Layers shifting with me too - please help

#1
Yes, I saw the thread posted by the other person who had problems with layers shifting. The problem is that I really don't understand the solution, other than possibly updating my firmware? According to my computer, the version of Arduino that I have installed is 1.0.5-12. That is the latest version, right?

I don't know how to check the stepper driver pentometers, and I don't know how to use the latest slic3r configuration file with slic3r RC2. What is RC2? I thought the program was just called "slic3r."

And if the problem lies with my extrusion multiplier setting, then I really don't know what to do. Gordon wrote,
" If you use the latest config file, and since you have a printer released prior to jan 30th 2014, make sure to go to "Filament Settings > Filament" and set the extrusion multiplier to 0.85"

I received my printer about a month ago (in May). According to Gordon, the new firmware uses the 1.0 multiplier. But maybe I need to change my multiplier setting?? Where is "filament settings > filament?" On the LCD screen of the printer? If it's there, I have no idea how to find it. Gordon, can you please give me the entire location of the setting; not just the last two folders?

So really, what do I do?

Here's a photo of what happened with my print. I tried to print the letter "B" in PLA. As you can see, the layers are shifted really badly:
 

Attachments

#3
Here is a 20 mm cube set to print in the center of your printbed. Hope this helps you narrow down where the problem is.

If you still see problems when printing this cube, I would imagine the issue could be stepper motor calibration. After printing the cube check if any of the motors feel either very hot or not hot at all. Unfortunately I don't know what's the best way to set the potentiometer, but hope this is a start.
 

Attachments

#4
I had the same problem too. I ended up turning down the potentiometer on the y axis to solve the problem. This procedure is described elsewhere in this forum. You will need to turn the potentiometer for the axis that is shifting (X or Y). The potentiometers are located on the Arduino board, which is located underneath the metal cover on the side of the printer. They regulate how much power is sent to the stepper motors. There is a good picture in Fig 14.1 of the GMAX assembly documentation. You will need a tiny screwdriver to fit into + shaped slot on the potentiometer. Be careful not to touch the screwdriver to anything else because you could short something.

Here's what I did to solve my problem. With a test print running (could be any print) I gently turned the Y axis pot (potentiometer for short) all the way counter clockwise. This reduced power to the stepper motor so it stopped moving. Then I slowly turned the pot clockwise until the motor started moving. Then I turned it a bit more (about 1/8 of a turn) just to be sure. My skipping dramatically reduced but I still got one or two in a 50mm high print so I turned it clockwise another 1/8 of a turn. That seems to have solved everything.

With hope, this is your problem because it is fairly easy to fix. My extruder was also making a ticking/thumping sound so I adjusted the pot for that too and the noise went away.
 
#5
Thanks to both of you!

I downloaded the cube file that matt_hova provided, and I started printing. Well, at that point, my printer literally started falling apart. The screws that hold the bed onto the rolly wheel thing popped out and fell on the table. I had to reattach them.

I got everything repaired, and I printed the cube. I think it turned out nicely:

Immediately after that, and without making any adjustments to the printer, I proceeded to print a different file. It's a half sphere, with its flat side bottom-down on the bed. Here's how the print came out:



Obviously, there's something terribly wrong. The picture I'm showing you is taken while the print is still on the bed, so that you can see its exact orientation. It's all over the place!! I "created" the file myself, if you can call it that: I went into my Tinkercad account and placed a half sphere on the grid, and then I exported it as an .stl file. From there, I brought the .stl file into Slic3r and exported it as a .gcode file. Here's the .gcode file:

So evidently the file didn't slice correctly at all! Maybe I don't have Slic3r configured correctly for my gGmax printer. I don't know how to configure it correctly. Can someone please help me with that? WHY would the file get sliced so wrongly??

Can sopmeone please download my file and print it, and see what happens on your printer?
 

Attachments

#6
I took a look at the gCode and it seems ok. The only configurations you need for slic3r are all on the gMax website's download page, once you've imported them into Slic3r and selected the particular combinations based on your nozzle size and choice of plastic, you're good to go.

How tight is the Y-belt under the bed? Don't take the bed off unless you absolutely have to, just reach push the belt down a bit and get a feel for it. According to the instruction manual the belt shouldn't be too tight, but I'd air on the tighter side rather than risk that being a possible reason for this.
 

