Never adjust your 3D Printer Z Offset again with Klipper and a plugin 

And thanks for the kind words. I wish you a great build!

Couldn't agree more! Thank you!

Biggest improvement I've installed on this printer so far. You will be happy. Thanks!

Glad I could help! Thank you for watching!

hi, can i ask what step you missed? i get a lot off errors taht i can all solve, because entry's where not filled in the autoz.cfg file or just were examples. After editting al errors out, now i get a message "No trigger on probe after full movement" in calibration step. Did you get the same error? how did you solve it? Thanks!

..and done on one printer, that was way less painful than I expected. To make it a bit easier, I even added calibrate_z at the end of my quad gantry leveling macro, just before the probe_dock ('m using the klicky probe, not euclid). Thank you for the video!

Wow, that was fast! If you get the chance, please thank Protoloft the creator of this code with a coffee. I agree with you, this was one of the few manual things left on my printer. I love it. I've been able to knock out prints since creating this video with a huge increase in consistency compared to my Voron prior to this change and also compared to my previous printer. I'm glad you found this useful and thank you for watching!

Thank you very much! Be sure to thank the creator of the software!

It's cool to know that others like the color! I'm glad you found it helpful and thank you for watching!

Glad it helped! Thank you!

Glad it helped! Thank you very much for watching and the kind words!

It's great that it's working for you. I have an older printer that isn't enclosed, I may give it a try. Thank you for watching!

Long time BLtouch user here, on multiple machines. You still need to set up your Z offset with the Bltouch. It's not hard, but it's not foolproof. Also over time it needs readjusted (especially after a pin bend or replacement). I always forget which way to go to do the adjustment so it's always a process for me. This macro eliminates that process. Also If you ever get gunk in your sensor pin (like a fan blowing fuzz from regular printing) the sensor will false trigger or hang up. A warped print can take out the sensor pin. I've ruined hundreds of dollars worth of bed surfaces with Bltouch malfunctions over the years. The Bltouch works great, until it doesn't. I really like using them when they work, and have bought a bunch of them (always authentic). But when they fail it's a big problem. I've ruined hundreds of dollars worth of bed surfaces with Bltouch malfunctions over the years, plus bent pins, and other crash related damages. The biggest advantage here is getting Bltouch results plus nozzle measuring without actually using a BLtouch.

Glad you like them! Thank you for the nice comments!

So glad it helps!

I did not consider that. Thank you for posting the solution!

That is a great point. I'm going to go back and measure the OT as you suggest here. Thank you very much for the tip. Thank you very much for watching!

I used the similar Euclid probe. It worked great. Thank you for watching and the comments.

Glad it was helpful!

I've been thinking about the same thing for the last few weeks. Sometime soon, I'm going to sit down and clean this all up. I'm planning some future changes for the printer and I think it will be the right time to optimize that aspect as well. Even as it exists right now, I'm so happy with the quality of the prints and the consistency, I'm going to enjoy it just a little longer before I start changing things... Thank you very much for watching and the great feedback!

[gcode_macro G32] gcode: CG28 ATTACH_PROBE QUAD_GANTRY_LEVEL calibrate_z DOCK_PROBE G28 Z

@@stuttgartbatman3788 Thanks! I will try that out ASAP. And Grüße vom Ostendplatz.

@@stuttgartbatman3788 Yes, Thank you!

@@timha4102 so ein Zufall 😁 was für einen Drucker hast du?

I find that as long as the nozzle is hot, and I brush it before the start, I don't seem to have problems with the z-switch. Or at least I don't think so. I was having many issues with the induction probes which may have been hiding a problem with the z-switch. So I suppose I will see soon enough. Do you adjust z-offset by hand? Thank you very much for the kind words and watching!

