Confused by the Leveling Print Bed Instructions in Manual

I thought that I had leveled my bed well enough when I first unpacked my printer, but I think that it needs to be tweaked just a little more. However in the the manual regarding how to level the bed. I’m confused about the following:

“To raise or lower the front right corner of the bed adjust the third screw from the left. Do not adjust the middle screw”

I’m not positive about which set of screws I need to adjusting. I want to take a guess that they are the on each of the four four corners of the print bed. Is this correct.

A picture would be greatly appreciated.

Thank You for your time
Wazman

Yes, you will want to adjust the 4 corner screws that hold the heated bed itself. the best way I’ve found to make adjustments is the “piece of thick paper” method, where you take a piece of paper and stick it in between the nozzle and the glass plate to where it touches but still moves smoothly. ideally you want to do this when the machine is warm since the metal expands when hot, and that is how you will be printing, but paper is flammable so it is better to get it close with the paper cold first and then adjust the Z axis after it is warmed up. a thin piece of metal such as a straightedge also works for this.

The screw on each corner that needs to be adjusted is the one with the spring.

So spinning the screw with the spring clockwise lowers that corner of the bed, right?

Correct. Screwing the screw tighter = down.

I’ve followed the leveling directions, as well as read online as much as i can…
however, i have spun the far right corner leveling screw all the way down (in the process of leveling) and am redoing the paper test and still cant seem to get all 4 corners the same. Do any of you guys have one leveling spring compressed much more then the others?

I’ve check the threaded rod measurements a million times and they appear the same? …i’m going to buy a micrometer soon.

I’m thinking maybe the initial 4 screws that hold the bed onto the frame are off? I unscrewed and re-tightened them but no luck.

Also, once i run the bed leveling test, the plastic should appear the same thickness and “density” in every spot of the test right? - in my case i can easily tell that it is off, but I once i adjust the corner that appears to be off and re-print i end up throwing the bed off even more.

Im assuming doing the paper test is a more accurate method then printing the calibration test, adjusting a screw and reprinting? Do you guys use a combo of these 2 methods?

Any input here or tips to help me? I feel like the more I adjust things the deeper i mess it up!

Thanks,
Aaron

You can check your leveling easily using this method to get you into the ballpark.

  1. Ensure your printer is on a level (using a bubble level), flat surface and the frame is level and flat and not sitting on any obstructions.
  2. Measure from the surface of your table to the bottom corner of the actual heated bed mounting platform (the aluminum plate that the heated bed retention screws screw into) at all 4 corners using a MM ruler. If all 4 corners are identical distances from the table top, now check the spring heights.
  3. Using a ruler again, Measure the height of whichever of the 4 heated bed screws seems to be about where you want it to be. Now measure all 3 other screws and ensure they have that same measurement.
  4. At this point ensure your Z axis stop is in the correct position so that the nozzle is about 1 thick business card space away from the top of the heated bed glass.
  5. Now run your calibration print. Be aware that the metal parts of the nozzle do expand the hotter it gets, so you literally may want to let it sit and warm up 15 minutes at temperature before you even think about running the calibration print. It is quite possible to calibrate with the machine at operational temperature, but still “cold” and then get a different calibration measurement
  6. If you find that the machine is still off, examine where it is off and adjust accordingly. If you are off at both corners on one side or the other of the X axis, you may consider turning the actual threaded rod to raise or lower that entire side rather than adjusting the screws. Ideally the X carriage itself shows level with the table top and the heated bed at that point.
  7. If its still off or still giving you trouble, take some pictures of the bed and screws and post them here and we can see if we can help you further.

I did, it was my front right corner. I actually had to cut the spring little by little to be able to compress it enough.

Here are some things that helped me when I first set mine up:

  1. Verify Z Axis Leveling (PG 49 TAZ 2.1 manual) Instead of using a 150mm ruler I used a 120mm machinist’s square
    Square.jpg
    because it can stand up by itself, freeing my hands. Also who can really measure accurately from the flat plane of the aluminum bed plate to a rounded edge of the X axis smooth rod? Since there is no spec for this distance you don’t really need a ruler, you need an accurate spacer between these two points. Put the Z axis high enough that the square sits under the smooth X axis rods, and little by little lower the Z axis switching from millimeters to tenths of millimeters as you get close. Don’t actually bring the smooth X axis rods into contact with the machinist’s square, stop 1 or 2 tenths of a millimeter above it, you will visually be able to see the gap. Then move the square to the other side of the bed and see if the gap is the same. I used a thick piece of card stock to slip between the square and smooth rod, the feel of the resistance on the paper told me how much space there was. You want both sides to feel the same. When you determine which side needs to go up or down, twist by hand the threaded rod to move that side of the smooth X axis rod up or down (make sure you turned off your motors before turning the threaded rod each time). Raise the Z axis up 20mm, then repeat the steps above until both sides are the same distance from the square. You have now leveled your Z axis to the aluminum bed plate.

