I’ve found that a very easy way to get your tramming dialed in all the way is to tram it in the way described in this thread or on the wiki. Then, do a bed levelling calibration and run a print that goes across the entire build plate on the first layer. Check how much the z-axis lead screws rotate as the toolhead moves across the plate. This is easier to see if you stick a tiny piece of filament or something to the lead screw to actually see how much it is rotating. Then, see adjust the plate depending on where it moved up or down (i.e if the bed moves down when the printhead is moving to the back, tram the back of the bed down a little bit). Repeat until your bed is dialed in well. Worked perfectly for me.
1 Like
Did you ever get this figured out? Im having this same problem over and over and over again in the exact same spot.
On another note I’ve been cursing the fact that there is no pause for user action G-Code [that we know of] that doesn’t move the toolhead to the chute like “M400 U1” does. So the tramming G-Code waits for a hard coded 30 seconds on each of the three spots, then goes over them all again two times with 15 second pauses. It would have been nice if you could fiddle with the gauge for any amount of time, then press “next” or something like that.
It did occur to me a while ago though, that you can do it this way:
-
Start the tramming G-code and wait until it pauses at the first spot. Then turn power off! Now adjust that point for any amount of time, until you are really happy with it.
-
Power on and run the G-code again. This time let it run until it pauses at second spot. Power off again and adjust.
-
Same procedure for third / last spot.
I’ll try this next time I feel a need to level the bed. Note BTW that this procedure is only sensible when using a 0.4 feeler gauge. If you’d set it to “barely touch” for the first spot, it would be awfully tilted when powering on again and all sorts of sad things could happen, possibly including nozzle scratching the plate…
This is sensible, and I’ve tried this wonderful model that puts a dial on the front right Z-screw for making it easy. I loved it but need to modify it because the cursor slips far too much at least on my printer. I believe the magnet sits on the bearing as opposed to the rod. One of these days I’ll revisit it and fix that problem. Some other size magnet and/or some part inbetween is needed.
How many bed adjustment knobs/screws are there? I see the two in front, and it seems like there’s probably one in the middle in back. Is that right, just 3?
1 Like
That’s correct, only 3 knobs for adjustment.
Can you please provide a step by step on how you use the feeler gauge?
I’ve tried to find videos but they don’t seem to be what you’re describing.
I’m sure there are basic youtube videos if you search a little more. For a real-life use case you can instead search youtube for how to set valve lash on a Moto Guzzi motorcycle because that too needs a feeler gauge 
Or setting points on old engines.
Anyway, when the toolhead stops a bit above the plate at a calibration point, you just stick the gauge in the gap. If you can’t, the gap is too tight and you need to adjust.
Conversely, if the gauge fits in the gap without any drag at all, the gap is too large. When the gap is correct, there is a slight drag but you don’t need much force. You’ll get a feeling for it.
Edit: It might be a good idea to just run the calibration procedure without actually adjusting any of the thumb screws. That should give you some base line and you can see how the current tramming is, without upsetting it. Do this for as many times you wish and you should quickly get some experience!
Excellent. This is extremely helpful. Thanks!
THANK YOU! Having a papper at 0.4mm thick solved the nozzle scraping across the print bed for me! After many sleepless nights, I’m finally back to pristine quality
Hi, my bed needed tramming which i did sucessfully a few months ago, but using a different method. Ive used the feeler gauge this time round as needed to do it again. The 30 seconds is not enough for me. My question, is can i run the g code and when it stops onbthe first corner, switch it off and manually adjust ?
I have an X1C
You could, but you will have to re-rerun the code, let it move to the first corner, wait until it goes to the second position, turn it off, adjust, re-run the code, wait until it moves to the third position, etc.
Try the way I described in this earlier post: Superior (non-destructive) bed tramming procedure - #19 by lkraus
The use of the wrench makes it much easier to make small, precise adjustments.
Using a light to find the “just touches” point is more distinct than trying to judge equal friction with a feeler gauge.
Please note that the bed tramming gcode file is viewable and editable with any standard text editor (such as Notepad). There are lines such as M400 S30 ; wait 30s (the thing after the semicolon is just a comment) and you can change that to eg. M400 S90 ; wait 90s or whatever figure you like. Note that you need to change at least three of such pauses in the file.
That makes perfect sense. I did use your method, just ran out of time 
very mich appreciate tou getting back to me !
I saw that and completly make sense but wasnt sure about the “M” number, but assume that just means “move” and speed of the move, but will give that a go 
thank you !
Can’t you use the old paper method?
Well, I wish I’d found this an hour ago before I’d moved the screws…
Hi Raz, would you recommend using a .4 feeler gauge on an a1 mini too? having awful problems with my print quality
I wouldn’t know, but I’d guess it’s the same.
Hi folks,
Although this topic seems to be “resolved” the comments here don’t quite match what’s currently on the BambuLab Wiki page. Specifically, the G-Code file now has a 0.1mm value rather than the 0.4mm that is mentioned in this post’s solution.
Looking into the G-Code file I can see there is a Z0.1
Compared with the 2022 G-Code file from earlier in this post where it is mentioned a Z0.4
The Z0.1 in the new file vs the Z0.4 in the older file seems to indicate that BambuLab has adjusted their tramming G-Code to use a 0.1mm feeler gauge aka a sheet of typical A4 paper.
Fortunately I don’t yet need to tram my A1 Mini’s bed as I have it only about 1-2 months, however based on my previous experience with Creality 3D printers, I’m bound at some point to need to tram the bed. Therefore having a reliable, consistent, effective and repeatable procedure is a good thing to have.
To that end, based on this newer G-Code file from the Wiki, does the 0.4mm feeler gauge still apply or is the sheet of paper a good option or an actual 0.1mm feeler gauge a better option?
I also have some 3D printed “feeler gauges” which I plan to use instead of the steel ones I can buy at any hardware store. I’d rather go with a plastic or non-metal feeler gauge as I’ve had a nasty surprise of shorting out my other printer’s bed temperature sensor while doing tramming with the bed heated. In hindsight, it was dumb thing to do to use a metal feeler gauge on a electrically heated bed but hindsight is always 20x20.
So, coming back to the topic, tramming should not be done with a 0.4mm gauge but with a sheet of paper, roughly 0.1mm thick OR a regular 0.1mm feeler gauge, and not with the bed heated. Also use the paper sheet/feeler gauge to measure the gap between the nozzle and the bed before modifying the screws so you get to see what the tramming of the bed is at the moment.
The one thing to mention about the G-Code file is that the S5 lines can be modified to increase the wait duration at each of the 3 points so you have more time to adjust.
Am I missing anything else?
Thanks.