jasondza
Posts: 15
Joined: Fri Jan 29, 2021 10:25 am

Second pico dead(ish)

Fri Feb 05, 2021 1:49 pm

First pico died here: viewtopic.php?f=144&t=301114

Second pico just seemed to die, but is possibly a zombie. Let me explain.

I've been working on my second pico for a few days, building a board with buttons and LEDs and a piezoelectric buzzer - something fun for the kids to play with. As part of that, I created a self-test routine that flashes LEDs and sounds the buzzer - something to help me know that the protoboard I'm working on is working properly.

A few minutes ago I plugged the board in and jumped into thonny, only to find that it couldn't connect. This was because the COM port was no longer showing up. I unplugged the board and plugged it back in again - same problem.

I thought the board was dead, but then I realised that the self test was running - LEDs flashing, and the buzzer very quietly playing it's mono tone.

So the code in main.py seems to be running, but the com port isn't shoewing up, and when I hold bootsel while booting I get no flash drive.

Sounds like the USB interface is MIA. I plan on attempting to access the pico using the debug interface once I solder a header on.

Any suggestions? Anyone had anything similar happen?

Jason

zmarties
Posts: 66
Joined: Fri Jan 22, 2021 6:53 pm

Re: Second pico dead(ish)

Fri Feb 05, 2021 5:11 pm

What OS are you trying to access the board from?

I had a problem on Windows 10 which was only solved by removing the USB device via settings, and then plugging the board in again so that WIndows recognized it afresh.

jasondza
Posts: 15
Joined: Fri Jan 29, 2021 10:25 am

Re: Second pico dead(ish)

Sat Feb 06, 2021 12:39 pm

emma1997 wrote:
Fri Feb 05, 2021 5:09 pm
I would suggest a new SD with standard PiOS installed and then following instructions in the Getting Started Guide. Otherwise we might assume a code bug which IME accounts for 99.9% of problems with Pi and other MCUs that have worked before. The hardware seems particularly robust considering how severely I've abused one of mine an no hint of failures or flakyness yet.
I don't understand what you mean by a new SD - the Pico has no SD card.

Also not sure what you mean by PiOS. Do you mean Raspberry Pi OS? You are aware that I posted this in the Raspberry Pi Pico forum with pico in the subject, and that the Pico has no OS, right? And no SD card.
emma1997 wrote:
Fri Feb 05, 2021 5:09 pm
Considering how many are out there w/none showing similar issues then it's starting to look like maybe OPs supplier is playing a joke. Or maybe OP is the prankster?
Or maybe the person responding to my legitimate problem has no idea what they are talking about, didn't bother to actually make sure they understood the problem before responding, and is just trying to make them,selves look informed and knowledgeable at the expense of someone else.

User avatar
nick.mccloud
Posts: 1228
Joined: Sat Feb 04, 2012 4:18 pm

Re: Second pico dead(ish)

Sat Feb 06, 2021 1:44 pm

I think Emma just had a Brain-Phut for a second - I managed to answer a Python question in C a few days ago.

I feel your frustration, I have a long standing track record of having things go south four hours before they are needed. Let's just say some client demos have literally been faked - I knew deep in my heart I could do it, I'd done it, it just stopped co-operating for me just as I needed.

Sometimes it's finger trouble, a loose connection or similar. And sometimes the hardware is from the purple patch in the manufacturing cycle - I've seen many boards that are bang on spec but as the vast majority were much forgiving, they looked 'wrong'.

I've read all the General & SDK post and many of the GitHub issues without any mention of Pico's going zombie. I'm not loaded, but I don't tend to treat £4 MCU boards with kid gloves. My desk is a nightmare of boards, comms modules, USB cables and PCB's exposing 5V and I haven't broken anything accidentally of late. I have set fire to a board over Christmas by not observing polarity so I'm still able to break things. What I'm trying to carefully say is that the Pico is a bombproof as an Arduino.

Can you rig up a PicoProbe as the debug/SWD port is plugged straight in to the RP2040 and bypasses pretty much everything, including, I believe, the boot ROM. That will give you some info for us to pick apart on the actual state of your Picos.
Pico/RP2040 ≠ Arduino
Pico = hot rod kit car, Arduino = hot rod kit car wrapped in cotton wool with buoyancy aids & parachute

jasondza
Posts: 15
Joined: Fri Jan 29, 2021 10:25 am

Re: Second pico dead(ish)

Sat Feb 06, 2021 7:45 pm

nick.mccloud wrote:
Sat Feb 06, 2021 1:44 pm
I think Emma just had a Brain-Phut for a second - I managed to answer a Python question in C a few days ago.
A Brain-Phut would be thinking I was talking about a pi zero or similar and responding as such.