  2. Leveling the print bed. Turn your bed temperature on to 85 and your nozzle to half of whatever you will extrude at (ABS extrudes at 220, so turn nozzle temp to 110). I find this expands these parts close enough to where they will be during printing, and because you didn’t bring the nozzle all the way up to temp, it will not slowly ooze filament while you are leveling. I use a thick card stock slip of paper as a feeler gauge at each of the four corners. I just go corner to corner getting the card stock to feel the same resistance when I slip it between the nozzle and bed. Turn your screws accordingly to raise or lower the bed, and just keep going corner to corner (clockwise or counter) until they all feel the same. If there is no contact or too much contact with the card stock at any corner, you need to adjust your Z end stop trigger so the nozzle sits lower or higher. Don’t worry if the nozzle is too high or low, it needs to feel consistent at each corner. Once everything is consistent you can change your Z end stop trigger adjustment accordingly. Always raise your Z axis 20mm before homing it.

  3. Now I home all my axes, and visually put a small .5mm space between the nozzle and print bed by adjusting the Z end stop trigger. It doesn’t need to be exactly .5mm, there just needs to be a small space. Now bring nozzle up to temp, home and print the bed_level.gcode file. You can see the first layer adhesion instantly and if it seems to look off, go ahead and stop the print, pull the filament off the plate, adjust your Z end stop trigger accordingly and print again. You can even use a digital caliper to measure the first layer height and get an idea of how thick it is. Your brain will start to associate a number with the results you see. Also I used a black permanent marker to put a small line perpendicular to the edge of the red Z end stop trigger, this way you have a visual reference of how much you are turning the stop trigger.

Hopefully that stuff helps, I remember it being the big three things that got me going. The first few days I had to repeat steps 2 and 3 above before printing, and after a while it kind of locked itself in. Now I will do steps 2 and 3 before a print, but don’t really have to adjust anything, just check to make sure it is still there. I’m still working on user:1013’s auto bed leveling stuff on the forum which should make most of this leveling confusion obsolete.

Wow…
Thanks so much guys. You don’t often run into this much help in the early days of joining a new forum!
Thank you for the tips, I am now no longer frustrated but excited to go home and try these various methods of leveling!

I’M going to stat step 1: checking the level of my desk. I have a feeling this may be where i am going wrong as i never checked the level of my desk to begin with. The fact that only 1 corner is off made me think this wasn’t the case.

Honestly thanks guys! I appreciate it, I will reply if i have more problems leveling.
PS: Auto leveling to come!? I’m going to search for this topic ASAP!

Thanks again,
Aaron :nerd:

2con: The main thread can be found here: https://forum.lulzbot.com/t/auto-bed-probing-and-compensation-with-g29-development/162/1 Can’t wait to see what you do with your printer!

If not, how do I go about adjusting the height of the aluminum plate from my desktop?
Im using a micrometer and I am getting ~1.5mm difference of the far right and left (parallel to the x-axis)

Looking at the unpacking manual here: http://download.lulzbot.com/TAZ/documentation/current/LulzBot_TAZ-Unpacking_Sheet.pdf you can see how the Taz is put together. you have the outer frame, which is resting on the table and should theoretically be level, then you have the Y axis assembly that screws down onto that outer frame. I’m guessing that join point is where your issue is. Check the demensions on the side that is off and see if there is any kind of obstruction, maybe a wire or something pinched between the two frame sections that would cause one side to be off significantly. If everything looks right, measure the thickness of the joiner assembly pieces and see if oone of the Y axis mount plates is off for some reason. Work your way up measuring from the table up to the metal bearing rods themselves. if everything is level up to the rods but the heated bed plate itself is unlevel then you might have a messed up rod bearing or bearing assembly. I can’t find a good picture of the Y axis bearings on a Taz so i’m not sure what they are using as the mount piece there, but somewhere in that assembly something is off 1.5mm. You should still be able to compensate for that using the bed leveling springs. but you will have much more success in the long term if you find the error and correct it now rather than working around it. You can always add shims on the low side if trimming the high point isn’t a possibility. You might also consider taking a bunch of pictures of the frame join points, the underside of the heated bed plate where the bearings mount, and any other contact point along both sides and posting them here. we may be able to see something that looks odd right away.

