Pico no longer found by Windows 8.1

I had successfully connected my Pico to the IDE many times on Windows 8.1 (even without ST's VCP drivers). I went to connect it again to push some updated code and Windows didn't recognize it. It still runs my code, but it no longer shows up in the COM list. When I connect it (even directly) in boot mode the red and green lights are lit continuously. I tried restarting the computer, I downloaded and installed ST's VCP drivers, and tested that one of my other Picos is found when connected. Has anyone had this issue before? Any ideas for next steps?

thanks

Try a different port, try rebooting computer. I've had both of those fix mysterious problems with various usb devices

... just to add, have you checked the back of the board to see if it's scratched? Some USB extension leads (Apple's ones) have a sharp spring on the shield that can scratch through the solder resist and silkscreen and cause a USB data pin to get shorted to ground - which has the effect you describe.

If that's the case there are some easy fixes - just use a different extension lead (normal ones have spring tines either side of the middle that won't short out). You can also cut trace on the back (I'll dig out the picture if you need) which is harmless, and which will make it work again with the original extension lead.

... the new boards (which should be coming into stock soon) have that trace (and micro USB outline on the rear) removed so people won't have the problem.

I checked the board, its not scratched. I rebooted a few more times and tried it in all the ports. Still doesn't see it.

under EVEN closer inspection –and different lighting– I found very tiny scratches (see attached). Also, I tried covering that side of the board with the piece of tape and reinserting it without success.

1 Attachment

Hmm, no, those scratches wouldn't be a problem.

Could you try it in a different computer to help to narrow it down? Hopefully when connected in bootloader mode in a different computer, the Red and Green LEDs should start pulsing - showing it has established a USB connection.

I tried it in a different computer and still got the steady red & green lights. I also tried a known working Pico which successfully connected.

Wow, ok. I'm not sure what to suggest then - and the 'contact' side of the board looks ok? It might be worth giving it a quick clean with an eraser.

If that doesn't sort it and you're willing to send it back, I could post you out a replacement? I'd be interested to see what's up with it.

The contact side looks fine, I tried cleaning it and still nothing. I'd be willing to send it to you. What address? Thanks Gordon!

Just a quick one - I managed to get hold of one that had a similar problem....

When the two LEDs are lit (bootloader mode) are they substantially darker than one of your other Picos in bootloader mode?

If so, it's probably because at some point the 5v pin on the Pico got shorted to ground, and it's damaged the diode that feeds power from USB. It'll still push enough power to drive the Pico, but not enough to get the Pico's voltage up enough to work on USB.

(You could check the voltage between GND and 3.3v with a voltmeter to see if that's the issue)

If that is the problem there are a few ways to get around (or fix) it:

• Always power it separately via the Bat pin (opposite side of the board to GND)
  
• Actually short out the diode (you then can't plug in a battery at the same time as USB though)
  
• Replace the diode (which actually isn't too painful)
  

The lights are bright (the one in question is on the right).
The voltages seem fine.
between GND and 3.3v it read 3.27v
between GND and VBAT it read 4.62v

1 Attachment

...like your serial #s: 33_93 and 33_39... Schnapszahl... just 3s, but with enough Schnaps 'sipped', they multiply and morphe...

Thanks for checking! Shame that's not it then - I'll send you a PM with my address...

Hi @Gordon do you know what went wrong with that Pico? It seems to have happened with another one. Is there a "fix" for it?

thanks

I just had another look at the one you sent back, and everything apart from USB seems fine on it - it works with a USB-TTL connector and the oscillator seems fine. In fact it turns out that just the pin A9 has got blown up.

To work around it, you need to connect to your Pico with a USB-TTL converter, then connect at 9600 baud. Type A9.set() then save(). Next time you plug in to USB it should work - but it does mean you can't type reset() and you won't be able to use the USB bootloader.

I've had a few returns, but yours seems to be the only one like this (the rest are just because people shorted them out and blew up the diode). The fact that it's happened again makes me think it could be some issue with how it's used.

How exactly was the Pico being powered when you were using it? My guess is it's being powered from a mains power supply and there is some kind of grounding problem so your Laptop's GND is way different to the Pico's - if you then connected the USB connector at an angle so only the 5V got attached, it could have put ~100v into the Pico's A9 pin I guess.

Maybe you could connect a volt meter (in AC mode) between the shield on a USB connector connected to your laptop and GND on the circuit with the Pico on it (without the Pico connected to USB) and see what it says? My guess is it'll be reading some quite big number.

I'll look at trying to fix this for the next batch of Espruino Picos I make (adding a resistor between A9 and USB), but as I say yours is the only one to have this problem, so I think it could be related to the way things are set up.

Hi @Gordon, the Pico has only been powered via USB. Either by my laptops (PC & Mac) directly or through extension cables, or with a 3-12v adaptor set to 4.5v. The next notch up is 6v and I didnt want to test if the USB can handle it. Do you have any ideas how I can prevent this from happening to another Pico?

Thanks

1 Attachment

Those selectable voltage power supplies are typically not suitable for anything other than weighing down a tarp or powering the crudest of consumer electronics. They are typically not regulated, so the voltage can be quite a bit higher than it's supposed to be, particularly under light loads (oh wait, the Pico is super low power...).

Yes, that'll be it... If you measure the voltage on it I guess you'll find it's quite a bit over 5v. It means the digital input that's connected to USB volts in order to measure when USB is connected gets blown up. I've added a resistor to the next set of Pico boards that get produced to protect them against that, but it'll be a while before those become available.

Honestly, I would throw away that white USB adaptor. The PSU is useful but being able to power USB devices from any voltage (including potentially reverse voltage) looks like a recipe for disaster.

If you want to power by USB, I'd just use a USB phone charger - they're probably a load cheaper than variable voltage supplies too.

If you want to use your power supply, don't connect it to USB, but instead connect it to GND and BAT pins (they're opposite each other, nearest the USB). You can then use anything up to around 15v without problems - and you can plug your computer into it while it is powered as well.

Those selectable voltage power supplies are typically not suitable for
anything other than weighing down a tarp

FUNNY!