Interface crashes randomly during upload, system files wiped

Robathome

Well, the subject line here really sums up the problem I'm running into… It's happened three times, but not consistently. I was able to get the printer running again after the first two events, and each event occurred while uploading different files. Nonetheless, the error has manifested almost-but-not-quite-exactly the same each time. At the risk of too much detail, here is the chain of events, as best as I can describe them:

Power on
Log in to web interface (password protected)
Calibrate delta

G28, G32, G32, G29, G28

Preheat hotend (Heater 1 Active temp > 180)
Prime extruder (Machine control > Extrude 20mm@5mm/s until extrusion appears, then retract 5mm)
Upload gcode file

During upload, an upload error is shown in the progress dialogue for maybe half a second, at which point the web interface crashed and reset itself (interface greyed out, yellow "firmware reset" progress bar appeared at the bottom of the page, then the page reloaded). However, there was no login required (not a good sign), and the printer was unresponsive.

The first time this happened, the extruder stepper started oscillating ~20mm@10mm/s back and forth until I manually powered off the system (web interface crashed, no e-stop button). The web interface appeared normal, but I knew something was up when it didn't ask for a password. The system remained unresponsive until I powered off the system for several hours, at which point it started up and responded normally.
The second time this happened, the system froze completely (unresponsive, no extruder oscillation), but the web interface remained online (no password required when reloading the page). E-stop commands triggered the "system reset" progress bar dialogue, to no effect. After sending G28 commands several times with no response, I checked Settings >System Editor >homedelta.g, and was shocked to find the file was blank. I copied the contents from a backup into the blank file, and the system immediately responded to G28 commands, and for all intents and purposes, behaved normally.
The third time this happened (about an hour ago), the web interface crashed during upload, again briefly showing the ERROR indicators in the progress dialogue. However, instead of resetting itself, the following message appeared in a red message box at the bottom of the screen:

[[language]]
Message: Uncaught TypeError: Cannot read property 'axesHomed' of undefined
URL: http://192.168.2.16/js/dwc.js
Line: 1584:30
Error object: {}

Reloading the page brings up the web interface (note: no password required), but the system was again unresponsive, even after software reset (using emergency stop button in web interface). Attempts to reload the web interface (via F5) came back with "This site can't be reached" (ERR_CONNECTION_REFUSED in Google Chrome).

This condition did not change after hardware reset (pressing RESET button on the DuetWifi board), or after a hard reset (PSU turned off for 30 seconds).

I connected the DuetWifi to my PC with a USB cable, and Pronterface was able to connect to the board on the first attempt (COM4@115200,"Connecting… Printer is online").

Sending M552 via Pronterface indicated that networking was disabled

[[language]]
Network is disabled,
configured IP address: 192.168.1.10, 
actual IP address: 0.0.0.0,
HTTP port: 80

Sending M552 S1 via Pronterface started up the WiFi server without issue

[[language]]
WiFi server starting up; 
DuetWiFiServer version 1.02; 
Flash size 4194304,
free RAM 31504 bytes,
WiFi Vcc 3.35V, 
host name: duetwifi, 
reset reason: Turned on by main processor; 
WiFi server connected to access point <nope>,
IP=192.168.2.16,
signal strength=-47dBm</nope>

Reloading the web interface appeared successful but did not require a password, and now G28 commands trigger a yellow message box (warning dialogue?) with the message "Macro file homeall.g not found". This was particularly concerning because the printer is a delta, and G28 normally calls "homedelta.g".

When I went to investigate the System Directory, it was completely empty. To be more specific, the frame that normally lists all the system files (Settings >System Editor) contained only the message "The current volume is not mounted" in a slightly-grey text.

At this point, I decided that this problem was way above my head. I hope this isn't too much detail, and if there is any information missing, please do not hesitate to ask. In every other respect, the DuetWifi has exceeded all my expectations, and if I have done something to damage it that would cause this error, I would buy a replacement board without hesitation. Hopefully that is not the case!

Here are some of the details that I anticipate might be useful:

Printer model: Kossel 250
Firmware name: RepRapFirmware for Duet WiFi
Firmware electronics: Duet WiFi 1.0
Firmware version: 1.17RC3 (2016-12-21)
WiFi server version: 1.02
Web interface version: 1.15

Firmware diagnostics:

[[language]]
M122
=== Diagnostics ===
Used output buffers: 1 of 32 (4 max)
=== Platform ===
Memory usage:
Program static ram used: 20260
Dynamic ram used: 83564
Recycled dynamic ram: 2672
Current stack ram used: 2872
Maximum stack ram used: 3828
Never used ram: 20748
Last reset 00:50:36 ago, cause: reset button or watchdog
Last software reset code: 0x5002, available RAM 12972 bytes (slot 2)
Spinning module during software reset: Webserver
Error status: 0
Free file entries: 10
SD card 0 detected, interface speed: 0.2MBytes/sec
SD card longest block write time: 0.0ms
MCU temperature: min 37.9, current 41.5, max 47.0
Supply voltage: min 2.1, current 24.6, max 25.0, under voltage events: 1, over voltage events: 0
Driver 0: standstill
Driver 1: standstill
Driver 2: standstill
Driver 3: standstill
Driver 4: standstill
Current date and time: 2017-04-21 08:47:32
Slowest main loop (seconds): 0.132088; fastest: 0.000000
=== Move ===
MaxReps: 0, StepErrors: 0, MaxWait: 0ms, Underruns: 0, 0
Bed probe heights: 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
Probe change coordinates:
=== Heat ===
Bed heater = 0, chamber heater = -1
=== GCodes ===
Segments left: 0
Stack records: 0 allocated, 0 in use
Movement lock held by null
http is ready with "M122"
telnet is idle
file is idle
serial is idle
aux is idle
daemon is idle
=== Network ===
WiFiServer is running
SPI underruns 0, overruns 0
=== Webserver ===
HTTP sessions: 1 of 8

deckingman

No doubt others more knowledgeable than I will be along shortly, but what stands out to me is the supply voltage (min 2.1) so I'm wondering if you have a power supply issue?

Robathome

Huh. I hadn't noticed that, but I'm hesitant to connect the two, unless the PSU is degrading very quickly. There were about seven or eight prints between the first and second occurences, and at least fifteen between the second and third.

deckingman

Yes, if it was the power supply itself, one would expect to have problems with heaters not getting up to temperature and so forth. The fact that it prints OK would indicate that is not the case, but it might be worth checking for a loose connection while waiting for a more knowledgable person to come along.

dc42

Yes, that undervoltage event points the finger at a power issue. Is your mains supply unreliable? My best guess is that you had a power brownout and the voltage dropped enough to interrupt the write to the SD card, so that the SD card got in a mess. You might have seen the lights dim at the same time, although that is less likely if you have LED light bulbs.

Another possibility is that you have a loose connection in your power supply wiring. Check that the screws in the VIN terminal block of the Duet are tight.

For the future I may implement monitoring of the second power monitor input on the Duet WiFi, so that if the 5V power falls below 4.0V then the board is reset. and won't try to start until 4.2V has been reached.