GORDON.LAPLANTE

Administrator
Staff member
#7
@Kristina

Try printing the sample gcode file which is now available on the download site.

Also, we found that with newer versions of slic3r config files don't load correctly when loading them from the menu. Basically most of the settings don't import causing problems. The solution is fairly easy:

Download the latest config files from the download site. You will now see several files. Each file represents a different profile all of which should be installed. You have to manually drop them into the slic3r config folder here. Below is how to find the folder in Windows.

You can find that folder in Windows by clicking "Start > Run". (Note in newer versions you can find run in "Start > All Programs > Accessories > Run". Also refer to google if you need to find it.

Then paste this into the "Run" window:
C:\Users\%username%\AppData\Roaming\Slic3r

You can also paste this in windows explorer.

This should open the slic3r config file location.

Then simply copy the config files into the respective folders. Make sure to backup your old files if you wish to keep the settings since this may replace the old files.


You can also use this very useful program to view your gcode files:
http://gcode.ws/
 
#8
Gordon, you wrote, "Download the latest config files from the download site." Well, I see three config files. How do I know which one is the latest?

Once I know which one to download, I'll be able to fiollow your other instructions very easily. Thank you!
 

Attachments

#10
Thank you, Ray

I followed the instructions and I proceeded to print the half-sphere shape again. Everything seemed to be going really well until I got to almost the end of the print. The top few layers shifted toward the front of the machine (toward me), on the Y axis.

What do I do now?
 

Attachments

#11
How tight is the belt under the bed? The only thing that comes to mind is making sure that the belt is not skipping by make sure it is tight (but not too tight), but I air on the tighter side.
 
#12
raykholo said:
How tight is the belt under the bed? The only thing that comes to mind is making sure that the belt is not skipping by make sure it is tight (but not too tight), but I air on the tighter side.
Yes, I remember you mentioning that earlier. It's not really that tight, but when you're a girl, it's a lot harder to make things tight than when you have the strength of a guy. I'll tighten it as much as I can.

However, I printed another file after the half sphere. It was a little statue of a lady standing. Granted, the file is probably too intricate to be printed at only 1" tall, but the statue had two layer shifts, and BOTH of those layer shifts involved BOTH the Y axis AND the Z axis. Therefore, I'm really thinking that this is a slicing problem rather than a belt problem.

In order to get to the bottom of that theory, all I'd have to do is print the half sphere again, and see if the same layer shift occurs at the same height. If it does, then it's a slicing problem. If it doesn't, then it's probably a belt problem.

I'll be back after I print it again. Thanks!
 
#13
OK, I printed the half sphere again, and sure enough, the layer shift happened at exactly the same height as the previous print. This leads me to believe that the problem does not lie in the tightness of the belt; but rather, in the slicing.

Can someone please download and print the new half-sphere file (sliced with the correct slic3r settings) so that they can see if they encounter the same problem?

I do want to add that the layer shift in the 2nd print wasn't quite as marked as the shift in the first print. Weird, right?

And also, if I do end up having to tighten my belt, what a b**** that's going to be! The area is so inaccessible that I'm going to have to remove the entire bed :(
 

Attachments

#14
Bump...

GOOOOOOOOOORDON, are you there?

I REALLY think there's a slicing problem at hand with the software. Since my last post, I've attempted to print a few different things, and they all have shifted layers here and there.

I'm going to download a file off Thingiverse and see how it prints.
 
#15
Skip to below the dashed out part

--
Don't worry about the belt then. More trouble than its worth for a simple diagnosis.

Any potential diagnosis I offer from this point on is purely speculative.

There's a chance that the tip of the nozzle is getting snagged on a part of the print. Although this shouldn't happen, and when it does it is typically attributed to operating at much higher temperatures than the Gordon-provided Slic3r settings of the gMax, the fact that the issue occurred at the same point multiple times is suspicious.

Why don't you try something like this: http://www.thingiverse.com/thing:199538/#files
--

Now that I read your post from a few moments ago, just I was about to submit the above, it could be that the acceleration is too high. Now, acceleration for the individual axes is addressed in the firmware, and we're not going to mess around with that. However, we can change some acceleration settings in Slic3r to test this theory out. The settings I am referring to are in the 'Print Settings' tab of Slic3r, and under that, the Speed section. Acceleration settings are at the bottom there, and in my case most of those are set to 500. Try changing the 500s to 350 or so. Save, Slice, and try another print.