So I was a bit wrong, the difference of 1.5mm was combined, so actually my aluminum plate was only 0.75mm off from my desk height.

I took another swing at it today with the help of my friend.
We started by re-leveling my desk, which helped a bit.
We found it very hard to accurately measure the distance from the x-axis rod to the aluminum plate because it was hard to hold the micrometer “the same angle” on each side and also couldnt get enough alumnimum plate to measure from . Instead we used my iPhone 5 (on end) as a spacer and found the x-axis was out! We aligned each corner as per the manual and ran many calibration prints. Each time we found one corner of the ABS to be too high or too low we would adjust that corner screw by an 1/8th turn and run the test again.
I started measuring the width of the first lay of the calibration test with my micrometer in each corner and doing micro-adjusts on each corner until the first layer width of ABS was between 0.3mm & 0.4mm. It seems to be level now, but I will keep in touch as I progress. I want this thing to be mint!

now the only issue i’m having is the plastic not sticking! I suppose its from all my greasy finger-prints over these test runs. Im currently going to buy a nice bottle of acetone…

PS> The psychical octo sent with the printer resembles a perfect print right? No settings are different from the one I run, right?

THe octopus is generally printed on your printer before they ship it to you as proof that the printer was working when they shipped it. That being said, parts do shift in shipping, especially during some of the negitive temperatures we’ve seen recently so don’t be suprised if a few things are off the first time you put it back together. If you are willing to take the time and effort to really dial in your printer, the extruder and all the subassemblies, you can achieve quality better than that. There are upgraded carriage and motion system components that the community has put together that you can print and improve your accuracy (different Z nut springs, auto levelrs, stiffer X carriages, etc). You can take the time to do a full calibration of your extruder (http://reprap.org/wiki/Triffid_Hunter’s_Calibration_Guide) and theoretically increase your individual layer accuracy. If you are willing to print slower you can switch out to a smaller nozzle and smaller layer height

Generally the settings for that octopus print will be the same ones they upload to the software area. If you are having trouble getting things to stick to the bed, bumping the temperature up slightly and printing with a 2-3mm “brim” layer can help if you don’t mind doing a bit of trimming. You can also try printing with additional perimiters for a more defined outer layer. I find for me it works best if I tackle one axis of calibration at a time. Sit down one session and dial in the Y axis as much as I can get it (belt tension, allignment, measurement, etc). Then next time dial the X axis in. trying to do them all at the same time leads to moving problem areas around rather than fixing them. Once you do get it all dialed in perfect though, the quality you can start to produce becomes orders of magnitude better.

I like to print that gcode toy dumptruck that lulzbot has in the download section from time to time and then compare the resulting print against the first one i ever printed just to see how much the print quality has improved since then.

Thank you so much! I am a very particular person so dialing in sounds like the perfect plan for me! I am learning so much every time I turn this thing on, way more then I expected.
your long term routine for dialing in my printer sets a new goal for me! thanks i appreciate the help
Aaron

Yes, leveling the bed can be confusing. At first it seems like a juggling act but there are some hacks. I have a TAZ 5. Can’t afford auto-leveling so what I am giving you is really basic.

One of the first things I did was to put a white mark on the Z offset nob. Now when I turn it I know how many turns I have made exactly and there is an equality for the number of turns vs. the offset. Think of it as a jar lid. Counter clockwise raises the lid -and the leveling screws. Clockwise lowers the lid. The more you turn the Z offset nob counterclockwise the higher the extruder head will stop away from the bed. If the leveling screws are too tight you want to raise them so you will have to raise the X axis by turning the knob in order to give you room to raise the bed under the nozzle.

Here is another hack for leveling the X axis. I know not everyone has a miter saw (had one handy) but you probably know someone who does. I cut a 3/8 inch piece of plywood (had one handy) about 3 inches long and 50.6mm wide. Experiment for what works for you. The manual shows that you should use the metal ruler to measure the x axis on either side of the print bed beginning from the aluminum. I think this is “a pain in the tookus.” Remember HS when in science class they taught you to use a ruler and everyone had a different idea? It was never consistent. Placing the “block” on the aluminum at the back end and lowering the X axis by manually turning the threaded rods will give you a consistent height for the x axis every time. No bending over and going “Humm, is this right? Well maybe.” Duh!