@@KapmansBasementWorkshop I adjust z-offset by hand and apply the result to the probe. I never do it again until I switch nozzle. And I rarely do. I used to brush it before the start but even then, it would sometimes be a problem with the Home being done on the nozzle. Like you, I don't have a decontamination brush (yet) so brushing is at the beginning of the process. As it stands now, I can start a print and walk away. Oozing at the end of the nozzle will get purged but will not mess up the z-offset. The only concern I have (and still need to test) is the variation in z-offset due to printing at various temperature. In theory, the nozzle is going to be higher (less expansion) when I print PLA vs ABS. 45 deg difference could shift the z-offset by 0.01 mm ish (still have to do the math). To be tested.

@@eridum That brush and purge bucket are high on my list. Very soon. I'm going to be experimenting with heat soak time to see if the Euclid probe has reduced the time needed as well. I was waiting up to 2 hours for the induction probes to stabilize. If I get an hour or more back, I will be thrilled with all of this. That said, I would think this modification would take care of the z-offset changes due to temperature changes. I wonder if it can be made to work without the z-endstop switch...Thank you again.

I'm sure many will find this helpful. Thank you for posting!

@@KapmansBasementWorkshop In order not to leave the others amid the wilderness here are some follow-up thoughts. Since Protoloft's equation is a linear one and the switch_offset constant gets used once only, it may also be used for KlickyPCB without hassle. My setup ended up with (autoz.cfg): [z_calibration] switch_offset: 3.25 It may occur 3.3 or 3.4 would prove to be fine for some. As per QGL and bed mesh (printer.cfg), my findings for 350mm bed are as follows, see ## notes: [probe] y_offset: 19.0 ## KlickyPCB distance tip-to-nozzle, towards back (Y+); was 25.0 (default) [quad_gantry_level] horizontal_move_z: 10 ## default 5 is too liitle, KlickyPCB skids/glides on the surface during move [bed_mesh] speed: 120 horizontal_move_z: 10 ## default 5 is too liitle, KlickyPCB skids/glides on the surface during move mesh_min: 35, 25 ## KlickyPCB position, not nozzle! 25 - probe.y_offset: 19.0 = 6 => axis goes to Y6 mesh_max: 315, 293 ## KlickyPCB position, not nozzle! 293 - probe.y_offset: 19.0 = 274 => axis goes to Y274 probe_count: 5, 3 Bottom line: Compliments and thanks to Protoloft and you, Kapman, to make and visualize the AutoZ idea. Helped me a lot during the build, though it was confusing (to) me at first due to the application of KlickyPCB instead of Euclid or KlickyNG probes. Great channel of yours, thx👌.

@@rozsikacsipke9880 Thank you for posting this and thank you very much for the kind words!

I don't think it will work with the induction probe for 2 reasons: 1. The induction probe does not measure the surface of the bed. It can only sense ferrous metals. So the induction probe measures the height of the steel flex plate under the PEI instead and it tells you nothing of the actual height of the surface. I have 3 flex plates, each one has a slightly different thickness of PEI. And if you have a plate with a rough side and a smooth side, each side has a very different thickness! By using klicky or Euclid, you will always get the surface location and can switch plates or flip sides and always get the correct a-offset! 2. The induction probe will either be a model that works at 7mm or 5mm distance. If you try to tap that against the z end-stop, your nozzle will crash into the bed.

It depends on how accurate your printer is. I believe I used the 0.2 model. But you can try it, and if it does not work, you can try one of the other sizes based on high tightly (or not) the parts fit into the jig. I admit, it is a bit confusing. If your prints tend to be oversized based on the design dimensions when compared to the actual print, you can target the larger jigs. Ideally it will fit perfectly.

Thank you for that and thanks for watching!

max_deviation simply is the allowable range in mm of the calculated offset. So if you want to allow 1.528, the max_deviation needs to increase to 1.6 or 2. Before you do this, adjust with the paper test and try to get it within 1mm. It's a safety mechanism. I'm glad the video has helped and thank you for the kind words!

@@KapmansBasementWorkshop okay so I think I got it. I have a glass bed but then put a 1mm sticker on it due to some issues and that caused the variance from when I originally set it up. So for instance correct me if I am wrong but if I use a flex plate and do the initial calibration with this plate and then use something that is 3mm thicker I think I would need the Max Deviation to be more than 3mm. Correct?

