CarloK
-
Posts
558 -
Joined
-
Last visited
-
Days Won
22
Content Type
Forums
Events
3D Prints
Posts posted by CarloK
-
-
The camera API is hardly used by anyone and that's why we only received the report of it being broken a few weeks ago. It is broken since v5.1.
As tinkergnome says, the API redirects to the mentioned camera stream link and I think Cura uses that directly.
-
There is a new testing release which is identical to the v5.2.11 release but has extra logging enabled for the WiFi related problems.
I discovered the WiFi module sometimes short circuits and the new log messages in this version can identify that problem.
@WesleySAV17 and @cabriobob could you try this on your printer and email me the log files again?
The image can be downloaded from here and is applicable to both the UM3 and S5.
This release is to be considered a testing version, i.e. only use it for non-critical applications as this release didn't undergo the normal quality check procedures.
-
I feel sad to hear that the attempted fix didn't work. The change did sound like a promising hypothesis.
Thanks for sharing the log files. The good news is we can exclude several causes now. The bad news is that we'll have to dig deeper and probably will come with another firmware with new logging messages to discover where it goes wrong.
It seems to go wrong at system startup where the WiFi hardware does not make itself known on the USB bus.
Fixing this WiFi connection problem has our priority and we are working hard on it.
-
@cabriobob and @WesleySAV17 see your private inbox for a new firmware to test.
Other people who are willing to help us fix this problem can send me a private message and I will mail a link to a new firmware version that possibly fixes the problem and also logs more information so we are better able to find the culprit if this latest change didn't fix the problem.
-
On 5/15/2019 at 8:46 PM, Lariniel said:
Excuse me sir, but i have installed the update and my screen keeps rotating and doing weird things. And still can't print anything, when the screen rotates the printer stops completely.
This is not the kind of rotated screen problem we have seen before.
1) When does this misformed screen show up? Is it straight from booting or later?
2) Can you send me the log files from your printer? Send here
-
The UM3 software is not available on github, so I have no clue to what you are actually asking.
Did you mean a UM2?
-
An intermediate update:
There are multiple problems with the WiFi that all result in people reporting the WiFi as failing, but have different underlying causes.
For the moment we are focusing on the most pressing issue where the WiFi can not be enabled at all, in the system log this manifests itself by a message "Cannot power cycle wifi". This was present in both logs sent to me.
We verified over a 100 printers (S5 and UM3) in the Ultimaker office but none showed the error which makes it difficult to fix this problem.
One clue we have is from a tester who reported that the bug is difficult to reproduce, but once he had it on a printer and then it remained for at least 5 reboots. We want to remove a software component that tries to restore the WiFi power state after a reboot, we don't use this feature but it might interfere with our operations (and less software is better). It just turns out that this software component is integrated in Linux and difficult to disable.
-
1
-
-
5 hours ago, cabriobob said:
It seems that our S5 has the same problems as WesleySAV17.
I'll send you our log files.
Thanks for sharing the log files. Yes, it's the same problem as WesleySAV17. We are going to look into this and report back next week.
-
19 hours ago, cmeier38 said:
I have no ethernet or wifi connection sindce 5.2.8.
I was happy that a new update 5.2.11 arrived. Tried to udate via USB. Then the message "The firmware has failed to verify, Please retry installation" popped up.
The releases 4.3.96 & 5.1.93 & 5.2.8 do reject firmware packages where the release date is too far off. Sounds like a great safety feature, but since your printer has no network connection it starts with the default Linux/Unix year set to 1970 and will reject all new firmware from 2019. This safety check has been removed again in the releases 4.3.97 & 5.1.94 & 5.2.11
There are the following solutions for you now being stuck on this firmware version:
1) Install a temporary ethernet connection and in a few minutes the time will be synchronized. This works even in most organizations with a firewall. Then install the firmware either over this network connection or from USB.
Create the ethernet connection by either:
a) For the time of the update, move your printer to a location where there is an ethernet cable.
b) Or have an ethernet cable across the room to an ethernet wall outlet. Looks ugly, but is only for 10 minutes. Ethernet cables do work up to a length of 100m.
c) Or connect a WiFi access point as a bridge to your WiFi and the ethernet of the printer.
