Nozzle scraping bed during beacon calibration
I'm starting to feel like the problem child...
New V4/400 build. Following the commissioning guide ran the "Beacon_Initial_Calibration and saved the config. Homed the toolhead and began the initial bed mesh calibration. As the toolhead made it to the back left of the printer the nozzle starting hitting the edge of the bed leaving a deep scratch.
Is there a step I missed in the initial calibration that I need to redo? Something to be checked on the toolhead? Rerun the initial Beacon cal? TIA
Is there a step I missed in the initial calibration that I need to redo? Something to be checked on the toolhead? Rerun the initial Beacon cal? TIA
58 Replies
Yes. You probably missed z_tilt calibration.
step 4 from initial calibration.
No, it was during the bed mesh calibrate of the Z_tilt ops that it occurred.
Bed mesh is one thing. Z_tilt is another...
and step 1 tells to
Insert "M84" on the Console, this will disable the stepper motors.
Proceed to manually adjust each lead screw to get the bed roughly levelled, it doesn't need to be perfect, the machine will perfect it later.
perhaps it was not enough
This is where I was. Homed the toolhead and then hit calibrate. The toolhead began moving across the bed, and then beyond it, scraping the bed surface as it came back across. Is there a parameter that needs to be set in Klipper that may have misdefined the extents of the bed?
before that, step 1
Did that, and saved the config
there is nothing to save on step 1.
Step 1 - Build plate preparation
It's normal to have the bed tilted after assembly. Insert "M84" on the Console, this will disable the stepper motors.
Proceed to manually adjust each lead screw to get the bed roughly levelled, it doesn't need to be perfect, the machine will perfect it later.
Move the toolhead to each Z position and adjust up or down as required.
I used a 4 foot level at that point to level the bed. I thought you were referring to the initial beacon calibration
Step 1 of initial calibration.
this is to have the bed at least in a decent position to avoid scratching
I didn't touch the bed after leveling. The guide said to level it, not get the bed below the extruder. Is that what I missed?
https://docs.ratrig.com/commissioning-guides/v-core-4#build-plate-preparation
you have to touch it before doing anything else!
There's no need to be patronizing here - I'm not a dumbass. Here's the text: Proceed to manually adjust each lead screw to get the bed roughly levelled, it doesn't need to be perfect, the machine will perfect it later.
Move the toolhead to each Z position and adjust up or down as required
The machine will perfect it later the operative term, eh?
No need to ne patronizing, neither to repeat myself 4 times! I'm sorry, I'm just trying to tell you what to do.
step 1 tells you to make the bed kinda leveled in order to later (the system) perform the z_tilt by itself.
If it scratched the bed when doing the z_tilt it only means it was not straight enough
So start again, but get the bed well below the nozzle? Is that the mistake I made?
no....
before doing anything else, send M84 to disable the bed screws
I used a 4 foot level on 3 axies I don't know how to get it any more level than that. Build surface is solid. Yes, I sent the M84 and manipulated the screws to level the bed. The nozzle was well above the bed at that point
then, by hand, rotate the bed screws to level the bed kinda close. move the printhead to near the 3 screws and use the printhead as a reference. Either with a ruller or something with a fixed size to make the 3 points close to each other
forget the level... The level tells you the bed is leveled. Not trammed with the frame. The refference is always the frame, not a level.
let me make a video, perhaps I'm not clear (not native)
Ok, now we're getting somewhere. Leave the M84 command active before proceeding to the z-level cal?
when you send M84 to disable the steppers, it will only enable them again if you home/move any of the axis
Ok, understood. So M84, move toolhead to each of the screws and adjust level such that the delta at each screw is roughly the same, and then proceed to the Z-test?
exactly!
then, BEACON_INITIAL_CALIBRATION
Awesome, thanks for the explanation, and apologies for the terse reply. It's been a long road to get to this point!
Step 2
no problem.
FYI - the video doesn't play.
freakin discord
if I upload directly it happens every time
now 😉
Thank you!
you welcome
anything bellow 5mm should be leveled enough
Beautiful explanation - you should add it to the Commissioning guide. Thank you again!
Probably the guide should be a bit more clear in some parts
I thought it was clear, but that's what I get for thinking...
