HAT install and Pi5 Red led

[IanRed]

So I have been using my Pi5 for a week or so awaiting the arrival of the Macarthur HAT which came the day before yesterday - thank you.
I installed the HAT according to the instructions with the GPIO extender so that the HAT is facing up and aligned with the PI5. All seemed well and very neat see photo attached.

Without doing anything else to the Mac HAT I plugged in the USB PSU to the PI5. The red light came on but I could not get the Pi to boot into Openplotter anymore.

Just in case the issue was the 12V on/off switch I disconnected the HAT from the Pi and all of the other peripherals and tried booting from a clean Rasp PI SD Card. Red light again.

It seems that the simplified installation has bricked the Pi, I have checked the voltages on the PI, I have 5V but no 3.3V. I clearly would like to avoid repeating the mistake. I am sure the GPIO was properly aligned. Full disclosure I have an SSD board underneath the Pi which has been working since original install.

A thought that occurred to me I installed a new active fan as the original one was faulty , I watched a number of videos on YT and the weight of advice seemed to support 4 pads one of which is over the PSU regulation. The new fan worked fine before the install of the HAT.    

Please may I have some advice and testing for the Macarthur HAT which was installed with the 12v to PI converter.

[Adrian]

Sorry to hear that. If the Pi doesn't boot without the HAT and even with a fresh SD card, I fear that the Pi might be dead/damaged.

A common way to damage a Pi is to accidentally short 3.3V and 5V pin on the Pi while it is connected to power. 

You could use a multimeter in continuity testing mode to check whether there's a short between pin 1 and 2 of the 2x20 header on the MacArthur HAT. Do this test with the MacArthur HAT not connected to the Pi or power! I don't expect to this to be short, as the HAT would fail production testing if that were the case. But let us know what you find.

   

[IanRed]

Sorry to hear that. If the Pi doesn't boot without the HAT and even with a fresh SD card, I fear that the Pi might be dead/damaged.

A common way to damage a Pi is to accidentally short 3.3V and 5V pin on the Pi while it is connected to power. 

You could use a multimeter in continuity testing mode to check whether there's a short between pin 1 and 2 of the 2x20 header on the MacArthur HAT. Do this test with the MacArthur HAT not connected to the Pi or power! I don't expect to this to be short, as the HAT would fail production testing if that were the case. But let us know what you find.

---

Adrian, thank you for the quick reply. The pins 1 and 2 are open circuit. I just wondered if plugging in the HAT with the buck converter connected but not powere would create any issues.

[Adrian]

There's an "ideal diode" circuit that prevents current flowing backwards from the Pi to the DC/DC converter. So no issue with powering the Pi from USB while the DC/DC module is mounted.

Did you ever get to the point of connecting the HAT to the 12V system or any of the boat systems? Or did the Pi die before that step?

[IanRed]

There's an "ideal diode" circuit that prevents current flowing backwards from the Pi to the DC/DC converter. So no issue with powering the Pi from USB while the DC/DC module is mounted.

Did you ever get to the point of connecting the HAT to the 12V system or any of the boat systems? Or did the Pi die before that step?

Adrian, 

No unfortunately when I plugged in the AC power supply the Pi came up red and never booted again

[IanRed]

Adrian, I purchased a new Pi5 and when I connected it to first my Pi4 as a sacrifice! and then mynew Pi5 I found that the Pi turned on for 3 seconds and then closed down. Using the AI tool it seems that the PI was not able to support the correct volatages. When I separated the HAT i discovered that the HAT has 12v, 5V but no 3.3 v. AI identifies this as a fault with the HAT. Could you review please?

thank you
Ian

[Adrian]

3.3V on the HAT is supplied by the Raspberry Pi, so that's expected. The one thing you could check is whether there's a short between 3.3V and GND.

Where does the 12V come from?

[IanRed]

Adrian, Thanks - the AI said otherwise so I followed its advice. There is no short between 3.3 and ground on the HAT that I can detect. The PI will boot up to open plotter on its own but when connected to the HAT it gets only 3 seconds before losing the 3.3v. Enable is 3.3v on the 12v converter and the power supply is from a bench supply which is regulated currently providing 12.4v. I have edited the MAX_CURRENT variable in config.txt. On the HAT of course I have 12v and 5v LEDs. When I connected the Pin 40 to 3.3v it did not allow the Pi to stay active as of course the 3.3v dropped. I was considering connecting the EN to pin 40 but have not done this yet.

As per the Wiki instructions, I have also edited the GPIO pins 26 and 21 to link the PI and the HAT

What I could do with is some method to chase the " I'm happy logic", I did not find anything yet on the forum - I am assuming that the Pi is unhappy with the detection of the 12v supply and therefore shuts down but it could also be detecting something different. would be nice to get some flashing LEDs or something to help. Perhaps we could tell the AI that the 3.3v is supplied by the PI not the HAT.

[Adrian]

That sounds like the power pins are not correctly configured and/or 12V SW is not connected. The Pi boots, then according to the OpenPlotter configuration checks whether external power is present (12V SW) and if not, shutting down.

Please make sure to read the documentation about power management on the MacArthur HAT, and follow it step by step. Asking AI, you never know whether information was omitted or made up.
https://macarthur-hat-documentation.read...power.html

[IanRed]

That sounds like the power pins are not correctly configured and/or 12V SW is not connected. The Pi boots, then according to the OpenPlotter configuration checks whether external power is present (12V SW) and if not, shutting down.

Please make sure to read the documentation about power management on the MacArthur HAT, and follow it step by step. Asking AI, you never know whether information was omitted or made up.
https://macarthur-hat-documentation.read...power.html

Adrian, Given a couple of days of troubleshooting I will overlook the have you plugged it in properly of your initial comment. I have been working on this as systematically as I can. Given that I have said that the HAT has 12v and 5v leds on all the time this seems to adequately dismiss your first comment.

The PI drops the power LEDs at the same time each cycle 5 secs in without the 12V 5V and CAN led dropping. Once the green LED goes out the CAN Led goes out. I have checked the logs and disabled the GPIO and checked it has been disabled.

When the HAT is powered via the 5V PSU that came with the pi and without the power board the PI stays up, for more than 15 minutes. The logs show that the PI has detected a reason to gracefully shut down.

VGCENCMD always results in an 0x0

[*]3-second cutoff on Pi 5 specifically, HAT LEDs never dip,vcgencmd get_throttled never logs undervoltage, unaffected by disabling gpio-shutdown, confirmed fixed by substituting the official Pi 5 USB-C supply while leaving everything else on the HAT identical. Is there are chance that at 3 seconds the PI is browning out on a transient that I cannot measure with  digital meter? 

[*]After a lot of troubleshooting I changed the bench supply to increase its output current and the problem went away. This is hopefully useful information for those who might face such a challenge