That might have been due to the g90 and I did not restart first. I think that might have been the issue but have to double check.

Assuming I understand what you experienced, I think that depends on how you have klicky or Euclid configured when homing, correct?

I would place it after both. Protoloft states this: "It's important to calibrate Z correctly for the created mesh!". Also, I did not create this macro, Protoloft did. If you find it useful, please send him a cup of coffee, the link is on his github. Thank you for watching!

@@KapmansBasementWorkshop thanks :) Your videos are so helpful, keep up the good work!

This change is the single biggest improvement on my printer. I recommend it strongly. The consistency is amazing and it simply adjusts itself, even when changing flex plates that are different thicknesses. One question: Is the BLTouch having trouble in the enclosed & heated printer like many claim, or is it working well for you?

Yes, it can. Have a look here: github.com/hawkeyexp/auto_offset_z Thanks!

It has it's own dedicated connection to the motherboard just as any of the other end stop switches do. Thanks!

The reason for not using the BLTouch is that it tends to fail due to the high heat inside the enclosed Voron 3d Printer. What is good for printing ABS seems to be bad for BLTouch. Also, the way the Voron z endstop switch is, you can change nozzles and it automatically adjusts so you don't need to calibrate a new z-offset.

@@KapmansBasementWorkshop thanks for the explanation, does it also apply to the version of bltouch with a metal pin? I mean, does the housing give up or the pin or is it electronics, thank you!

@@artemnikitin2806 My apologies, but I just don't know the answer to that one.

This is great to hear. And thank you for posting the solution!

Good question. You can save the settings, but I choose not to. My reasoning? When I start my first print of the day, I don't heat soak my printer. As soon as the bed is hot enough, the printer homes, QGL, home and then calculates the z-offset and the print starts. But as the printer frame, gantry and linear guides absorb the heat, expand throughout the day, the dimensions of the printer change slightly. So I found that running the process again for the next print generates more consistent 1st layers because a new z-offset is generated for each print. So for me, it did not make sense to save those settings. And since the z-offset is calibrated automatically, I'm not doing to work, the printer is doing it for me. True, it takes a little longer to start a print, but for me, that's way better than heat soaking the printer for 30 to 60 mins prior to the 1st print and it gets me a perfect 1st layer each print. Thank you for the kind words and watching!

I believe you can just use the "SAVE_CONFIG" command. The printer will reboot and you will need to run the auto-z again. But it should be fine from there. One comment. While this is useful for switching bed plates, the reason why I run this at the start of each print is to compensate for thermal expansion. Since installing this, I do not heat soak the printer. When cold, I start a print and as soon as the bed and nozzle are at the correct temps, it automatically calibrates and starts printing. Once done, the printer cools down a bit, I prep the bed for the next print and it now auto calibrates as the printer is no longer cold, but warm and the frame and bed might have changed slightly due to the temp changes. I get very consistent 1st layers since doing this, without heat soak. The net effect is that calibrating prior to each print actually saves time! Thanks for watching and the question!

Great to hear! I'm glad it helped. I do plan on more mods. Quite a few in fact. I expect to start work on a purge bucket relatively soon. I have a few others I will complete 1st. Hopefully some of the others I have created might help out in as well. Thank you very much for watching and the great feedback!

Which purge bucket design are you looking at?

@@KapmansBasementWorkshop really anything that also works with trident, if you don’t have one all good watching a install on a 2.4 will still help

@@GREGELZZ I would think that 90% of the time, what works for the 2.4 should work for the trident. I believe the next 3 to 5 projects should work just fine with the trident.

It sure can be used that way. It works great. Especially if you are moving from an induction probe that requires warm up time. This requires none of that. Thanks for watching!

I see no reason why you can't. I would certainly try it. It just comes down to the physical limits of your z axis.

I think you can. As long as you have klipper, but you also need that nozzle end stop as well as either a Euclid probe or Klicky.