Calling me a "prankster" when there is nothing to support the idea that I was being anything less than sincere is not a Brain-Phut.
nick.mccloud wrote:
Sat Feb 06, 2021 1:44 pm
Can you rig up a PicoProbe as the debug/SWD port is plugged straight in to the RP2040 and bypasses pretty much everything, including, I believe, the boot ROM. That will give you some info for us to pick apart on the actual state of your Picos.
I'll be honest, I don't know what a PicoProbe is in the context of a Pi Pico. More information would be great (a google search only shows me the GDB product).

As to connecting to the debug port, as mentioned, I plan on doing that as soon as I can. I have no more headers to attach (can't just pop out during lockdown), and the pico docs make it clear that you need a good connection, so crocodile clips are out of the question. I'm also not desperate enough to solder directly to the board. I'm probably going to make a supply run on Monday. I'll pick up the headers then and report back here.

FWIW, though, the previous Pico failed to respond at all. I'm hoping for more this time.

jamesh
Raspberry Pi Engineer & Forum Moderator
Raspberry Pi Engineer & Forum Moderator
Posts: 28657
Joined: Sat Jul 30, 2011 7:41 pm

Re: Second pico dead(ish)

Sun Feb 07, 2021 12:08 pm

jasondza wrote:
Sat Feb 06, 2021 12:39 pm
Or maybe the person responding to my legitimate problem has no idea what they are talking about, didn't bother to actually make sure they understood the problem before responding, and is just trying to make them,selves look informed and knowledgeable at the expense of someone else.
Post deleted. Please use the report button if you feel a post contravenes our rules (Basically, be nice to each other and don't be a dick!)
Principal Software Engineer at Raspberry Pi (Trading) Ltd.
Working in the Application's Team.

User avatar
nick.mccloud
Posts: 1228
Joined: Sat Feb 04, 2012 4:18 pm

Re: Second pico dead(ish)

Sun Feb 07, 2021 12:20 pm

jasondza wrote:
Sat Feb 06, 2021 7:45 pm
I'll be honest, I don't know what a PicoProbe is in the context of a Pi Pico. More information would be great (a google search only shows me the GDB product).
Appendix A of the Getting Started with Pico - I appreciate that it's mostly C stuff, but there is always benefit in glancing at the table of contents for all the documentation as it gives the best picture of the landscape.

Short croc cables would be worth trying.
Pico/RP2040 ≠ Arduino
Pico = hot rod kit car, Arduino = hot rod kit car wrapped in cotton wool with buoyancy aids & parachute

jasondza
Posts: 15
Joined: Fri Jan 29, 2021 10:25 am

Re: Second pico dead(ish)

Tue Feb 09, 2021 3:22 pm

Some feedback.

For reasons that I do not understand, the zombie pico was working on Monday morning. I plugged it in, and it worked without me having to do anything else.

The only thing that was different on Monday morning was that my Windows 10 PC had been disconnected from power for about 5 hours (we had load shedding).

It's worth noting the following:
  1. I had previously tried rebooting my machine and it made no difference.
  2. I had previously shut down my machine, and it made no difference (of course, though shut down it still had power running through it)
  3. I had previously worked with two other Picos, plugged into the same port using the same cable, and they worked.
  4. I plugged it into a second machine that also did not detect the COM port or the BOOTSEL virtual flash drive.
So it appears that one of the following brought the Pico back:
  • It likes Mondays. Tell me why the pico likes Mondays. I can see no reasons 'cause there aren't no reasons, what reasons do you need to be shown? (Sorry)
  • My main and standby PCs being without power for a couple of hours did something to the PCs (I reject this hypothesis)
  • The Pico somehow "reset" x hours after the problem. This seems the most logical to me, but it still makes little sense.
The only thing that would allow me to understand this is if the Pico was somehow electronically compromised, and in some way the EM fields being generated by the computers and other electronics was preventing it from discharging. The total blackout for a few hours would then explain why it "reset" and was working again.

Alternate theories welcome. Just be aware that I tested extremely thoroughly before concluding there was a problem with the Pico.

