Leveling the bed (e.g., doing the impossible)

Check here for instructions and tips on how to operate your printer.

Leveling the bed (e.g., doing the impossible)

Postby dclark » Wed Sep 23, 2015 5:56 pm

Hello gCreate community,

I am the exasperated owner of a gCreate 1.5 XT. In the distant past I had thermister issues that have been rectified for the moment. My current challenge is getting the bed leveled so I can produce a large format print - which was the only reason I purchased this printer in the first place.

I am trying to print half of a rectangular (almost square on the bottom) container. It measures 340mm x 330mm, which should be well within the size of my dual extruder 1.5 XT print area. Unfortunately, the acrylic bed is impossible to level. I say "impossible" simply because I've spent the last 45 or so hours of my working life trying everything I can to get this bed level. No matter what I do, I can't seem to adjust for the too-high back/center and front/center (the middle center is a different height) and a too low middle/left and middle right.

My question for the group is three-fold:

1) Has anyone been able to print something with a similarly sized first layer? (e.g., 340mm x 330mm)

2) How did you level your bed?

3) If you found the supplied acrylic bed unusable (as I am finding), what options did you turn to?

I am really frustrated by the difficulty of getting this machine to work reliably. I have successfully produced five prints in the last 10 months:

- A 2cm x 2cm x 2cm hollow cube.
- Two 1/4 sized early prototypes of the storage tray/box I am working on.
- A 1" x 1" x 1" cube with 40% fill.
- One 1/4 sized later prototype of the storage tray/box I am working on.

I can't seem to get anything bigger to even successfully print the first layer. There are other issues I could go into with the reliability of the print head, etc., but I think I could get past those if this defective print bed would level.

Any help you can provide is appreciated.

Thanks,

Doug Clark

P.S., for those of you wondering what process I've used, I've detailed it, and the results of several measurements, below:

The process I have developed to try and keep track of what is going when leveling my print bed is:

1) adjust one or more (but usually one) of the seven adjustment points, which include:
- back/left wingnut
- back/right wingnut
- front/left wingnut
- front/right wingnut
- left z-axis coupler
- right z-axis coupler
- z-axis endstop screw

2) Auto home the print head

3) Turn off the printer (so I can move the bed around, carefully, by hand)

3) using numbered pieces of laser printer paper (to as accurately as I can account for differences in paper thickness), measure and record the distance between the bed and the print head at 9 spots:
- back/left
- back/center
- back/right
- middle/left
- middle/center
- middle/right
- front/left
- front/center
- front/right

4) Evaluate the current measurements, choose your next adjustment, and return to step 1

Here are the results of the last 13 measurements:

KEY:

+ : a piece of paper fits under the print head and moves with no resistance
. : a piece of paper fits under the print head with a minimal amount of resistance (I can feel it, but just barely)
- : a piece of paper fits under the print head with a noticable amount of resistance
-- : a piece of paper fits under the print head with a significant amount of resistance
--- : somehow a piece of paper slipped under the print head, but it is very difficult to move.
0 : I cannot fit a piece of paper under the print head.


Starting point (I'm trying to wrap these diagrams in "code" blocks, but don't know if that will work):
Code: Select all
   -------------------
   |     |     |     |
   |  -- |     |  -- |
   |  -  |  0  |  -  |
   -------------------
   |  -- |     |     |
   |  -  |     |     |
   |  +  |     |  -  |
   |  +  |  -  |  +  |
   -------------------
   |  -- |     |     |
   |  -  |     |     |
   |  +  |  -  |  0  |
   -------------------


Explanation (I'm just putting one of these here in long-form sentances so you can hopefully understand what my diagrams mean):
back/left - two pages fit under the print head, one with a noticable amount of resistance, another with a significant amount of resistance.
back/middle - no pages fit under the print head
back/right - two pages fit under the print head, one with a noticable amount of resistance, another with a significant amount of resistance.
middle/left - four pages fit under the print head, two without any resistance, one with noticable and another with significant resistance.
middle/center - one page fits under the print head with noticable resistance
middle/right - two pages fit under the print head, one with no resistance, and a second with noticable resistance.
front/left - three pages fit under the print head, one with no resistance, a second with noticable and a third with significant resistance.
front/middle - one page fit under the print head with noticable resistance
front/right - no pages fit under the print head.

Adjustment:
Twist the back/left wingnut clockwise, 1/8 of a turn (this should raise the back/left of the bed up slightly, among other things)

Result:
Code: Select all
   -------------------
   |     |     |     |
   |     |     |  -  |
   |  -  | --- |  +  |
   -------------------
   |  -  |     |     |
   |  +  |     |     |
   |  +  |  -  |  -  |
   |  +  |  +  |  +  |
   |  +  |  +  |  +  |
   -------------------
   |  -- |     |     |
   |  -  |     |     |
   |  +  |  -  |  -- |
   |  +  |  +  |  -  |
   -------------------

Not what I expected with such a small, localized change.