2) Connect Cura to your printer.
Get a pc with Cura and connect this with an ethernet cable to your printer. You can use any ethernet cable (straight or crossed both work). Once Cura recognizes your printer it will synchronize the time and the USB update can be started.
3) For the UM3 you can install a recovery image from SD card. For the S5 this is not supported by Ultimaker since there is a lethal high voltage inside the machine. For the S5 you can ask your reseller to do this.
-
1
-
-
6 hours ago, WesleySAV17 said:
I email you the log file
Thanks for sending the log file.
From a quick look it seems like there is a problem in discovering the WiFi hardware module. This WiFi module is connected to the USB bus and here something goes wrong. We will look into this further and let you know in a couple of days.
-
There was a bug where the XY calibration pattern would apply the last set offsets to the patterns where it should have reset them to zero during the pattern printing. This was fixed in the v5.2.11 release of today.
-
1
-
-
There are at least two known problems with the filament retraction at the end of a print. These problems have been fixed in a beta-release but did not undergo the full test procedure.
I'm sad to say that the UM2 development is in an 'end of life' state which means only huge bugs will be fixed. The mentioned beta-release is there for almost a year but gets no priority.
I'm looking for people to give this beta-release a try. So far there has been little response, but the reponse I did get was positive. With more positive results from beta-testers this release can be made into an official release.
The UM2+ firmware can be found here: https://github.com/Ultimaker/UM2.1-Firmware/tree/UM2.1_JarJar/releases/beta
And the release notes: https://github.com/Ultimaker/UM2.1-Firmware/commit/f6e69344c00d7f300dace730990652ba614a2105
Please give this a try on one of your printers if your model is a UM2+. You can upload the firmware through Cura. Reverting to the original firmware is always possible, the original firmware is provided with Cura.
Report your test results back here. Without positive reports I can't get the momentum for releasing this update.
-
1
-
-
It is possible you received a PVA spool where the NFC tag is damaged, or not programmed at all.
In the top right corner there is a 'X' mark where you can skip the Welcome procedure and then manually do the steps yourself.
-
In 5.2.11 we fixed a WiFi problem where two modules could be loaded in the wrong sequence and hence failing, a software problem called a 'race condition'. Those problems are difficult to tackle since a lot of external factors make the code fail or pass.
So, when you people report the WiFi still fails I'd like to get more details since all our internal tests succeeded and 'it fails' isn't enough info to get us searching in the right direction.
Could you please send me your log files to: https://community.ultimaker.com/messenger/compose/?to=347407
-
An easy work around for the XY-calibration starting to print without filament is to manually push the filament forward at the start of the print procedure. Annoying, yes, but luckily a procedure that has to be executed seldomly.
-
Glad to hear my tips helped you find the problem.
Kind regards, Carlo
-
It sounds like a coincidence for the buildplate temperature sensor having problems, especially the ER14 is an indication for a bad electronic connection. Check for a loose wire near the glass plate, loose connector on the other cable end, or a broken wire inside the cable.
An extra check you can do is to go to the 'temperature' web page on your printer. For example:
http://10.183.0.164/info/temperature.html
Here you see a real time graph of all temperature sensors. Do these values make sense? A broken wire will be reported as a temperature of 700C. Just try fiddling with the bed cable and see if readings change here.
Downgrading from 5.2.8 is a bit difficult because of the changes to the file system you will have to do the recovery procedure which involves preparing an SD-card and opening the bottom cover of your printer: https://community.ultimaker.com/topic/20024-recovering-a-bricked-um3/
-
Var/spool/cluster is reserved for print files so you can re-print objects from within Cura Connect, and also the print-queue is here when Cura Connect is functioning as a print job spooler for multiple printers.
-
1
-
-
4 hours ago, Swire said:
Is there a recovery.img of an older build or is my issue unrelated to build versions?
A link to a recovery image for the UM3 can be found in the first post in this thread. The recovery image is currently at 4.3.3 which has proved itself as a stable version for almost a year now.
Your problem is reported in less than 1% of the printers that downloaded the v5.2.8 version. We are working hard to figure out what's different in those printers. Your information that the printer works after installing the recovery image, but then fails some time later is a valuable clue into discovering the underlying problem.