Now for the bed leveling screws. I have over tightened and stripped the aluminum leading to replacing the aluminum bed. You could sauder a hex nut to the back of the aluminum bed but that might eventually break. Orrrr, just tighten the bed leveling screws down till they are snug and given that the x axis is level go on to the next step.

First I home when the bed is warm. Expansion. Also paper will burn at Fahrenheit 351. (Never read the book) That is 233 C. You don’t need your bed or your nozzle that warm when you level. That said, “Home All.” Watch to see if the nozzle forces the bed down. You don’t want that. Note that if your nozzle forces the bed down you can fix that by turning the Z offset nob counter clockwise. Do it one turn then “home all.” Make sure the bed is not forced down.

This is why I put a white mark or “tick” on the nob. I can tell how many turns I make and get pretty precise. If the bed is not forced down - here is the fun trick- you can slip or try to slip a piece of paper under the nozzle or your can use a bent finger gauge (No paper, no fire). If you use a digital caliper, (you should have one because it really helps.) you will find that the thickness of a “folded piece of paper” is .18 mm. Trust me. Use your finger gauge and get .18 mm between your bed and n nozzle. It’s always reliable and it’s slippery where paper is rough. You will feel the difference but if the bed is not forced down when you slip the gauge under the nozzle you’re golden. Business card stock varies in thickness (from .27 to .35mm) so, in my belief, it’s unreliable.

Now for the bed screws and springs. Yep! I note the same thing. One or more corners will seem to be screwed down tighter than others, or the bed corner is higher on one side than another. It’s called spring tension governed by the spring constant and I personally think it varies slightly from one spring to another. Unlevel bed corners catch on threads and stick. I have to get an order of springs in one package and replace all of them to check but this has been a bother to me for a long time. …and yeah I have replaced my springs before. Same problem.

Another problem is that the bed corner gets caught on the “inverted” screw (mentioned above) and to level you have to push the bed corner down. Sometimes it will pop and the corner will level out or the corner will be visually unlevel. You will have to "dink’ with it by raising the Z offset nob then re-leveling. The point is to give yourself room to raise the bed levelers. That is to say if the distance between the bed and the nozzle is too tight raise the Z offset by a counter clockwise turn, check the space between the bed and the nozzle with your paper or finger gauge and if you have to because the bed is pushed down, repeat until the (Here is the trick with the leveling screws being too tight.) nozzle is raised sufficiently so as to ease the turning of the leveling screws and all four corners are even with the finger gauge at .18 mm between the bed and the nozzle. One other thing, I used a longer inverted “stabilizing” screw on the bed corner. I just have a little more flexibility when adjusting the bed height. I call them “post screws” because they look like posts and I think they should be smooth at a level above the aluminum because the threads serve no purpose and the levelers need to slide so they don’t get caught when leveling the bed with the leveling screw.

One final thing before I confuse you more. There is an exact number of turns of the Allen wrench per fraction of a mm you raise the bed corner. Sorry, I have forgotten it for the moment but it will help you when you realize that you can divide precisely a complete turn and equate that with an increase in bed height. One turn equals x mm of height. It’s important if you are making molds and can get within .04 or in some cases .02 mm. Takes a math nerd.

A point of clarification. When I say “That is to say if the distance between the bed and the nozzle is too tight raise the Z offset by a counter clockwise turn, check the space between the bed and the nozzle with your paper or finger gauge and if you have to because the bed is pushed down,” I meant to say, "just give yourself a bit of space between the nozzle and the bed using the Z offset knob so you have room to raise the bed, at which point you will find it a bit easier to level the bed, by turning the leveling screws counterclockwise. I really hope that makes sense. I have made a lot of mistakes and what I offer here is what I have learned.

1.) The height of the aluminum plate is fixed. You can only raise the glass by turning the leveling screws on the bed corners if you are using an TAZ 5 as I am.
2.) To heck with the ruler. A digital caliper will save you tons of frustration and well they cost maybe 20$ You just spent 2 grand on a 3D printer so don’t be a tightwad. Save yourself some grief.
3.) if you are “auto leveling” let the machine do the work. Look on the LCD in the firmware.