FW 3.0.5 RC  

Re: FW 3.0.5 RC

Now I have reset again, i checked the difference between nozzle and probe and the probe is 0,7mm higher then the nozzle.

After I reset I printed V2 calibration and started the batman gcode to further adjust the nozzle. After that I started the whistle gcode just to see it it started, thats the one that failed for me earlier, and it started so I stopped the print.

No I ran the x/y calibration and got a different message. Saying something like "All OK x/y is slight misaligned" I different message then before. It didn't mention anything about it being corrected automatically.

So I started the whistle again and after the bed leveling it just stops and I get the message again "Bed level failed, sensor triggered to high waiting for reset" and nothing happens.

What may be causing this, I'm starting to get a bit frustrerad.

Thanks for the help.
Copied from the first post:

NOTE
Make sure your start g-code has this sequence
G28 W ; home all without mesh bed level
G80 ; mesh bed leveling
If it normally runs the mesh bed levelling twice or more, it might cause error message to show up.

Re: FW 3.0.5 RC

Now I have reset again, i checked the difference between nozzle and probe and the probe is 0,7mm higher then the nozzle.

After I reset I printed V2 calibration and started the batman gcode to further adjust the nozzle. After that I started the whistle gcode just to see it it started, thats the one that failed for me earlier, and it started so I stopped the print.

No I ran the x/y calibration and got a different message. Saying something like "All OK x/y is slight misaligned" I different message then before. It didn't mention anything about it being corrected automatically.

So I started the whistle again and after the bed leveling it just stops and I get the message again "Bed level failed, sensor triggered to high waiting for reset" and nothing happens.

What may be causing this, I'm starting to get a bit frustrerad.

Thanks for the help.
Copied from the first post:

NOTE
Make sure your start g-code has this sequence
G28 W ; home all without mesh bed level
G80 ; mesh bed leveling
If it normally runs the mesh bed levelling twice or more, it might cause error message to show up.

Yes i saw that, so the gcodes supplied on the SD card won't work with this firmware after the calibration is done?

Re: FW 3.0.5 RC

I ran one of the files I have sliced, which I know have the correct start code and everything worked just fine.

Thanks for the help. I assumed the stock gcodes where a safe bet to test with. My bad.

Keep up the good work

Re: FW 3.0.5 RC

I ran one of the files I have sliced, which I know have the correct start code and everything worked just fine.

Thanks for the help. I assumed the stock gcodes where a safe bet to test with. My bad.

Keep up the good work

We don't know why it crashes when second mesh bed levelling is ran before print. The printer now halts with error message for example when printer fails to detect one point to prevent damage.

Re: FW 3.0.5 RC

I had this same error, but upon reading this topic i understand why, since it did bedleveling and then again.
Will you provide new sliced models for new users to download ? Since now all the ones on the SD card fail.

Also the time it took to print Adalinda is stated as 6 hrs, but in fact it took 7hrs 15 minutes.
I started the print 8PM (20.00hrs) expecting it to finish by 2.00AM, it was 3:18AM hahaha.
But the print was awesome! Thanks for all the improvements in the firmware.

I was so proud when it said my X/Y was perpendicular, as i had some issues aligning it properly.
One of my rods is sticking out way more (like 2mm) then the other to get it straigthened, but probably is ok now, since i got this message ?

Last thing i noticed. I downloaded the NEW FFF settings for Simplify3D (today) the file: Original Prusa i3 MK2 RC2.fff
But it still had the Merge all outlines into a single solid model option CHECKED ?
So something is wrong there maybe in the file ? I used Import FFF and then took the unzipped file from the ZIP file.
Just so you know!

But the printer is the best i ever owned (next to the Prusa Mendel, the Kossel Mini and now the Mk2!)
Amazing detail. Everyone is drooling over the Adalinda dragon 😉

Re: FW 3.0.5 RC

So after doing several successful prints with the 3.0.5 firmware, out of nowhere on my next print, I started getting the "Bed leveling failed. Sensor didn't trigger. Debris on nozzle? Waiting for reset." error message. I tried doing the print several more times but to no avail. Then, I tried doing a calibrate z and that failed as well. Next, I power cycled the printer and did the calibrate z again and it did work... but only once. I've tried doing calibrate z other times and it fails and it fails on different points, too. So far, the points that have failed are front right, middle right, and back right. All these rights have me curious... hmm. But this has never happened to me on other firmware versions. I also tried factory resetting, which didn't work either. I think I might revert to the previous version to see what happens, even though I'll lose the perfect X/Y calibration 🙁 .

1 Hour Later:
I reverted to 3.0.3 and everything is right in the world, lol. No issues at all, but my X/Y is off a bit again, haha. Not sure what's up with the firmware though. It's a head scratcher.

