND, has any portion of the system functioned correctly so far??
 

wayneh

Joined Sep 9, 2010
18,085
Something I haven't taken the time to sort out: Does your project involve connecting different, not-isolated power supplies? That always presents the potential problem of a ground loop. Significant current can flow between the power supplies and often something will be damaged by the unexpected current.
 

MisterBill2

Joined Jan 23, 2018
27,159
Certainly it looks like "ground" from a number of different sources are tied together, as I look at the MANY SEPARATE BLOCKS.
So it may be that there is, exactly as WayneH has mentioned, some unanticipated and unfortunate problems as a result.
In addition, what I see in the various details, are a large number of bypass capacitors connected directly across the supply voltage. The result isthat a failure of any one of those capacitors could overload the power source.

IF this 3-phase inverter is actualy assembled from many modules, testing each module separately could show where the problem is located.

AND, to the TS, please consider that following a circuit presented in the arrangement of the "FINAL.PDF" is a serious challenge to follow. Also, without a connected circuit shown, the possibility of missing some connection is much more likely.

The purpose of a system drawing is primarily to allow an understanding of the system function, both for the design verification process and for future diagnostic efforts. With this drawing representation that is not possible. Thus it may be that there is a connection error at some point in the assembly that is producing the power supply collapse.
 

MisterBill2

Joined Jan 23, 2018
27,159
If the problem is solved then it would benefit this community to know how and what the problem was. Problem solving is aone of the main purposesfor this site, as I understand it. Also sharing KNOWLEDGE AND INSIGHTS.
 
Top