Layer Offset after resume printing

Thoughts about Raise3D, 3D printing and making in general.
rapid3d
Posts: 4
Joined: Wed Sep 06, 2017 1:15 pm

Layer Offset after resume printing

Postby rapid3d » Wed Sep 06, 2017 1:41 pm

Hi,
I have a N2 model up and running fine for the past 3 months, Now after power failure when the printing resume there is a significant layer shift (Shift 1-2 mm leftside)and its spoils our component.
I haven't had this issue before, resume printing was running great.

Can someone please help me to find out what could have gone wrong. power failures will be common in my area for sometime and i need to get this right. Kindly advice.
Attachments
Offset.jpg
offset 2.jpg

User avatar
Brandon
Posts: 100
Joined: Fri May 05, 2017 1:22 pm

Re: Layer Offset after resume printing

Postby Brandon » Thu Sep 07, 2017 1:37 am

As I have not seen this I'll ask our engineers and get back to you on this.

User avatar
Vicky@Raise3D
Posts: 2429
Joined: Fri Mar 25, 2016 3:54 am

Re: Layer Offset after resume printing

Postby Vicky@Raise3D » Thu Sep 07, 2017 1:28 pm

I would recommend to check the endstop switch, since every time before resuming the print, the extruder will back to origin position first.
If only left side, then it should be the X one. At the left rear corner behind the sliding block and its belt.

User avatar
Brandon
Posts: 100
Joined: Fri May 05, 2017 1:22 pm

Re: Layer Offset after resume printing

Postby Brandon » Thu Sep 07, 2017 3:54 pm

Can you post a picture of your setting screen ( gear in the upper right of the touch screen, machine tab) ? Check your X & Y homing switches are either of them bent or damaged? If you know you are in a "bad power area" I'd strongly suggest a good UPS. Bad power could kill your electronics.

rapid3d
Posts: 4
Joined: Wed Sep 06, 2017 1:15 pm

Re: Layer Offset after resume printing

Postby rapid3d » Thu Sep 07, 2017 5:32 pm

Vicky@Raise3D wrote:I would recommend to check the endstop switch, since every time before resuming the print, the extruder will back to origin position first.
If only left side, then it should be the X one. At the left rear corner behind the sliding block and its belt.


Hello Vicky

Thanks for you replies, as suggested i checked the limit switch and it looks fine.
Attachments
File_000(1).jpg

rapid3d
Posts: 4
Joined: Wed Sep 06, 2017 1:15 pm

Re: Layer Offset after resume printing

Postby rapid3d » Thu Sep 07, 2017 5:50 pm

Hello Brandon & Vicky

Thanks for you replies, @ Vicky - as suggested i checked the limit switch and it looks fine.

@brandon - The settings page looks normal. But i noticed a bend in the rod (Which is near to that camera usb) photos and videos attached for you reference. During printing the corresponding coupler moves close to 2-3 mm up and down.

I dono how this bend occurred all of a sudden, because the printer stays on the same place where it was installed first. Do you think do to this the layer shift might happened.? and i printed a 300mm long flat stick like structure today and there were significant bumps observed.

yes i already planned to install an UPS, I am planning on to buy a N2 plus in near future, so UPS should be able to backup both. May i know the power consumption of the N2 & N2 plus.

If i need to replace the rod, kindly help me with the procedure. :(

This video gives you an idea of the bend.
https://drive.google.com/drive/folders/ ... sp=sharing
Attachments
File_000.jpg
IMG_20170907_225345987.jpg
File_000(2).jpg

rapid3d
Posts: 4
Joined: Wed Sep 06, 2017 1:15 pm

Re: Layer Offset after resume printing

Postby rapid3d » Thu Sep 07, 2017 5:58 pm

Component with series bumps printed today. :cry:
Attachments
File_000-001.jpg

Jetguy
Posts: 1782
Joined: Tue Mar 22, 2016 1:40 am

Re: Layer Offset after resume printing

Postby Jetguy » Thu Sep 07, 2017 6:23 pm

Firmware 1.1.1 has a 60 second timeout on steppers.
When you pause, you stop sending commands to the motion control board and marlin firmware. This stepper timeout is invoked after 30 seconds of no activity. Then position is lost and a shift can and likely does occur.

This is specific to a pause function but there are situations changing filament and other aspects where a user can invoke this timeout and just being unaware that it exists or it can happen is a problem in and of itself.
Last edited by Jetguy on Thu Sep 07, 2017 8:26 pm, edited 2 times in total.