Also, this is the start gcode of the print that failed but finally succeeded on 3.0.3 if you're curious:
/*
; generated by Slic3r 1.30.0.25-prusa3d-release_candidate_1_3-win64 on 2016-07-16 at 22:53:30

; external perimeters extrusion width = 0.45mm
; perimeters extrusion width = 0.45mm
; infill extrusion width = 0.45mm
; solid infill extrusion width = 0.45mm
; top infill extrusion width = 0.45mm
; support material extrusion width = 0.35mm

M107
M83 ; extruder relative mode
M104 S180 ; set extruder temp
M140 S80 ; set bed temp
M190 S80 ; wait for bed temp
M109 S180 ; wait for extruder temp
G28 W ; home all without mesh bed level
G80 ; mesh bed leveling
M104 S250 ; set extruder temp
G1 Y-3.0 F1000.0 ; go outside print area
G1 X60.0 E6.0 F1000.0 ; intro line
G1 X110.0 E7.0 F1000.0 ; intro line
G21 ; set units to millimeters
G90 ; use absolute coordinates
M83 ; use relative distances for extrusion
G1 E-0.80000 F2100.00000
G1 Z0.200 F7800.000
G1 X112.827 Y93.036 F7800.000
G1 E0.80000 F2100.00000
M204 S1000
G1 F1800
G1 X113.549 Y92.350 E0.03710
G1 X114.513 Y91.558 E0.04645
G1 X116.185 Y90.627 E0.07127
G1 X116.623 Y90.444 E0.01766
G1 X117.413 Y90.159 E0.03125
G1 X118.599 Y89.875 E0.04543
G1 X119.244 Y89.774 E0.02429
G1 X120.416 Y89.692 E0.04376
G1 X129.690 Y89.689 E0.34532
G1 X130.093 Y89.701 E0.01500
G1 X131.590 Y89.881 E0.05615
G1 X131.960 Y89.959 E0.01407
G1 X133.746 Y90.549 E0.07002
*/

James

Re: FW 3.0.5 RC - FAULT REPORT

OK, it's not the biggest fault, but it is an annoyance...

Since V3, the LCD display has shown "USB Printing" on the 4th line when printing via OctoPrint/Pronterface etc. This overwrites the printing status contained within the GCode or generated via OctoPrint.

It is fairly obvious to me that I am printing via USB, but occasionally I do want to see the status of the print; I don't believe that the currently displayed text is at all helpful.

Incidentally, I did also report this when it first appeared.

Peter

Re: FW 3.0.5 RC - FAULT REPORT

OK, it's not the biggest fault, but it is an annoyance...

Since V3, the LCD display has shown "USB Printing" on the 4th line when printing via OctoPrint/Pronterface etc. This overwrites the printing status contained within the GCode or generated via OctoPrint.

It is fairly obvious to me that I am printing via USB, but occasionally I do want to see the status of the print; I don't believe that the currently displayed text is at all helpful.

Incidentally, I did also report this when it first appeared.

Also a "Factory Reset" does not reset the "Live Adjust Z value".

Peter

Re: FW 3.0.5 RC

I'm a bit confused.

I tried multiple times to go to 3.0.5 RC, but the result is always the same, the process is stuck with the LCD screen showing 4 lines of white squares, or some errors as mentioned above in this thread. Going back to 3.0.3 is pretty quick.

I downloaded several times the file on the initial message of this thread to check if the file is corrupted, but the zip file have no error and the content is the same on each attempt. I tried to connect to various USB connectors, USB2 or USB3, same result. Did I miss something ?

Re: FW 3.0.5 RC

I'm a bit confused.

I tried multiple times to go to 3.0.5 RC, but the result is always the same, the process is stuck with the LCD screen showing 4 lines of white squares, or some errors as mentioned above in this thread. Going back to 3.0.3 is pretty quick.

I downloaded several times the file on the initial message of this thread to check if the file is corrupted, but the zip file have no error and the content is the same on each attempt. I tried to connect to various USB connectors, USB2 or USB3, same result. Did I miss something ?

While I updated it took a very long time and then I got some avrdude.exe error (or something similar) and I just had 4 lines of squares on my lcd. I pressed update again in the tool and this time it was successful and no problems since.

Re: FW 3.0.5 RC

Christophe

Worked perfectly for me; do you have the latest driver package installed?

Are you connecting to the printer via a USB hub or directly to the PC?

Peter

Re: FW 3.0.5 RC

Hi Peter,

Yes I have the 1.7.3 package, and I do not use USB hub.

What is strange is that going to 3.0.3 works flawless at every attempt, only the 3.0.5 RC react this way and I never ever once successfully installed it.

Re: FW 3.0.5 RC

Hello.

I am responsible for the development of the X/Y calibration in the Prusa3D firmware. I was on vacation the last week. Let me answer all the questions at once.