Did you update the printer to a newer version after installing the recovery image from SD card? This might help us discover if the problem is related to the new v5.2.8 or is also present in the v4.3.3.
What you can try to do is to execute 'Cura Connect reset' from the Maintenance menu. Please report the results back here.
-
Thanks to all people who sent me the log files!
For the WiFi not working we found a problem at startup where two modules are loading in a wrong order. The network connection manager (ConnMan) is supposed to reset the WiFi electronics, but the software to toggle the USB power is not loaded yet. This requires a software update that we are working hard on and want to combine with the other most pressing issues.
The stalled printers is something we are still investigating. Log files for this are still welcome.
-
21 hours ago, Trueblondegod said:
- XY calibration fails with the same error. 😞 yeah unless i cancel the active leveling in settings then it starts printing the calibration grid.
That's a known bug. The abort should abort the whole XY calibration. Now it continues with a bad leveled bed and might at wrong heights.
21 hours ago, Trueblondegod said:I can manually level and it prints ok. The calibration is right on the +20/-20 limit. at either end. is that normal to be so far off?
I guess you mean the XY calbration? When there are no calibration values known it shows the value on the upper left scale, which happens to be '-20'. Run the XY calibration to set a correct value, or if you don't care a lot about absolute accuracy for the second nozzle, then set it to zero (good enough for 90% of the dual material prints). You can always calibrate again later.
-
If you can get your printer to dump the logs from the menu, then it will dump logging data from the last 5 reboots, or up to a maximum total size. Whichever comes first. So yes, even after a reboot the log dumps are valuable.
If you can't dump from the menu, then there are two other options:
- The web interface (shows a shorter list of logging data but is still helpful)
- SSH: journalctl -n 20000 --no-pager > dump.txt
-
Summarizing the info you gave in the other thread, combined with your above info:
- These problems started after updating to v5.2.8
- When you run the Bed Level sensor test from the Diagnostic menu it is reported as 'Low sensor performance
- You changed the cap board, but results remain the same.
- XY calibration fails with the same error.
I'm reasonable sure there is an electrical problem with the bed level sensor.
When you say you replaced the 'cap board', then I assume you mean the board below?
This board is rarely a problem since it contains no electrical parts. The only known problems are related to the connectors. So I'm not surprised that changing this board didn't help. The electronics for processing the data is on the pcb inside the print head. This is known to fail but is difficult to replace since you have to dismantle the whole print head. Not an easy task because of the many loose mechanical parts. Last time I did this, I lost a mechanical spring which never showed up again.
When your printer is still under warranty I suggest to contact your reseller.
Otherwise, double check the wiring from the capacitive sensor board to the printhead pcb. These wires did get reports from cut through, bad connectors, to even melting together.
But, most likely it is the capacitive sensor chip that's damaged. The printhead pcb is available as a service part for about €85.
All said, the active levelling sensor is a nice feature, but in practice with only 3 points levelling you get similar accuracy with manual levelling. On a UM3 the levelling feature is a gimmick, on the S5 with the larger build plate it is a necessity. On my UM3 I disabled the active levelling by setting the frequency to never, it will then use the manual levelling data. This gives faster print starts and I only calibrate once a month.
-
8 hours ago, Trueblondegod said:
i started another thread on ‘er45 low sensor performance’
It's good you started a new topic for this issue since I don't think it is caused by the new firmware, except perhaps that we now show an error for edge cases that were problematic but ignored silently in the previous firmware.
Let's discuss further in the other topic.
Problems with latest firmware
in Firmware
Posted
@Rhsin I'm sorry to hear about your display problems. We did receive a few more of these reports but from the total number of printers that upgraded to v5.2.11 this is only a handful. We are trying to figure out where it goes wrong and from your long search we can conclude it is a software problem, perhaps in combination with a certain hardware revision.
Did you perhaps save the log files? Otherwise, are you willing to go through the process of updating again to v5.2.11 and dump the log files to USB when the display fails? You can send the log files in a private message.
For the time being I suggest you remain at v4.3.3. Do not upgrade to 4.3.97 as that is not stable for printing but is only a stepping stone to the v5.x releases.