M4V2 + SATA HAT won’t power on unless replugging one HDD

[pa67]

Hi,

I’m having a power issue with a NanoPi M4V2 with a SATA HAT.
My setup is:

- NanoPi M4V2
- SATA HAT
- 4 HDDs
- 12V 4A power supply
- OS: OpenMediaVault

This setup worked fine for a couple of years with no issues at all.

Suddenly, it doesn't start anymore: the power LED lights up for a second and then turns off, no boot.
I began troubleshooting and found that if I briefly disconnect one drive and then immediately reattach it, the NanoPi M4V2 powers on without issues.

I made another test: I swapped the NanoPi M4V2 with a NanoPi M4, keeping the same SATA HAT, HDDs, and power supply, with this setup, it starts every time without any issue.



Thanks

[Wasibbelf1950]

Hi,

This looks very much like a startup power/inrush current issue.

When the system powers on, all four HDDs try to spin up at the same time. The spin-up current of 3.5" drives can be quite high (often around 1.8–2.5 A per drive on the 12V rail). With four drives, the initial current spike can easily exceed what a 12V 4A power supply can deliver, even though the average consumption during normal operation is much lower.

The fact that the board boots if you briefly disconnect and reconnect one drive supports this theory. By doing that, you are effectively staggering the spin-up of the drives, which reduces the peak current at power-on.

A few additional observations that might help narrow it down:

The NanoPi M4V2 may have slightly different power management or protection behavior compared to the NanoPi M4, making it more sensitive to short startup voltage drops.

The PSU might also have degraded over time (capacitor aging), reducing its ability to handle short high current peaks.

If the SATA HAT does not implement staggered spin-up, all drives will start simultaneously.

Things I would try:

Test with a higher current PSU (for example 12V 6A–8A).

Measure the 12V rail during startup to see if there is a voltage drop.

Try booting with fewer drives connected and add them one by one.

Check connectors and cables for resistance or oxidation.

Since the exact same setup works with the NanoPi M4, my guess would be that the M4V2 power input circuit is a bit more sensitive to the startup transient.

Just my thoughts.