>> DO NOT run calibration routine when any USB program connected ( Pronterface etc )
> Does this statement from the previous version still apply to this version?

The Marlin firmware has problems when Pronterface or another monitoring application queries the 3D printer continuously while the printer firmware is running a lengthy procedure, which itself issues G-codes internally. The unprocessed G-codes issued by the monitor application fill the buffer and the G-codes issued by the 3D printer itself internally will be lost. I have fixed this issue in the 3.0.5 firmware hopefully.

> X/Y compensation if axes are not perpendicular: Would it be possible to show the compensation results/values?

We could show the compensation values, but we tried to keep the user interface simple.

Currently we recognize three skew values: perfect, slightly skewed and severly skewed.
When one let's the printer draw a line along the X axis at Y=0, the printer shall follow the silk screened line on the print bed.
If the error is less than 0.5mm over the bed width (250mm) without software correction, the printer is considered to be built precisely.
If the skew is higher than 0.5mm, but lower than 1.1mm over 250mm length (corresponds to a skew of 1/8 degree), the "slight skew" message is issued as a result of the calibration.
If the skew is higher than 1.1mm (corresponds to a skew of 1/4 degree), the "severe skew" message is issued.
Only in the first case the firmware will NOT correct the skew of the axes, because there is a chance that the builder built the printer more precise than we may measure with the P.I.N.D.A sensor. In other cases the printer firmware will correct the skew of the axes.

> Seems pretty fine! My X/Y calibration was perpendicular!
> I could also remove my added "x-stop-enhancer" now that the points are located better.

Congratulations! I am happy it helped.

> one more thing, now the bedleving fails. "Sensors triggerd to high waint for reset" how do i fix that?

> So after doing several successful prints with the 3.0.5 firmware, out of nowhere on my next print, I started getting the "Bed leveling failed. Sensor didn't trigger. Debris on nozzle? Waiting for reset." error message. I tried doing the print several more times but to no avail. Then, I tried doing a calibrate z and that failed as well. Next, I power cycled the printer and did the calibrate z again and it did work... but only once. I've tried doing calibrate z other times and it fails and it fails on different points, too. So far, the points that have failed are front right, middle right, and back right. All these rights have me curious... hmm. But this has never happened to me on other firmware versions. I also tried factory resetting, which didn't work either. I think I might revert to the previous version to see what happens, even though I'll lose the perfect X/Y calibration 🙁 .

As a last step of the X/Y calibration, the height of the 9 sensor points are measured and stored into the EEPROM. The next time when the Z calibration runs before each print, the height difference between the 1st and the other points is calculated and compared to the difference stored in the EEPROM. If the difference is higher than 0.5mm, the printer will display an error message and halt. There is a good reason for it. If the height difference is more than 0.5mm, you likely have a mechanical problem somewhere.

Let me stress this: The XY auto calibration is a procedure using the printer stepper motors and the induction sensor for measuring distances. The calibration procedure works very well, if the printer is built correctly. That means, if the printer is rigid, the pullies do not slip and the axes move freely over their full span. If that is not the case, the assumptions the calibration procedure makes do not apply, and the calibration procedure cannot work reliably.

First, is the print head clean? During the calibration procedure, the print head must NOT touch and push the print bed away. If you place a thin paper under the print head during the calibration procedure, there shall be no friction. If the print head is touching the print head, you may need to lower the induction sensor very slightly and re-run the calibration procedure. This actually happened at one of our printers we use at Prusa3D internally. We did not notice the problem until we applied the 3.0.5 firmware with the auto calibration to that printer. In case the print head touches the print bed during the calibration, it pushes the print bed during the bed leveling as well and you will get the infamous "Bed leveling failed" message with the 3.0.5 firmware. With the 3.0.3 firmware, you "just" get an uneven thickness of the 1st layer if you are lucky.

Are all the nuts of the frame tight, or does the frame wobble? If the frame wobbles, the movements of the print head may not be reproducible. If the frame nuts are not tight, the Y frame may twist by following the table surface, and if you move the printer on the table, you may need to re-run the X/Y calibration.

Is the P.I.N.D.A. sensor screwed tightly, or does it wobble? Just be careful to not break the plastic tab off while while tightening the P.I.N.D.A. nuts.

Are the screws on the belt pullies tight, or do they slip? If the pullies slip, the measurements the firmware makes by counting the steps of the motors will not be reliable.

Could the end stops be reached by the X and Y carriages? Do you hear the stepper motors skip steps, because they crash into another mechanical part of the 3D printer? This may apply to the Y axis, when the Y assembly is very skewed or shifted and the print bed is near the Y end stop.

Are the linear bearings of the print bed held tightly by the zip ties, or does the print bed wobble?

Vojtech

Re: FW 3.0.5 RC

Hi Peter,

Yes I have the 1.7.3 package, and I do not use USB hub.