Next adjustment:
Twist back/left wingnut counter-clockwise 1/4 of a turn. lowering the back/left corner):

Result:
Code: Select all
   -------------------
   |     |     |  -- |
   |  -- |     |  +  |
   |  -  |  -- |  +  |
   -------------------
   |  -- |     |  -- |
   |  +  |  -- |  +  |
   |  +  |  +  |  +  |
   -------------------
   |     |     |     |
   | --- |  0  |  -  |
   -------------------

Next adjustment:
twist back/right wingnut clockwise 1/8 of a turn, raising the back/right corner slightly.

Result:
Code: Select all
   -------------------
   |     |     |     |
   |     |     |  -- |
   |  -  |  0  |  -  |
   -------------------
   |  -  |     |  -- |
   |  +  |     |  -  |
   |  +  |  +  |  +  |
   -------------------
   |     |     |     |
   | --  |  -- |  -  |
   -------------------

Next adjustment:
Rotate right z-axis coupler counter-clockwise 1/8 of a turn, lowering the right end of the extruder rail slightly.

Result:
Code: Select all
   -------------------
   |     |     |     |
   |  -- |  0  |  -- |
   -------------------
   |  -  |     | --- |
   |  +  |     |  -  |
   |  +  |  -- |  +  |
   -------------------
   |     |     |     |
   | --- |  0  |  -- |
   -------------------

Next adjustment:
rotate z-axis stopper 1/8 of a turn clockwise, which should raise the print head slightly.

Result:
Code: Select all
   -------------------
   |     |     |     |
   |  -  |     |  -- |
   |  .  |  -- |  -  |
   -------------------
   |  -  |     |     |
   |  +  |     |  -  |
   |  +  |  -  |  +  |
   |  +  |  +  |  +  |
   -------------------
   |     |     |     |
   |  -  |  -  |  -  |
   -------------------

The back and front seem close to the desired height, but the middle is too low. Maybe the wingnuts are too tight?

Next adjustment:
twist all wingnut screws clockwize 1/4 turn, which should raise the entire bed up on all four corners.

Result:
Code: Select all
   -------------------
   |     |     |     |
   |  0  |  0  |  -- |
   -------------------
   |     |     |     |
   |     |     |  -- |
   |  -- |  0  |  -  |
   -------------------
   |     |     |     |
   |  0  |  0  |  0  |
   -------------------

Can't really tell how much of a difference that made, as most of the bed is too high to allow paper under the extruder head.

Next adjustment:
rotate z-axis stopper 1/4 of a turn clockwise, which should raise the print noticably.

Result:
Code: Select all
   -------------------
   |     |     |  -  |
   |     |     |  +  |
   |     |     |  +  |
   |  -  |  -  |  +  |
   -------------------
   |     |     |     |
   |     |     |  -  |
   |  -  |     |  +  |
   |  +  |  -  |  +  |
   |  +  |  +  |  +  |
   -------------------
   |     |     |     |
   |     |     |  -- |
   |     |     |  +  |
   |  -  |  -  |  +  |
   -------------------

How in the world did my adjustments cause the right side to get so out of balance?

Next adjustment:
rotate back/right and front/right wingnuts 1/8 of a turn clockwise, which should raise the right side of the bed.

Result:
Code: Select all
   -------------------
   |     |     |     |
   | --- |     |     |
   |  -  |     | --- |
   |  +  | --- |  -  |
   -------------------
   |     |     |     |
   |  -- |     |     |
   |  -  |     |     |
   |  +  | --- |  -  |
   |  +  |  .  |  .  |
   -------------------
   |     |     |     |
   |  -  |  -- |  -  |
   -------------------

This seems almost OK, but the dip in the middle left makes it impossible to start the left edge of my print.

Next adjustment:
twist back/left wingnut 1/8 of a turn clockwise, which should raise the back/left corner of the bed.

Result:
Code: Select all
   -------------------
   |     |     |     |
   |  0  |  0  |  -  |
   -------------------
   |     |     |     |
   | --- |     |     |
   |  -  |     |  -- |
   |  +  | --- |  -  |
   -------------------
   |     |     |     |
   |  -- |     |     |
   |  -  |  0  |  -- |
   -------------------

How did raising the back/left corner of the bed also raise the front center of the bed? (sigh)

Next adjustment:
rotate z-axis stopper 1/4 of a turn clockwise, which should raise the print noticably.