I have my m401 & m402 defined in a file called euclid.cfg and then include that file name at the top of my printer.cfg file. If you wish, you could define m401 & m402 directly in your printer.cfg file. You should be able to do the same with Klicky. If they are already defined, and you want them called automatically, you can add them to the QGL if that is being called prior to each print in your print start. So something like this: [gcode_macro QUAD_GANTRY_LEVEL] rename_existing: _QUAD_GANTRY_LEVEL_ORIGINAL gcode: M401 ; deploy Euclid Probe if needed _QUAD_GANTRY_LEVEL_ORIGINAL ; check bed level M402 ; dock Euclid Probe However, I would assume you have m401 & m402 configured correctly previous to installing the auto z calibrate.... Let me know if I'm even getting close to the answer you are looking for. Also, for M401 & m402, have a look at my video about installing the euclid probe.

@@KapmansBasementWorkshop to be honest i'm staring with a 2.4 r2 build. i got most of the parts in, its going to be a 300 x 300 build. so i'm gathering more info about the probe and some other problems that this printer has. After watching your video i'm going to skip the induction probe and ordering the euclid probe. That's why my question was how and where to put the commands i have a core 3 from ratrig now i want a printer that can print abs And to be honest kind a new with klipper still learning and sometimes a little help is very appreciated.

@@petarbajai In that case, you may find these two interesting. They represent some key things I learned building this printer. For Euclid: ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-Z7x4S_HMlvc.html For Klipper and other software: ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-XDQji9IFZ18.html When placing the magnet to the bed, use the flattest surface you have. Hopefully thick glass and weight with and overnight cure: ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-xarmgigvHrU.html And this is an absolute must once software is installed and printer is running, the klipper input shaper will provide a huge increase in print quality at high print speeds: ru-vid.com/video/%D0%B2%D0%B8%D0%B4%D0%B5%D0%BE-hXgl9-lXn7Y.html Some things I learned along the way and I hope they will answer most of your questions. Thank you for watching!

@@KapmansBasementWorkshop For Klicky probe change: M401 with Attach_Probe, M402 with Dock_Probe

I don't save it because I run this prior to starting each new print. But you can certainly choose to do so. Thanks for watching?

Yes, that can be changed. I used the Euclid instead of the klicky, but have a look at your m401 macro. The z height is set there.

I believe that is what you would need to do. Here are the requirements of this modification: github.com/protoloft/klipper_z_calibration/wiki/Requirements

So don't take this as known facts, it's more superstition.... I have heard conflicting stories about QGL clearing the bed mesh prior to actually executing the QGL. I have also heard stories about Klipper not automatically loading a bed mesh after a QGL runs. I have heard similar stories about G32. Also, the Voron secondary tuning instructions where it states that it's probably a good idea to run a bed mesh calibrate in your print start prior to each and every print states this: "Then, place BED_MESH_CALIBRATE in your PRINT_START macro, AFTER any homing and leveling routines like G28, G32, QUAD_GANTRY_LEVEL, and Z_TILT_ADJUST." That's not the same as simply loading a bed mesh, but it's similar. So, does that backup the statement that QGL or G32 clear the bed mesh and don't reload it? I don't know. This (at the time) seemed like a sure way to be sure while being lazy and not testing the theory to be sure. I wish I had a better answer for you. It's a good question and thank you for asking and watching!

Thank you for watching and glad it's working!

I don't think so, unless you change the auto-z calibrate code itself? The config file simply asks for the location in x & y. It doesn't really give you those options.

If you go into your euclid.cfg file, just change the z height to a value that works for you. If you need a more complicated path you can add that gcode there as well. This will be the macros for M401 & M402. Both of them set the z height. You should be able to adjust enough to pass over the z end stop. For the macro, to call it, you should probably modify the the G32 macro or the start_print macro in your printer.cfg file. In G32 I have it set to do a home, qgl, home and then calibrate_z.

