oops, updated wifi server bin and printer is now offline
-
If you can't connect to wifi I'd definitvely first try USB to see if the Duet is responding, then transfer a new wifi server binary to the sd card and retry the flashing.
-
Thanks. That's what I was dreading. I only ever connected to the Duet Wifi board over USB once, while glued to the instructions, when I first set it up around three years ago. I never needed to do it again, so I've forgotten how that goes. Moreover, I've reinstalled Windows since then, so I'll undoubtedly have to go through the process of getting the USB drivers, com ports, all that malarkey set up properly again, then hope that once I have a terminal talking to the Duet Wifi board, I am able to figure out what I need to do. What I'm hoping is that the server itself actually works, and I just need to reinsert the instructions on connecting to my wifi network with ssid and password and the like. If it's more than that then I have some serious learning to do.
-
You can also power Duet off, take the SD card to your computer, and manipulate files that way. That does involve understanding the directory structure of the card... and that is not super well documented. At the same time, it is not too hard to figure out.
If I'm ever "painted into a corner" I mess with the card. I literally don't know if the USB port even works on any of the Duet's that I have.
-
I'd recommend moving the sd card to the computer to place a new wifi server binary on there in any case. While you're there you could create a trigger or macro to perform the wifi flashing if you really don't want to use the usb port; but a micro usb cable and terminal software isn't that much voodoo-like.
-
I'm not sure if there's a way of doing this 'blind'. Can M997 be run from config.g? You could try copying the firmware 1.17 DuetWifiServer.bin (I hope you have it, I can't seem to find a 1.17 release on github) to the SD Card sys/ folder on a PC, back up your existing config.g, and create a new one that just has:
M997 S1
Put the SD card back in the Duet, turn on, and wait at least a couple of minutes for the wifi firmware to flash. Then turn off, and on a PC replace the wifi flash config.g with your original config.g.
CAVEAT: This could be risky! I don't know if this will work, or leave your Duet in an unusable state! It would be safer to go the more traditional route and send the commands over USB using YAT: https://duet3d.dozuki.com/Wiki/Installing_and_Updating_Firmware#Section_How_to_recover_if_the_main_firmware_and_WiFi_firmware_versions_get_out_of_step_on_your_Duet_2_WiFi
Alternatively, update the main firmware to match the wifi firmware. See https://duet3d.dozuki.com/Wiki/Installing_and_Updating_Firmware#Section_Upgrading_a_Duet_2_Wifi_to_firmware_1_19_or_later_from_1_18_2_or_earlier
Don't forget to update to the compatible version of DuetWebControl too, if you do this. And check for changes you may need to make to your config.
Ian
-
Thanks all. I know I have a micro usb converter somewhere so I can do that. I'm not dead set against the terminal emulator over USB, but if it's a simple matter of making sure the right files are placed in the right directories, then the SD card method is probably going to be easiest for me to figure out.
The stupid thing is I decided to do this (again, not thinking too clearly) right in the middle of some testing I was doing to establish and set the pressure advance parameters that would work best for my printer. Now that process is interupted until I can get the printer back online again. If I can penetrate the mental fog a little I believe my logic was hey, firmware 1.17 was a long time ago, I'm messing with pressure advance, maybe I should get the latest firmware to see if it changes any of the ways I'm adjusting my printer, so I went off and braincrampedly did this wifi server update instead.
I appreciate all the suggestions. Tonight I'll wrestly my way through this under the assumption that by some means I cannot necessarily predict in advance I will eventually get this figured out and have the printer online again.
-
@droftarts said in oops, updated wifi server bin and printer is now offline:
I'm not sure if there's a way of doing this 'blind'. Can M997 be run from config.g?
Uncertain, wiki says wifi module is not enabled until after config.g; which is why I suggested a macro or trigger to be run post config.g. On the other hand one of the recent wifi configuration threads seems to suggest running M552 S0 from within config.g can work around that. Yet again its possible M997 takes control over all the relevant enable/reset signals, but a macro or trigger avoids all the ifs and buts.
USB is likely easier as you get feedback as you go along.
-
Wow, well a lot of things changed somehow.
I pulled the SD card and configured the latest versions of all the firmware files in /sys as the webpage says, created a /www directory (there wasn't one already) and put all the new web control files in there.
Had a little trouble getting Windows to recognize the Duet Wifi when I attached to it with USB, but eventually got it. Updated the firmware and the wifi firmware.
Anyhow, I can connect to it again, and it's got the 2.03 firmware.
Some things changed. At first I couldn't home the axes because I got the error message that says that insufficient axes are homed. Googled that and found the Duet page that says the M564 H0 command will allow this, so I added that to my config.g file and it rebooted.
Reconnected, and hit the "Home All" button, and both the X and Y axes moved in the opposite direction and ended up crashing into the frame! The Z axis homed correctly.
That's kind of wierd. Did people upgrading from some older firmware suddenly reboot with the new firmware and have their homing macros work in exactly the opposite direction?
I'm looking now to see what might be causing this, and what command I need to change to make it home correctly.
But hey, I'm not printing again yet, but at least it's back online and I can connect to it and do things through the web browser. That's progress.
-
@sethipus
Wow, I also just noticed that it's treating the X axis as the Y axis, and the Y axis as the X axis!That is, hitting a X+10mm moves the print head toward the rear of the printer (instead of to the right). Hitting Y+10mm moves the print head to the right!
-
Maybe we should have said you should read the https://github.com/dc42/RepRapFirmware/blob/dev/WHATS_NEW.md from top to bottom to see if there are any major issues that will affect your upgrade.
-
@sethipus If you scroll down the list in the link that @bearer provided, to RRF 1.19, you'll see this in the upgrade notes:
"Important! On a CoreXY machine, if upgrading from a version prior to 1.19beta9, you need to reverse the Y motor direction in the M569 command. Similarly for CoreXYU machines."
Do that and you X Y movement will be restored.
-
@deckingman said in oops, updated wifi server bin and printer is now offline:
@sethipus If you scroll down the list in the link that @bearer provided, to RRF 1.19, you'll see this in the upgrade notes:
"Important! On a CoreXY machine, if upgrading from a version prior to 1.19beta9, you need to reverse the Y motor direction in the M569 command. Similarly for CoreXYU machines."
Do that and you X Y movement will be restored.
Thanks! I suppose I probably should have read all of the release notes of all the versions of the firmware I just skipped over.
Anyhow, that did it! Btw @deckingman I watched your video about your pressure advance and whatnot, and I was really astounded by your printer. That's just awesome. I just ran that pressure advance testing python gcode generator last night and got some funky results (massively chunky sections where the print speed was 5mm/s, and I mean massive, like 3 or 4 times as wide as when it was at 100mm/s). I'll post about it in that other thread.
I gotta run a print now, just to make sure everything else is working right again. Guys, I appreciate the help. Thanks for helping me get my D-Bot/Duet Wifi back online and apparently functioning properly again after my update screwup from last night.