Hey everyone, I just installed my DIY modchip which is trace-for-trace compatible with the Rebug SwitchME M0. I have been having an odd issue ever since I installed it, and the issue persists whether I'm using Fusee Suite or my own minimalist modchip code Hekate-NIMBOOT.
Essentially, charging works, docked mode works, etc. But when I try to plug the Switch into my PC (Windows 10), it continuously disconnects and reconnects, at seemingly random times between 0 seconds and 10 seconds. When running software to make the SD show up as a USB drive, I can verify that the drive appears and disappears continuously.
Here is a pic of the install. The D+ wire is hidden under green electrical tape and the D- wire. (I was packing them in a tight space next to a grounded metal plate and poking the insulation with sharp tweezers, so it seemed a good precaution.) As you can see, the wires are short and nearly even in length. I have never had a failed injection, so I know the wires are good for data transfer...
In my injector Hekate-NIMBOOT, I even go a couple of extra steps to set USB-specific hardware registers that disconnect AND disable the USB system BEFORE going to sleep. So I am basically certain that, if it is possible for the SAMD21 to not interfere with USB while sitting "disabled and asleep" on the line, it should be happening here.
I suppose my next course of testing is to wire the USB lines to a blank PCB to see if the layout acts like an antenna or something. But I doubt it.
Has anyone with any type of unpatched v1 internal payload injector had this issue or something similar? Have you used USB with your SAMD21-based ("Trinket M0 clone") modchip without issue?
Essentially, charging works, docked mode works, etc. But when I try to plug the Switch into my PC (Windows 10), it continuously disconnects and reconnects, at seemingly random times between 0 seconds and 10 seconds. When running software to make the SD show up as a USB drive, I can verify that the drive appears and disappears continuously.
Here is a pic of the install. The D+ wire is hidden under green electrical tape and the D- wire. (I was packing them in a tight space next to a grounded metal plate and poking the insulation with sharp tweezers, so it seemed a good precaution.) As you can see, the wires are short and nearly even in length. I have never had a failed injection, so I know the wires are good for data transfer...
In my injector Hekate-NIMBOOT, I even go a couple of extra steps to set USB-specific hardware registers that disconnect AND disable the USB system BEFORE going to sleep. So I am basically certain that, if it is possible for the SAMD21 to not interfere with USB while sitting "disabled and asleep" on the line, it should be happening here.
I suppose my next course of testing is to wire the USB lines to a blank PCB to see if the layout acts like an antenna or something. But I doubt it.
Has anyone with any type of unpatched v1 internal payload injector had this issue or something similar? Have you used USB with your SAMD21-based ("Trinket M0 clone") modchip without issue?
Last edited by nimaid,