Result:
Code: Select all
   -------------------
   |     |     |     |
   |  -  |     |  -- |
   |  +  |  0  |  -  |
   -------------------
   |     |     |     |
   |  -  |     |     |
   |  +  |  -  |  -  |
   |  +  |  +  |  +  |
   -------------------
   |     |     |     |
   |  -- |     |     |
   |  +  |  -- |  -  |
   -------------------

I need to somehow lower the back/middle and front/middle and raise the left edge and right middle...impossible.

Next adjustment:
rotate back/left and front right wingnuts 1/8 of a turn counter clockwise, which should lower the back/left and front/right corners.

Result:
Code: Select all
   -------------------
   |     |     |     |
   |  -  |     |  -  |
   |  +  |  -- |  -  |
   -------------------
   |     |     |     |
   |  -- |     |     |
   |  -  |     | --- |
   |  +  |  -  |  -  |
   |  +  |  +  |  +  |
   -------------------
   |     |     |     |
   |  -- |     |  -- |
   |  -  |  -  |  -  |
   -------------------

I got rid of the zeroes, but I'm not sure what to do next...
These users thanked the author dclark for the post (total 0):
dclark
 
Posts: 10
Joined: Wed Sep 24, 2014 10:17 pm
Has thanked: 0 time
Been thanked: 0 time

Re: Leveling the bed (e.g., doing the impossible)

Postby AlexD » Thu Sep 24, 2015 5:36 am

Hi Doug,

Wow first of all that is a very detailed analysis of your levelling process! I can sympathise with your frustration, i've had a tough time levelling the acrylic bed on the gMax at my work too. Getting it level is tricky but not impossible, that said a level bed will not help that much on an un-even bed surface. If you think that the acrylic is un-even (ie is not flat, has dips / raises), then it might be a suggestions to flip to the other side, rough the surface and begin again.

You mention adjusting the Z Axis Screws? You shouldn't be needing to do this. Grab a tape measure and make sure that the distance from the 20 x 20 V Slot to the top of the printer frame is the same, you could also measure from the top of the X Axis parts, to the bottom of the Z Axis Couplers, make sure they are the same.

Once you know the X Carriage is true to frame you can remove that from the equation, assuming the acrylic is flat you should be able to get it level by doing what you have done already.

Its worth mentioning also that newer marlin firmwares have Z Babystepping, so you can adjust the first layer real time without touching the Z Endstop. So you can use this to get the Endstop dialed in perfectly.

If you feel upto the task you could also retro-fit Auto Bed Levelling to the printer, i've got some extruder parts you could print if you decided to.

Alex

PS. Re your question about acrylic alternatives, you can always print on glass using hairspray, 3M painters tape etc. I use heated glass on my home built one.

On the work printer i noticed really quickly that we were wearing a 'low spot' in the acrylic bed by printing smaller objects in the middle all the time. Luckily we have a massive calibration sander at work (a brand new one), when they demo'd it to us they showed it sanding the writing off of a sheet of paper. Therefore i am able to ensure that the acrylic is PERFECTLY flat. Although when screwing the bed back down you can still warp the surface uneven.
These users thanked the author AlexD for the post (total 0):
User avatar
AlexD
 
Posts: 78
Joined: Wed May 13, 2015 10:07 am
Has thanked: 2 times
Been thanked: 10 times

Re: Leveling the bed (e.g., doing the impossible)

Postby gCreate_Kyle » Mon Sep 28, 2015 7:28 am

Hi Doug,
Your acrylic sheet most likely is not flat. This happens when the acrylic was exposed to some heat. Usually from a hot truck in shipping . Unfortunately acrylic warps slightly from even just a little heat.

A year ago I was testing the thermistor with a hair dryer, Yes a bad idea. In doing so I damaged by the acrylic sheet. The center of the bed warped slightly down. I could only successfully print in the center of the bed. It took me a while to figure out what happened but when I replaced the acrylic sheet I had much more success with bed leveling. There were still some high spots in the acrylic. I sanded them down with a 40 grit sand paper. I have been using this acrylic sheet ever since. I have successfully printed in the entire area of the bed many times. You should be able to purchase acrylic from your local hardware store. The gMax uses a 17 7/8 X 17 7/8 sheet.

I hope this helps. Email me at support@gcreate.com if you have questions for me.
These users thanked the author gCreate_Kyle for the post (total 0):
R. Kyle Gress
gMax Technical Support
3D Printing/Electronics Enthusiast
User avatar
gCreate_Kyle
 
Posts: 169
Joined: Mon Feb 02, 2015 8:47 am
Location: Lancaster, Pennsylvania
Has thanked: 11 times
Been thanked: 18 times


Return to gMax Operation Instructions

Who is online

Users browsing this forum: No registered users and 1 guest