@@KapmansBasementWorkshop got it thanks. Any idea why I have to use error_if _probe_deployed first before I can use Euclid probe?

@@nicklong9985 I sure do. Early on I had not adjusted it very well and it would sometimes drop the probe. As soon as it notices the probe is not fully connected, it raises an error and stops. If it did not do that, it would probably destroy your bed, print head or both when doing a bed leveling. Trust me, you want that error handling code. I can think of 4 different occasions where it would have been very bad if it were not there. Include the code!

@@KapmansBasementWorkshop yes I want to use the code. I’m probably not explaining the problem very well. What I wanted to know is what is causing the error?

@@nicklong9985 If at any time the probe is supposed to be attached and it senses both magnets are not making full contact to the Euclid, it will raise that error.

If you don't want to use Euclid, you could use the Klicky probe instead. Also, someone seems to have gotten BLTouch working right here: github.com/hawkeyexp/auto_offset_z But I can't use a BLTouch due to heat issues, so I have not tested this. But you can certainly give it a try.

You could but what would you gain by doing so? It would just be a more complicated way of doing what your already doing.

I believe in one of the other comments above I posted a link for someone that did this for th cltouch. That might work.

It's great question. I tried to clarify that since I didn't think the instructions were that clear, and I was confused at first as well. In your autoz.cfg file there is the switch_offset value. The starting value should be the value taken from the specs of the microswitch you used. However, this is probably not the final value you want. You will test that switch_offset value with the sheet of paper. You need to keep testing and adjusting switch_offset (and rebooting klipper) until the paper friction feels right (or feeler guage) and you have the nozzle at the correct height for your 1st layer of a print. One you have proven the switch_offset is correct based on the paper test and possibly some minor adjustments based on an actual print. You are done and you will never need to do it again. You can change nozzles, switch bed plates, print at different bed temperatures and the printer will now automatically adjust itself to the correct height! Honestly, its by far the best improvement I have made to my printer. Also, I strongly recommend that you install a nozzle brush (I created a video as well) and setup the printer to brush the hot nozzle prior to calibrating the z-offset. That way you won't have filament on the end of the nozzle that might cause an incorrect reading with auto-z.

@@KapmansBasementWorkshop Okay, that makes sense! I'm doing that mod next and will be following along your video. Then the runout sensor, so on and so forth ha

I purchased it from here: preciseprinterparts.com/Printer-Beds_c_11.html I highly recommend it.

I've got a very different start macro these days. And I've been testing other features, so my newer one does not apply. Have you taken a look at: github.com/Kapman1/Kapman1/blob/main/printer.cfg and github.com/Kapman1/Kapman1/blob/main/autoz.cfg or even better reviewed the latest from the creator of the auto-calibrarion software? github.com/protoloft/klipper_z_calibration

I've heard that some have had good luck with that probe. I've got no experience with it....yet. Thanks!

There is complex code in the prusa to compensates for the temperature variations that effect the pinda probe. That is why it works so well. The pinda probe isn't the magic. The code is. Same as with this style probe.

I used this: mods.vorondesign.com/details/t1DBVlcUBbdEK6habEsVzg You can also purchase kits with all the parts if you need to.

I love it as well. It's soooo cool that you got this working with the CRTouch! I may need to try that with my other printer! I wish I thought of it. Thank you for posting back and thank you for watching!

@@KapmansBasementWorkshop Thanks for the great video! It was very helpful in setting z_calibration up on the V2.4. I was going to put a klicky probe on the custom build but the mount they had for the EVA wouldn't work with my hotend, it was lower than the nozzle. Since I already had the CRTouch on that carriage, I wanted to see if I could use it for z_calibration. I printed out a nozzle stop like the one the V2.4 uses for use as the z endstop and tried without any modifications but it just wouldn't work. I got to thinking and came up with the adapter that allows the nozzle to contact the pin and the metal pin of the CRTouch to contact higher, fooling z_calibration about the "switch" being lower than the nozzle, increased the switch_offset and it worked beautifully. Here's the adapter I came up with, feel free to use it or not, however you'd like to: social.thangs.com/m/65919 It should be tight on a 5mm pin, tap it gently until it hits the inside stop. I made it fairly tight so that with the built in stop, it won't move again after installed. Play with switch_offset to dial in the correct height after running z_calibration. I was working on the printer bed and had to dial mine in again, it's currently 2.04, so a much bigger offset than with a microswitch but it works.

