Duet powering and unresponsive webinterface

Hello,

So far my whole ormerod printing experience has been pretty good. I still have to mess around with Slicer settings etc but overall I'm very pleased with my printer .

Now I do have a question and also an annoying problem:

The question: Can the duet board only be powered by the USB cable and hence through a USB port? I mean that if I unplug my USB cable all the LED's on my duet board are not lit and it seems that it simply does not have power. Is it meant to work like this or is something wrong with my wiring? I remember the instructions saying something along the lines: "We will power the duet with the USB cable for now." Assuming that later on you would not need the USB cable anymore and would be able to run it of a PSU an network cable.

The problem: The web interface! It works but keeps crashing. Currently I am running the latest of dc's firmware (I believe it's version dc42i.bin that flashed on the duet) and I'm running his last version of the web interface as well. I just keeps ending up in a not responding state. Sometimes right after I launch it, sometimes after a few minutes. I am unable to upload a file to duet because it just stops responding in the middle of the upload. Any ideas why this is happening? Could this be related to the power issue? I am running the printer off a fairly old laptop and read something about some USB ports not providing adequate power to the duet. Though pronterface and arduino don't have any problems.

Cheers!

Rory

Hi Rory

You may have missed this part of the commissioning instructions: [reprappro.com]
The important part is to put a jumper on the JP9 (5V_EN) pins. With USB disconnected, when you apply mains power you don't get a light on the board any more, as the LED that did come on was for the extra fan, which is now off by default in newer firmware. However, if you have ethernet plugged in, you'll get lights on the socket, and you should get the Y endstop light too, if the bed is away from the endstop.

If you have an early kit (December 2013) with red printed plastic parts, there is a manufacturing fault that stops the 5V regulator working. See first point here: [reprappro.com]

Ian
RepRapPro tech support

As for powering the board, most people run it off the 12V supply via the on-board 5V regulator. There are a couple of jumpers on the Duet to control where it gets power from. If you want to run it off the 12V supply (as most do), you put a jumper on the "5V_EN" connector, near the bed heater ribbon cable connector. Described in the docs here.

Regarding the web ui, in the last few versions I have sometimes had the Chrome tab freeze up, while the printer keeps going as normal. Just have to close the tab and reopen it (reloading the page does not seem to work). Not sure what causes it, but it's no big deal for me. All it does is mess up the remanining time estimates a bit. Not sure of the frequency of it, but I do think it happens for most long-running prints if I leave the browser tab open.
It has happened once or twice during gcode file upload as well, and the same "restart web page" has done the trick there too. I also deleted the "half-uploaded" files before re-uploading the couple of times this has happened.
Have not tried other browsers, using latest regular Chrome version on Mac.

Not sure if this is what you are experiencing, or if it is due to "uneven" power through USB. Does the Duet crash/restart when you do this, or is it just the web UI?

(meh, this is what happens if I spend too much time writing a reply, Ian is too fast for me)

Edited 1 time(s). Last edit at 08/18/2014 04:42AM by jstck.

Quote
jstck
Regarding the web ui, in the last few versions I have sometimes had the Chrome tab freeze up, while the printer keeps going as normal. Just have to close the tab and reopen it (reloading the page does not seem to work). Not sure what causes it, but it's no big deal for me. All it does is mess up the remanining time estimates a bit. Not sure of the frequency of it, but I do think it happens for most long-running prints if I leave the browser tab open.
It has happened once or twice during gcode file upload as well, and the same "restart web page" has done the trick there too. I also deleted the "half-uploaded" files before re-uploading the couple of times this has happened.
Have not tried other browsers, using latest regular Chrome version on Mac.