Sadly, the other dead pico is still dead (although it did pop up a com port for about 3 seconds when I tested it).

I do still intend on getting a picoprobe up and running.

User avatar
davidcoton
Posts: 6256
Joined: Mon Sep 01, 2014 2:37 pm
Location: Cambridge, UK
Contact: Website

Re: Second pico dead(ish)

Tue Feb 09, 2021 11:43 pm

jasondza wrote:
Tue Feb 09, 2021 3:22 pm
So it appears that one of the following brought the Pico back:
You forgot to allow for the phase of the moon.
The only thing that would allow me to understand this is if the Pico was somehow electronically compromised, and in some way the EM fields being generated by the computers and other electronics was preventing it from discharging. The total blackout for a few hours would then explain why it "reset" and was working again.
Nothing there that a tin foil hat won't cure.
Location: 345th cell on the right of the 210th row of L2 cache

jasondza
Posts: 15
Joined: Fri Jan 29, 2021 10:25 am

Re: Second pico dead(ish)

Wed Feb 10, 2021 8:24 am

davidcoton wrote:
Tue Feb 09, 2021 11:43 pm
You forgot to allow for the phase of the moon.

I did! Does anyone know where I can check the moon phases and their effect on electronic equipment?
davidcoton wrote:
Tue Feb 09, 2021 11:43 pm
Nothing there that a tin foil hat won't cure.
Well, I was planning on building a small Faraday cage to put the pico in if this happens again, but a tin foil hat would be so much easier. Do you think it would work? And how would I block the bottom of the hat, where the neck normally goes?

Seriously, though, can anyone offer an idea of what may have caused the problem?

jasondza
Posts: 15
Joined: Fri Jan 29, 2021 10:25 am

Re: Second pico dead(ish)

Wed Feb 10, 2021 12:18 pm

Only marginally related - my other Pico (the one I thought was DEAD dead) is now also working again. Except I'm pretty sure I fixed it, so not too much of a mystery there.

BenBB
Posts: 7
Joined: Fri Apr 02, 2021 9:52 am

Re: Second pico dead(ish)

Sun May 09, 2021 12:38 pm

Sunday 9th May 2021

I have a similar issue with a PICO
Was working 100% been using it to drive a 128x64 OLED display; SPI 0; 4 wires and powering it from the USB 5v VBUS

GP2; 3; 5 = SCK; MOSI; CS
GP15 for DC

(I have a reset switch wired to the PICO so no power cycle needed to reboot)

Then I plugged a Serial converter into the PICO; 4 wires GND TXD RX and 3v3 (from PICO PSU) it draws less than 1mA..100uA more likely
GP0 = TXD (to converter RXD)
GP1 = RXD (to converter TXD)

No software written yet !

Then I rebooted the PICO and SPI0 was dead; working for weeks now suddenly dead
I switched over to SPI1 and that worked; back to SPI0 and dead
All voltages perfect; 5.0V and 3.2v..

Then I rebooted the PICO and that was it... PICO was now dead
Removed the RS232 converter and did all above... no change; PICO was dead;

When you look under Windows device manager USB devices:
it says: actually 2 of these show up...

Windows has stopped this device because it has reported problems. (Code 43)
A request for the USB device descriptor failed.
Windows has stopped this device because it has reported problems. (Code 43)
A request for the USB device descriptor failed.

That tells me the PICO cannot boot into its "boot mode" for some or other reason;

- My suspicions: which I doubt :) is that static killed it.... but why the gradual death...
- Something happened with Micro Python that killed the bootloader ??
- Shared pins UARTTX and RX shared with SPI0 RX.. SPI0 CSn, this I can maybe think could kill SPI0 ????? if the RS232 converter; which 3v3 levels and NOT 5v; MAX3232 did something to it

My philosophy is always this: what changed last time before it happened...
- so my logical assumption is thus my rs232 converter killed it for some or other reason

As for now I have another PICO and will try that later this week; have to solder the header on first
I think the PICO is still VERY new and there are many bugs still in the pipes... silicone included; suppose this is V1.00000000000000

Tx Ben

pica200
Posts: 275
Joined: Tue Aug 06, 2019 10:27 am

Re: Second pico dead(ish)