Jetguy
Posts: 1782
Joined: Tue Mar 22, 2016 1:40 am

Re: Layer Offset after resume printing

Postby Jetguy » Thu Sep 07, 2017 6:31 pm

Second is, power loss recovery is not meant to be perfect. Understand that the fact it works at all is pure luck as there are a number of assumptions at play. These are: that the part does not warp or shift while the power and heat are off, that the homing sequence is perfectly accurate and repeatable for XY (hint, it's not), and that Z axis from friction alone does not move.

power failures will be common in my area for sometime

You should never take an electronic device and power it unprotected on an unstable source of power. If your power is that likely to fail, MINIMALLY you need to place an uninterruptible power supply (UPS) between the power and the printer. At least one user managed to blow the fuse in the power supply from constantly inrushing the power supply with repeated power loss -seen here in this thread viewtopic.php?f=2&t=518&hilit=fuse

Again, let's just cut to chase here. Anytime you resume a print some period of time after it started, there will be defects. This would be true in CNC machining as well because machine heat, expansion and contraction are enough to matter and make minor blemishes. Depending on a gimmicky feature repeatedly because of a poor power source is simply bad for the printer and the worst plan ever for printing quality prints.
Again, power loss recovery is nice to have, but it will never be be perfect. In fact, don't just take my word for it, other users say the same basic thing- power loss recovery, while a nice to have feature, cannot guarantee layer alignment or perfect print recovery after the printer has been powered off- all bets are off. viewtopic.php?f=2&t=593&p=11511&hilit=fuse#p11511 Examples in that thread was one user said the part is likely to separate from the print bed as the glass cools and the other indicated even worse problems with ABS as chamber temp dropping would cause warping and all kinds of problems.
Last edited by Jetguy on Fri Sep 08, 2017 12:18 am, edited 2 times in total.

Jetguy
Posts: 1782
Joined: Tue Mar 22, 2016 1:40 am

Re: Layer Offset after resume printing

Postby Jetguy » Thu Sep 07, 2017 7:05 pm

The recovery loss technical details.
So at power loss, the first thing that happens is 24V is dropping from the power supply. That power loss first affection the motion control board running Marlin and since the mega2560 processor and usb to serial converter are powered from the 24V source regulated down, the steppers and marlin all just basically cut out. Since the pcDuino front end LCD computer is batter backed, it remains powered and running. So a power loss event is detected as the combination of power loss, USB to serial device dropping off the USB enumeration, the lack of the last line of completed gcode being returned across the USB to serial interface. At that point, the front panel system makes the decision that this is a power loss event and saves a resume point for the gcode file, the executes an orderly shutdown of the Linux operating system and the LCD screen goes dark.

Upon power resumption, the system boots normally, the pcDuino and Marlin firmware on the motion board both start, the pcDuino checks the USB to serial interface and expects a valid "ready" signal from the marlin firmware indicating there are no hardware faults. Then as the Raise 3D graphical user interface starts, it sees this previous power loss recovery save point and prompts the user with a screen to either resume the print, restart the print from scratch (requires the user to clear the build plate), or just ignore the job and go to the normal home screen.

When you choose to resume, the front panel uses a modified combination of some scripted start.gcode and some elements from the print file. This is why the starting gcode in your printer profile has the M1001 in the starting gcode to let the system know when parsing that file where the starting gcode section ends. Again, that is the "magic" that makes or breaks this function. Again, key here is that the new starting gcode sent to the motion board on a print recovery is NOT the verbatim code at the start of your original print file. It is a specific modified print recovery script. This script homes XY only and does not touch Z axis. This is because Z axis homes to minimum and with an object on the build plate, it would crash that into the upper gantry if it tried to home. XY homing is not sub mm accuracy as it still homes relatively fast. In addition this assumption that the bed did not move, the part did not warp, the exact perfect alignment at the point of stopping is now still vaild- sorry but that is far beyond the control of Raise 3D. Great it it happens to work the blemish is not too visible, but under no circumstances should a user expect perfection. There are simply variables such as the bed position locked only by the very friction of the system that are unrealistic from user expectation to demand that is perfect each and every time.

User avatar
Vicky@Raise3D
Posts: 2429
Joined: Fri Mar 25, 2016 3:54 am

Re: Layer Offset after resume printing

Postby Vicky@Raise3D » Fri Sep 08, 2017 8:54 am

Have checked your video, the status of the axis seems not normal.
Please shoot a ticket onto our support system: help.raise3d.com/support/home and advise us where did you get the machine from.

For the power consumption of N2 and N2 Plus, the highest can reach will be about 500W.

frjez
Posts: 40
Joined: Tue Jan 24, 2017 2:55 am

Re: Layer Offset after resume printing

Postby frjez » Sun Sep 24, 2017 3:01 am

Agree with jetguy 100%. In addition to his excellent explanation... When you get a power outage the printer stops in one specific place over a part it's being printed. But since the hotend and the nozzle doesn't cool down immediately almost anytime the nozzle makes a huge blob of melted plastic in one spot. In worst scenario, the nozzle after a certain time cools down and gets literally glued to the part.

The proper way of solving power outage for Raise3D would be to include a small UPS that last just 15 seconds!!! In this time the UPS should send a signal to a control board that should move the printhead to a home position.

The bended sliding road in your video though.... This is a nice example of Chines engineering and quality control!

frjez
Posts: 40
Joined: Tue Jan 24, 2017 2:55 am

Re: Layer Offset after resume printing

Postby frjez » Sun Sep 24, 2017 3:16 am

So to make an advice. I would suggest you to buy a UPS quite a large one if your place is experiencing longer el. outages and if you want to finish a part in one go! But, as the UPS' tend to be very expensive the bigger they are the other option would be to buy a tiny UPS that last for a few minute. All UPS start an alarm when the el. goes off. When you hear the alarm manualy pause your printer and wait for remaining electricity from the batteries to drain. The Rais3D "solution" will take over. This way at least you will start from the Home position!!


Return to “General”

Who is online

Users browsing this forum: No registered users and 6 guests