John, I too have noticed that web page reload has been less reliable recently and opening a new tab works better. There were a bunch of changes between 0.78h and 0.78i that may relate to this (i.e. 0.78i may have made things worse). If you have time, please can you try both versions and see if you can detect a difference.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

It is very difficult to make any kind of scientific measurement on it. It happened for most "long running prints" on the earlier firmware I was on (0.78e I think it was, and whatever web ui was current at the time). I think it has happened once or twice since I upgraded to 0.78i a couple days ago, but I haven't been surprised by the frequency of it.

A guess of mine is that the difference between reloading the page not working, and closing + opening a new tab working, is that there are some tcp connections or something still left "open" that are broken.

Is there any useful status I can grab if it happens? Like telnetting to it and dumping something, or checking some particular javascript state in the browser?

Quote
jstck
Is there any useful status I can grab if it happens? Like telnetting to it and dumping something, or checking some particular javascript state in the browser?

Best option is to install Wireshark on your PC and record a Wireshark trace of a connection attempt. You could also connect Pronterface and look for any error messages sent to the USB.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

I'll start by rigging tcpdump (never got capturing inside wireshark to work well on mac, but tcpdump does the same thing) for next time I do a longer-running print and save it if something "breaks".

Quote
droftarts

You may have missed this part of the commissioning instructions: [reprappro.com]
The important part is to put a jumper on the JP9 (5V_EN) pins. With USB disconnected, when you apply mains power you don't get a light on the board any more, as the LED that did come on was for the extra fan, which is now off by default in newer firmware. However, if you have ethernet plugged in, you'll get lights on the socket, and you should get the Y endstop light too, if the bed is away from the endstop.

That will probably be it, no idea how I missed that. Thanks!

Quote
jstck

Regarding the web ui, in the last few versions I have sometimes had the Chrome tab freeze up, while the printer keeps going as normal. Just have to close the tab and reopen it (reloading the page does not seem to work). Not sure what causes it, but it's no big deal for me. All it does is mess up the remanining time estimates a bit. Not sure of the frequency of it, but I do think it happens for most long-running prints if I leave the browser tab open.
It has happened once or twice during gcode file upload as well, and the same "restart web page" has done the trick there too. I also deleted the "half-uploaded" files before re-uploading the couple of times this has happened.
Have not tried other browsers, using latest regular Chrome version on Mac.

Not sure if this is what you are experiencing, or if it is due to "uneven" power through USB. Does the Duet crash/restart when you do this, or is it just the web UI?

It's pretty much exactly as you describe. The duet itself does not crash, only the web interface. I have to restart chrome and it will respond again but I will indeed lose track of my printing time etc.

Quote
dc42
John, I too have noticed that web page reload has been less reliable recently and opening a new tab works better. There were a bunch of changes between 0.78h and 0.78i that may relate to this (i.e. 0.78i may have made things worse). If you have time, please can you try both versions and see if you can detect a difference.

Mine worked fine until a few days ago. I tried both versions a bit back and only the latest "i" version seemed to be working fine at that time. This was until like two days ago when it started crashing again. I did notice that Windows pushed some updates... Don't know if that could have anything to do with it. I'm on Win7 btw.

I had that problem when I had a conflict with IP number.

/Joar

I'm also on Win7, and I've only noticed the problem in the last couple of days. So maybe it does have something to do with Windows updates, or possibly Chrome updates. The symptom I notice most is that after re-flashing the firmware, the old browser session seems to be unable to reconnect. OTOH if it is already connected, then I can do as many page refreshes as I like, and it always comes up again with the red Connect button, just as it should.

---

Large delta printer [miscsolutions.wordpress.com], E3D tool changer, Robotdigg SCARA printer, Crane Quad and Ormerod

Disclosure: I design Duet electronics and work on RepRapFirmware, [duet3d.com].

http://googlechromereleases.blogspot.be/

I have no clue what it's all about but you might!

I also get fairly frequent browser crashes. One thing I have not yet tried is to disable my antivirus software and see if that has any effect - I've experienced problems before with that (I use AVG). Modern antivirus programs interact with web sessions to a greater or lesser extent.

Dave
(#106)