What is strange is that going to 3.0.3 works flawless at every attempt, only the 3.0.5 RC react this way and I never ever once successfully installed it.

Christophe

I really don't know what is happening then. It does appear to be PC related, so it may be worthwhile removing the driver within "device manager" and re-installing it; I know that does not explain why 3.0.3 loads properly but it's the only thing I can can now think of.

Sorry 😥

Peter

Re: FW 3.0.5 RC

It seems that whenever I turn my printer off, the position that it is at gets reset to 0,0,0. So if it's all the way to the right on the Y axis, I can't make it go to the left because it thinks it is at 0. So I move it all the way to the right (as it's clacking the whole time because it can't go any further), and only then can I make it go back to the left.

Anyone else seeing this?

Re: FW 3.0.5 RC

> It seems that whenever I turn my printer off, the position that it is at gets reset to 0,0,0.

That is certainly happening and it is correct.

The printer knows its absolute position by first reaching the end stops and then counting the steps of the stepper motors. When you switch the machine on, the printer did not actuate the end stops yet, therefore it does not know its correct absolute position.

> So if it's all the way to the right on the Y axis, I can't make it go to the left because it thinks it is at 0. So I move it all the way to the right (as it's clacking the whole time because it can't go any further), and only then can I make it go back to the left.

After the machine power up and before trying to move the axes in the negative direction, you have to let the machine run to the particular end stop. Normally this procedure is executed by the printer at the beginning of a G-code, but you can do it manually by going to menu, Settings->Calibrate Z. After that, the printer will know the position of its axes and it will only allow you to move the machine axes in the machine working space, so you shall not hear the clacking sound anymore.

Vojtech

Re: FW 3.0.5 RC

Sorry, my mistake. 😳

I had thought that it saved its position on power off on the previous firmware, but now that you mention that, that doesn't make sense because it easily could be moved while the machine is off.

Probably should have learned the intricacies of the current fw before trying a RC version, I guess.

Re: FW 3.0.5 RC

> It seems that whenever I turn my printer off, the position that it is at gets reset to 0,0,0.

That is certainly happening and it is correct.

The printer knows its absolute position by first reaching the end stops and then counting the steps of the stepper motors. When you switch the machine on, the printer did not actuate the end stops yet, therefore it does not know its correct absolute position.

> So if it's all the way to the right on the Y axis, I can't make it go to the left because it thinks it is at 0. So I move it all the way to the right (as it's clacking the whole time because it can't go any further), and only then can I make it go back to the left.

After the machine power up and before trying to move the axes in the negative direction, you have to let the machine run to the particular end stop. Normally this procedure is executed by the printer at the beginning of a G-code, but you can do it manually by going to menu, Settings->Calibrate Z. After that, the printer will know the position of its axes and it will only allow you to move the machine axes in the machine working space, so you shall not hear the clacking sound anymore.

Vojtech

As the RAMBo has the sockets for MAX-end stops as well wouldn't it have been best to include them in the design to prevent wear and tear on the motors, belts and other parts? It would also speed things up then as well... As all new endstops were included with the kit so it would be easy for people to use the old one's off their MK1's and then just send out a few to those who bought it pre-built... They could be held on with zip ties like in the old days (on the MK1)... Bit hodge-podge but better it works well than looks pretty! New Kits could have new parts designed and printed for them to incorporate them, those that want to could then print them out and update their existing machines as well.

I did wonder if it was possible to add back the z-endstop to act as an emergency cut off, so if the probe fails or isn't calibrating, it stops the z-motors before the nozzle can hit the bed and just use the Z-MAX socket to connect it as it is all done programmatically...

Re: FW 3.0.5 RC

The maximum end stops do not help really. The minimum end stops are in use during the work space calibration only. Once the work space is defined by the printer model and the minimum end stops, the machine shall not reach the hard hardware limits during the normal operation.

The Atmel CPU does not have enough horse power to check the end stops during the normal operation. Checking the end stops during the normal operation could potentially introduce jitter into the motor movements, therefore the test is off in the Prusa3D firmware during printing.

A mechanical low Z end stop could be useful to avoid crashes during the Z homing if something goes wrong, but the new features in the upcomming 3.0.5 firmware shall achieve a similar effect. First, the new 3.0.5 firmware finds the induction sensor target points, then it stores a reference height over each of the induction sensor target points into non-volatile memory. If something goes wrong during the normal Z calibration, the firmware will now find out, as the new Z heights of the induction sensor over the induction target points will differ from the reference significantly.

All in all, the less sensors, the higher the reliability and the less troubles.

Vojtech

Re: FW 3.0.5 RC

Thanks for the info vojtěch I didn't know about the changes in the firmware as I'm a bit nervous of using a beta version when everything is currently running smoothly. Sounds like you've got it al covered! 😀