I’m not sure. I had my laptop crash 3 times now from a USB port issue…I am 90% sure. I am not sure if it is the board, an ESP32 thing, or maybe just the esp32 I am testing with. If I pull the power from the board, my laptop locks up.
And something isn’t right. My Last GRBL test was ~80mm/s raster with laser, this is not even doing 5, marlin hits 20+. There is so much new stuff here I am sorting through a lot of info. This could be config, firmware, or hardware?
So I am a network admin. What kind of crash?
Two times it just locked up and I had to reboot by holding the power key for a long time. The third time I was able to use the keyboard to reboot, but that is when I reset the esp with the buttons instead of just pulling the power.
Some sort of spike down the USB? I have never ran an esp wired so I am not sure where to start with this one.
Shouldn’t be, I don’t think they draw anywhere near enough to worry a standard USB port, even with the extra draw from the add-on board.
I would try run it out of the FluidNC board and see if it does the same. If you’ve got a powered USB hub somewhere, that might also be a useful troubleshooting step. That’s not something I’ve run into before, but I tend to use all of mine through USB just long enough to flash them with ESPHome, then it’s all WiFi after that.
I’m thinking the power supply sending juice the wrong way down the USB. It happens every single time I unplug the power cord to the test board before the USB.
Well that can have that affect. Can you put tape on the positive of the usb?
Let me try a USB hub
Well no, even then I do not want to send one out and pop someone’s laptop.
What did Jno have to say, I wonder if he coukd see something in the schematics. Or maybe Bart?
Other options:
You could try clipping the D1 diode off the board with a pair of side cutters. That diode will stop the USB from attempting to power the board and should mean that there is no current flowing from the board back into the USB connector unless your voltage regulation is ‘way’ off. You could power the board up without the PC connected and check the voltage on the 5V rail. If it’s less than 5.4V, I wouldn’t think that was your issue…
@jono035 what, you dont like electric tape, lol
Can you measure the USB power signals on the test board with a DMM or oscilloscope?
If using a DMM, Put the - lead from the DMM a return on your laptop (say the USB outer shell, or a jakpost if you happen to have a VGA or serial port on the laptop.) Put the + lead on the USB +5V pin on the test board. Then pull the power to the test board and see what happens.
I wonder if you’re getting leakage somewhere like an ESD diode or something that’s impacting the USB port.
They’re all things that can be tried. I probably wouldn’t go down that route personally due to a general distaste for electrical tape and the potential to cause other issues like gumming up my USB ports or getting poor/intermittent contact with the signal lines. It’d definitely be a good ‘last resort’ kinda check, though. I’d probably sooner DIY an unpowered USB adapter, use a USB isolator that I have sitting around or even open the cable, spread the braid and carefully clip/insulate the 5V cable. If I did use the electrical tape, I’d definitely mark that cable as being ‘bad’ for the future, just to be sure. USB contacts are really a kinda shite design, anyway. Very little contact force, very little ‘wipe’ to them.
In order, I would:
Try the ESP32 off the board.
Check the 5V on board with the ESP32 on the board but not connected to the PC.
Try a powered USB hub if one is floating around. If not, try the electrical tape method or sacrifice a USB cable as above.
Try a different ESP32 in a different board. If that doesn’t do it, try see if the issue chases the board or ESP32. Then try to figure out what’s going on with that.
Over-current on the 5V line isn’t normally a big deal with USB ports, they tend to just trip a PTC and stop providing power. Entirely hanging a PC is a pretty bizarre failure mode in my experience.
I suspect when Ryan is pulling board (motor) power, there’s a collapse happening and that is driving the USB port ABOVE 5V, causing the issue. That’s a subtle issue if so.
Okay…good and bad news.
While I had the USB unplugged and was testing the board my laptop blue screened. Picked it up and it is burning up. So maybe i messed it up traveling. That means the test board should be okay…but my nice little laptop might get torn down tonight.
I did check the 5V rail with two DMM’s and they both show good on plug and unplug.
So now just the M4 issues.
So I have a Scope, and a USB logic analyzer. If there might be a way to see what is going on I am open to suggestions.
M3 and all the other stuff work fine. Just inline laser commands.
Yeah, that sounds more likely. I was really confused as to how you’d be crashing a laptop with USB power conditioning issues. That’s remarkably unlikely given the setup on that board.
Just so I understand- you tested the USB +5V while connecting and disconnecting board (motor) power?
With the laptop powered off? Or disconnected?
What motor power voltage are you testing with?
I wouldn’t discount that you had a repeatable issue with plugging / unplugging board power and now your laptop is “burning up”.
It did happen exactly when I unplugged though 3 times…like try three was a laptop test and it did crash. Weird.
