USB 3.0 and GPS Interference

[mwolter805]

Hello, thanks for the great project! Looking forward to keeping video, being able to review my trips and knowing I'll have video in case of an accident.

My setup is a 2025 M3LR Highland with TeslaUSB running on a Rock 4C+ with DietPi. The data drive is an external SSD connected via a USB 3.0 cable. Dual cable setups with one cable for the USB A 3.0 OTG to the hub and another cable from the USB C port for power to the same hub. Power pin on the USB A cable is covered with Kapton tape. Hub is the YONZEE Glovebox USB Hub. All connections to the SBC are USB 3.0 except power to the USB C port.

Anyone experiencing GPS interference issues? With the TeslaUSB SBC connected navigation GPS goes all wacky after driving for a bit. Nav shows I'm 50 to 500 ft to the right which then causes all kinds of problems with FSD. Removed the SBC and everything returns to normal. Looked it up and it appears USB 3.0 can interfere with GPS but not sure if it's the USB 3.0 signal or something coming from the Rock 4C+.

[deepcoder]

I've not worked with the Rock 4C+ in my M3 Highland LHD, however, I went thru several USB hubs in my glove box before I found ones that did not cause GPS to exhibit similar behaviors to your experience. I have a rather wonky setup up that is working well for me now, where I have the USB 3 hub in the front center console bay, with a USB 3 cable from the USB port in the glove box routing down to the center console.

Good hunting!

[marcone]

I wonder if the problem isn't CPU frequency rather than USB3. The Rock 4c+ has cores that nominally run at 1.5 GHz and 1.0 GHz, which are close to some of the GPS frequencies. USB3 has a base frequency of 2.5 GHZ IIRC, which seems more likely to interfere with wifi than with GPS.

[mwolter805]

Thanks for the help @marcone and @deepcoder.

Moved the SSD to a different USB A 3.0 extension and everything appears to be working. Drove around town yesterday and didn't experience GPS issues. However, previously the issue didn't appear until about 20 miles into the trip. Have a longer trip tomorrow and will know.

Checking CPU frequency and it appears it could overlap GPS

# cpu

 ─────────────────────────────────────────────────────
 DietPi CPU Info
 Use dietpi-config to change CPU / performance options
 ─────────────────────────────────────────────────────
 Architecture |     aarch64
 Temperature  |     28 °C / 82 °F : Who put me in the freezer!
 Governor     |     schedutil

                 Current Freq    Min Freq   Max Freq
 CPU0         |      816 MHz      408 MHz    1008 MHz
 CPU1         |      408 MHz      408 MHz    1008 MHz
 CPU2         |      408 MHz      408 MHz    1008 MHz
 CPU3         |      816 MHz      408 MHz    1008 MHz
 CPU4         |      816 MHz      408 MHz    1512 MHz
 CPU5         |      816 MHz      408 MHz    1512 MHz

If the drive tomorrow does not go well I'll lower CPU max frequency.

Thanks again!

[mwolter805]

200 mile drive yesterday and GPS worked properly. Appears to have been caused by the USB A Male to USB A Female cable.

Closing as the issue appears to have been resolved. Thanks for your help.

[marcone]

Reopening this so it's easier to find for people having similar issues

[deepcoder]

FYI, I have noticed that since release 2025.20.3 the USB port in the glovebox seems to be behaving differently. I have not had a chance to test fully, however at first pass look it appears the port is delivering less current. I know that Tesla added the ability to turn on USB ports when the car is 'off' in this or recent release. I have not played around with this as yet, so I am ASSuming my car is in the default mode where the power to the ports is off when car is off.

[THX723]

Tesla's recent change to allow charging ports to remain ON does not include the data port used for the dash cam. Read the latest config file for Keep Car Awake options.