I just received the ATXRaspi 2 with usb headers. Solder on a switch with a blue led ring like in the pics. The Raspberry Pi is a B+ with the usb current setting set to 1.2A. hanging off the back of the pi is a usb harddrive. I'm powering the rig with a 2A wallplug that supplys 5.12 volts. When inserting the ATXRaspi 2 into the mix the voltage drops from a static 5.12V to 4.97V during boot which causes the USB drive to fail on start up. I was hoping there was a modification I could make to the ATXRaspi 2 to keep the voltage from dropping. Any ideas or suggests would be greatly appreciated.
Extra info / back story: I'm building a mobile media server to serve up videos to the kids on long trips. Rev 1 of this project ran off a voltage converter in the car with the usb drive on a separate powered usb hub. Rev 2 runs off a 11000mah battery with the usb drive being powered directly from the raspberry pi thanks to the up'd current limits of the B+. It works great and streams for about 12 hours. The next step was to add a shutdown circuit. The rig pulls about 700 - 800 ma during max use. I understand spinning up the drive probably pulls more for a short time but my inline usb volt amp meter is not registering it. I tested with both the battery and wallplug. same results.
So you have a *mechanical* hard drive. That probably spikes a few amps when it starts up, something the ATXRaspi has not been designed for, and probably can't sustain for that hard drive to go through it's long starting sequence. In most such scenarios I think others have powered their poewr hungry devices not through the Pi which is not recommended anyway, but through a separate powered hub. But i understand your situation.
I think you might be able to solve the problem by adding a large electrolytic cap on the 5V input, there are 2 unpopulated headers where the 2.1barrel jack is, you could solder a cap there.
If I get some time I might try this with a USB hard drive and see if I can replicate and if I can find a solution.
Thanks for the capacitor suggestion Felix! I think it's right on the edge of working. I'll give the Cap idea a shot, but i'm afraid it's the voltage drop thats causing the failure. I did try some shorter cables. It helped a little but not enough. I haven't fully looked into to it, but I was wondering if the Mightyboost might work in this situation?
Voltage and Current at the Pi with the OS shutdown
5.16V .05A
Booted with USB drive attached (fails to mount and fully detect drive)
5.02V .60A
Unplug USB drive
5.11V .25A
You know, you might be able to solve the problem by slowing down the inrush current going into the Pi. You can do that with a ferrite bead as seen in this video: https://www.youtube.com/watch?v=vvuVvcYw6_8
To slow it even more, wrap the 5V in wire one or more times through the ferrite.
If that doesn't help it means your drive is really hungry and demans that its spike is met by the power supply.
Not sure if the MightyBoost would solve the issue since it has it's own circuitry which acts similarly to the one on the ATXRaspi output. But generally the voltage boost on MightyBoost is around 5.2V so ... it might be enough to get your drive to boot.
Solved:
I wanted to report back what ended up working.
I created my own USB connections to the battery and Raspberry Pi. The other ends I solder directly to the ATXRasPi. I used 24AWG stranded wire.
Voltage under load 5.17V
Current: 50 - 80 mA
I got a feeling all my usb cables are probably 28awg. Hope this helps someone else.
That makes a lot of sense, thank you for the update!