I've made a small power supply board to act as a mains relay control for an amp and a 5V supply for a RaspberryPi. This borrows heavily from Jos Van Eijndhoven's design that I've used successfully: RelaiXedPassive -- Documentation
http://www.vaneijndhoven.net/jos/relaixedpassive/RelaixedPassiveSchPcb-20160901.pdf
My schematic and PCB layout attached.
This uses a Mean Well IRM-05-5 1A 5V (and now a IRM-10-5 2A 5V) AC/DC module, with some filtering, to give the +5V supply, and an OMRON relay controlled via a GPIO-connected FET.
Problem: I get random reboots on the attached RaspberryPi - connected at the end of a 2m MiniDIN 6-core shielded cable.
The Pi can work fine for many hours, then just reboots. This happens whether the relay is energised or not. It seems to be more frequent when the relay is energised, however.
Sometimes, when the Pi immediately comes back on, it fails to connect to the WiFi - this is indicative of a drop in voltage, so I believe.
This behaviour does not occur using an external 2A 5V supply (with the GPIO connection removed), and is consistent across memory cards etc. So I'm happy it's not the Pi itself, nor the software.
To keep noise on the audio lines under control I had to connect the GNDD to the chassis / safety earth, so the supply is no longer floating.
Any ideas what might be causing this?
http://www.vaneijndhoven.net/jos/relaixedpassive/RelaixedPassiveSchPcb-20160901.pdf
My schematic and PCB layout attached.
This uses a Mean Well IRM-05-5 1A 5V (and now a IRM-10-5 2A 5V) AC/DC module, with some filtering, to give the +5V supply, and an OMRON relay controlled via a GPIO-connected FET.
Problem: I get random reboots on the attached RaspberryPi - connected at the end of a 2m MiniDIN 6-core shielded cable.
The Pi can work fine for many hours, then just reboots. This happens whether the relay is energised or not. It seems to be more frequent when the relay is energised, however.
Sometimes, when the Pi immediately comes back on, it fails to connect to the WiFi - this is indicative of a drop in voltage, so I believe.
This behaviour does not occur using an external 2A 5V supply (with the GPIO connection removed), and is consistent across memory cards etc. So I'm happy it's not the Pi itself, nor the software.
To keep noise on the audio lines under control I had to connect the GNDD to the chassis / safety earth, so the supply is no longer floating.
Any ideas what might be causing this?
Attachments
Last edited:
Yeah, I intend to put a scope on the 5V line and leave it to trigger on the FET GPIO pin changing from high to low, or something like that. Testing with a DMM indicated 4.95-5 so thought I was good.
I tested it using shorter, thicker cables for the connections with no joy.
Someone on the Raspberry Pi forum suggested removing the LCL filter entirely, so I'm giving that a go.
I tested it using shorter, thicker cables for the connections with no joy.
Someone on the Raspberry Pi forum suggested removing the LCL filter entirely, so I'm giving that a go.
So by blindingly following Jos's BoM I have most likely caused the problem. I guess he used a Pi filter to take remove any switching noise. And this is probably fine with the low current draw in his design.
The inductor I used has an I_sat of 1A with a R_max of 0.7ohms. So if the Pi were to suddenly run some background process, the current draw would increase, the voltage drop across the inductor would also increase and could lead to the Pi browning out at ~4.3V.
The system has been up for ~17hrs and operating as expected during that time. Pencil this down as a learning experience for next time...
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.