And I try to make it as clear as possible with my english level. Sometimes words can’t replace a vid or I lack some vocabulary
There is a big discussion for a loooong time about the therm leveling… It can introduce a lot of confusion. I would rather use tramming with the frame. But it is what it is.
I'm happy finally having a running machine. Just followed through with the 3 points - we'll see where it goes.
Managed to complete the Z-tilt without scarring the bed. Is it normal for the scan to only encompass about a third of the bed? Toolhead is only traversing 1/3 of the bed, and I'm getting "missing clusters"
Is it before a print or when calibrating the mesh on mainsail?
I've not made it out of the Commissioning guide yet. It went off the bed while calibrating. I've been manipulating the default (configurator) Klipper values for x and y and I think I'm getting close. I'd assumed that the x/y min and endstop values were standard for the printer you selected in the configurator. I've changed the values such that when I home x and y the positions are listed as 200, 200. When I enter 0 for a position for either I'm almost to where the toolhead gets to the front left corner of the bed per Klipper guidance.
I hit emergency stop here
With the original Klipper parameters the first photo shows the position of the printhead after homing x and y individually. The nozzle is right around 205 mm from the front edge of the bed, but it's only about 85 mm from the left side of the bed. Mainsail is telling me that the position of the nozzle is x=200 and y=200. Sending x and y to 0,0 individually via Mainsail, this is the where the printhead is at 0,0. At this stage I'll run with any thoughts or suggestions you might have to get the printhead over the left corner at 0,0. Would this be a likely explanation of why the printhead goes off the right side of the bed during the Z-tilt calibration?
Can you share ratos-debug.zip with the current configs?
Certainly - let me go pull it up.
Appreciate you taking a look.
Adding to the confusion here - x location to a decimal place when the Klipper axis values are all whole numbers? There's something really flaky going on with control on the x axis. Actual location of the toolhead is roughly half of the Mainsail-posted value. The Y-values seem to be pretty close whether homing, or sending it to the origin. Is there a multiplier in the software that needs to be nixed?
Homing positions are directly dependent of the endstop_position and position_min
If both are whole numbers how do you end up with a position to a tenth? Is it hitting the endstop and then gauging the distance based on turns of the pulley?
Checked movement of the toolhead this morning. Issued a Mainsail command to move on the x axis by 50 mm - It moved 25. Still using the originally configured Klipper axial constraints. What would you think the best place to start troubleshooting? The motors are to be responding to direction correctly, so would you think this is a mis-applied software multiplyer, or more like a miswired stepper motor jumper?
should not be half using the stock configs.
Check the pulleys if they are 20 or 16 teeth
this is not a wiring issue, for sure. Also, config is working correctly on everyone else. I will check your configs soon
I had that thought early on. The bags say 20T, but I guess I need to actually count.
just measure the diametre over the teeth
As best I can get my calipers in between the supports the left pulley ~ 12 mm
12.25
Sounds correct
So is there something else that would cause that stepper to halve it's distance that I can change?
Mr./Ms. Pereira - I am somewhat pleased to inform you that this situation was directly attributable to a condition known as "short between the headsets". There is some confusion in Section 12 of the build guide whether to set the jumpers for "SPI" or "UART" mode - the board photos in the guide show both. I reversed my initial config from SPI to UART and two things occurred: 1) x and y move independently; and 2) the toolhead homes to 200, 200 and straps out to the same off the bed. Sending the head to 0,0 puts the toolhead in the front left corner as it should be.
Thanks for your time and patience - onward through the fog!
Hello
I just checked the guide. Step 18 of section 12 tells clearly to set them to UART, Inserting 1 jumper, and shows where it should be… I’m not sure where you found both infos.
But I’m glad it is solved
Look at the photo of the board on the next step - Step 19. Do I follow the text or the photo - if you're a nube (rube) like me you go with the photo. C'est la vie...
in the previous step says very clear to use 1 jumper. Exactly the part I marked...
Why would you need to set them first to UART and in the next step change to SPI?
AND.... this is the first line of step 19!!!
But, ok! I will make yet another board image without the jumpers just for step 19
Or to use the same image from step 18
@Cruz can you use the step 18 image in step 19 of electronics guide? Looks like it is creating some confusion
Or if you prefer, I can edit the CAD file, remove the unwanted jumpers and export it
will take care of it