Directly Power 5V to USB
joris 2020-04-14 16:17:44 1595 Views2 Replies
PROBLEM:
One of my LattePanda's stopped working (LattePanda 4G/64GB). When I connect it to a power supply and start the board It does not boot. When the Lattepanda is started it start to short-circuit the power supply (it is not shorted when not started). I power the device via CN2 connector, when measuring this connector it is not shorted. When measuring the white sensor connectors I do measure a short-circuit. This makes me believe that I have most likely blown up an internal voltage regulator. I connected quite some devices via USB: IR camera, Thermal camera (via USB hub), 3D camera (Realsense D345). The lattepanda is also build into a metal housing; the bottom shield is connected (with artic silver) to the metal housing for cooling. Since the housing is closed, there it no a lot of other cooling, the board might run fairly hot.
It might be that I was drawing to much current via the USB connectors, it worked for some time. But together with the heat it might have died on me.
I bought a new lattepanda, but I want to make sure I'm not going to have the same problem again.
PROPOSED SOLUTION:
Instead of only providing power to the CN2 connector I also power the USB devices with the same 5V source directly. The intent is to bypass the voltage regulators on the lattepanda for the USB devices. I however do now know how the lattepanda reacts on getting 5V injected via the USB ports. I might try to add a Schottkydiode to the 5V going back into the lattepanda. This will also drop the lattepanda's provided 5V ensuring the current directly comes from the power supply and not the lattepanda
QUESTION:
1) Is this a workable solution?
2) how many current can be drawn from the USB ports without breaking the board
One of my LattePanda's stopped working (LattePanda 4G/64GB). When I connect it to a power supply and start the board It does not boot. When the Lattepanda is started it start to short-circuit the power supply (it is not shorted when not started). I power the device via CN2 connector, when measuring this connector it is not shorted. When measuring the white sensor connectors I do measure a short-circuit. This makes me believe that I have most likely blown up an internal voltage regulator. I connected quite some devices via USB: IR camera, Thermal camera (via USB hub), 3D camera (Realsense D345). The lattepanda is also build into a metal housing; the bottom shield is connected (with artic silver) to the metal housing for cooling. Since the housing is closed, there it no a lot of other cooling, the board might run fairly hot.
It might be that I was drawing to much current via the USB connectors, it worked for some time. But together with the heat it might have died on me.
I bought a new lattepanda, but I want to make sure I'm not going to have the same problem again.
PROPOSED SOLUTION:
Instead of only providing power to the CN2 connector I also power the USB devices with the same 5V source directly. The intent is to bypass the voltage regulators on the lattepanda for the USB devices. I however do now know how the lattepanda reacts on getting 5V injected via the USB ports. I might try to add a Schottkydiode to the 5V going back into the lattepanda. This will also drop the lattepanda's provided 5V ensuring the current directly comes from the power supply and not the lattepanda
QUESTION:
1) Is this a workable solution?
2) how many current can be drawn from the USB ports without breaking the board