Sun May 09, 2021 12:57 pm

Try unplugging all USB devices and then only plug the Pico in. Does it show up as drive? There is a known bug mentioned in the datasheet where other devices on the same hub can make USB on the Pico/RP2040 fail.

hippy
Posts: 9959
Joined: Fri Sep 09, 2011 10:34 pm
Location: UK

Re: Second pico dead(ish)

Sun May 09, 2021 1:05 pm

BenBB wrote:
Sun May 09, 2021 12:38 pm
Windows has stopped this device because it has reported problems. (Code 43)
A request for the USB device descriptor failed.
Might not be same but sounds similar to my problem whereby Raspberry Pi OS couldn't read the descriptors - viewtopic.php?f=144&t=311024

That's a PIco which hasn't had its GPIO connected to anything, is in an insulated box with just USB socket and BOOTSEL button exposed, had been working just fine until it didn't.

rupert_powell
Posts: 4
Joined: Thu Mar 23, 2017 2:26 pm

Re: Second pico dead(ish)

Fri May 14, 2021 10:52 am

I have used a Pico on this project.

https://www.youtube.com/watch?v=cGVxqfhBk5Q

All has been working fine for around 1 month of development, and then a couple of weeks of actual use.

But now the Pico died.

I will be replacing it tomorrow and the post mortem will follow.

It's a bit disappointing for its first real outing. Until now I have used Chinese-Arduino Nanos for this sort of project. I must have 20+ in the field, as well as 10 big Raspberry Pi's - all without a moment's trouble.

I hope it's something I did - but if I can't find anything it will make me a bit twitchy about using them in future.

Report to follow......

hippy
Posts: 9959
Joined: Fri Sep 09, 2011 10:34 pm
Location: UK

Re: Second pico dead(ish)

Fri May 14, 2021 1:50 pm

BenBB wrote:
Sun May 09, 2021 12:38 pm
Shared pins UARTTX and RX shared with SPI0 RX.. SPI0 CSn, this I can maybe think could kill SPI0 ?????
From trawling the RP2040 datasheet I couldn't tell whether SPIx CSn is an output or input when SPI is enabled. If an output then there's potentially an issue if the MAX3232 TX is connected to an SPIx CSn output without a current limiting resistor ...

Code: Select all

  MAX3232                               Pico
.---------.                       .----------------.
|      RX |<-------- UARTx TX <---|---> SPIx RX    |
|      TX |--------> UARTx RX --->|<--- SPIx CSn   |
`---------'                       `----------------'     
It would likely be fatal if the +/-10V TX intended to go to a PC COM port were mistakenly connected to the Pico.

dshadoff
Posts: 47
Joined: Wed Apr 28, 2021 3:12 am

Re: Second pico dead(ish)

Fri May 14, 2021 2:34 pm

For those instances where the Pico is having trouble connecting as a USB device, trying another USB cable may be helpful.
Some cables are flaky - or rather, the connection at the Micro-USB side can be inconsistent with some cheaper cables.

Separately, I have also found cheaper USB cables can act as lightning rods, causing reboots when a small amount of electrical noise happens, even at a distance. (i.e. electrical spark 1 to 2 metres away).

BenBB
Posts: 7
Joined: Fri Apr 02, 2021 9:52 am

Re: Second pico dead(ish)

Sun Jun 13, 2021 1:39 pm

Another PICO bricked
This is frustrating; the Pico`s are so unreliable\
I used thonny; dumped some code as Main.py and all was hunky dory
I use a 127x32 OLED display and SPI interface
Working for hours
I then made a change to the code and dumped it again; then suddenly the PICO did not show up as comport any more..
even updating UF2 files it can dump Micropython to the PICO as a USB drive but the its gone

:( :( :( :( :(

hippy
Posts: 9959
Joined: Fri Sep 09, 2011 10:34 pm
Location: UK

Re: Second pico dead(ish)

Sun Jun 13, 2021 1:53 pm

BenBB wrote:
Sun Jun 13, 2021 1:39 pm
This is frustrating; the Pico`s are so unreliable\
It might not be the Pico causing problems but the host OS.

The CDC drivers used by Raspberry Pi OS can break when the Pico is unplugged, the virtual serial port disappears, such as when the Pico is reset or BOOTSEL mode is entered. That can freeze a Pi completely, prevent it from being rebooted or even shutdown without a power-pull. That's not the fault of the Pico.

Have you tried rebooting you host system to see if the Pico comes back to life ?

Return to “General”