[Help Needed] DC Noise Affecting Arduino – SPWM Inverter


Author Message
KeepIS

Guru

Joined: 13/10/2014
Location: Australia
Posts: 2134
Posted: 06:11am 27 Mar 2025      

Nice to see another new forum member in the Inverter area.

Just took a break from completely rebuilding my Solar charger system over the past 5 days, whew!

There are some posts here about Noise and the Arduino 328P, but obviously it would be applicable to almost any Micro including the EG8010, which I have also used.

The Controller board designed by Wiseguy, which is the board I'm using, was designed with a ground plane. In my previous two build I carried out extensive testing with respect to Inverter power stage induced noise, which can sometimes be increased or even overridden by Solar charger noise on the main DC bus feeds - especially at high solar input power.

As you noted, the DC bus, both positive and negative carry very high Inverter currents and the resulting magnetic fields can play havoc with a Micro controller board.

I had a quick glance at the DSO captures you posted and did not notice anything unusual with respect to Inverter noise.

I found an Aluminum ground plate slightly larger then the Controller board, and mounded about 10mm below the PCB made a very noticeable improvement to Inverter noise injection, however it must be connected to the Common Main Bus Negative input on the inverter with a minimum of 2 gauge, preferable 0 Gauge cable, and no more than 200mm in length, no other connection with the exception of the PCB ground plane (if available) via a Cap, is allowed to touch the Aluminum ground plate, including metal work in the inverter.

In the latest build the PCB, the combination of small Arduino Nano board mounted on the Controller board, is placed on a ground plate, even though it's a good distance from the Power BUS and Dual Power stages, as noneyabussiness mentioned, cable layout is important to the extent of not running any cables to the Controller parallel to the Main DC BUS unless there is substantial seperation, separate earths back to the main input BUS point is a good idea.

I also use isolated power supplies for extra circuitry and driving LEDS and Meters, LCD etc, Wiseguys power board uses fully isolated SUB supplies on the Power board.  

The result is as follows using the Nano Controller:

The serial USB COMS is 100% stable, there is no interference carried across to the USB port of a PC or Laptop that is connected to the Nano Controller, even by 10 meters of USB repeater cable.

I assume you are referring to a different screen than the LCD screen I use in my Inverter, this is a separate Nano board driving a 4 LINE LCD, obviously there are zero problems with this approach.

The Nano has proven to be 100% reliable in my inverter, running under extreme current and power levels without a single glitch.