Mark, I`d be interested to see how you mounted this on with the CR touch I`m running a BLTouch on an Anycubic. Just wondering how I could adapt this for it.

@@goodmandavis I'll check and see if I still have the files/photos of what I did. I lost a lot of videos and photos recently due to a corrupted drive and personal failure in backing that drive up so I'm not sure I still have anything so no promises. I've made changes to that printer since and don't use the CR Touch anymore, changed the hot end and modified the carriage which allowed me to add an annex engineering probe.

It sounds like you might not have the correct coordinates. In your cfg file you should have something like this where you set the positions. These were my settings. Yours will be different depending on your measurements as shown in the video. Even if your bed is 350mmx350mm like mine, they probably won't match exactly since our beds are not exactly bolted down in the same place. [z_calibration] nozzle_xy_position: 227,350 ; A X, Y coordinate (e.g. 100,100) of the nozzle, clicking on the Z endstop. switch_xy_position: 232,328 ; A X, Y coordinate (e.g. 100,100) of the probe's switch body, clicking on the Z endstop. bed_xy_position: 170,170 ;default from relative_reference_index of bed_mesh switch_offset: 0.519 #...the rest of your file follows....

@@KapmansBasementWorkshop Mine are different. But everything looks the same except the values: UPDATE: I figured it out 😀 [z_calibration] I commented out this line because it gave me the error "Unable to parse option 'clearance' in section 'z_calibration'" clearance: 5 * z_offset from the "probe:z_offset" section The line is needed but needs the ## after the value, so: clearance: 5 * ## z_offset from the "probe:z_offset" section I don't know why they don't have it written that way in the documents but either way I'm one step closer to having a fully functioning printer equivalent to my Prusa MK3. Many thanks for your help.

@@Hopeinformer I'm glad it's working now. Thanks for watching!

@@KapmansBasementWorkshop Do you have a venmo or Zelle? I'd like to send you a little thank you.

@@Hopeinformer I truly appreciate the gesture. But I don't have either. And even if I did, I would ask you to send the little "thank you" to the creator of the Klipper plugin Protoloft: ko-fi.com/protoloft

My guess is the micro switch in the z ends top might need to be replaced. It could also be the klicky micro switch. I have had to replace the z endstop switch multiple times…

hi, a question, did the rest of the setup went well? I have a Klicky and had to change quit a lot in the autoz.cfg before it was error free. Now at calibrate "No trigger on probe after full movement" and process stops. Did you not get this error? Thanks!

@@richardtepas4832 That message usually comes up when the microswitch isn't being sensed properly. It can be the sign of a continuity problem with the wiring of the wrong pins are configured.

@@KapmansBasementWorkshop ok thanks, into that. Klicky works flawless, so i think it might be something else. Did you only have to change 2 or 3 settings in the config? Because i had to change a lot more

@@richardtepas4832 Most of my changes were in the cfg file for the automatic z calibration. I have my files in github. The video description contains the link to those files if you wish to see them.

Thank you! Have you had a chance to step through this: github.com/protoloft/klipper_z_calibration/wiki/How-To-Configure-It On this printer, over time I've moved into a very different direction as I continue to try different tools. I've got no limit switches anymore. I have plans in the future to get back to this on a new printer I'm building, but it probably won't happen for several months at least.

Now that you mention it, I should have weighed both options! Thank you for watching.

Good question. You can set this up so it saves the calibration or you can do that manually. However, I have chosen to not save the calibration. I have the printer re calibrate at the start of every print job so there is no need to save. By running prior to each print, minor adjustments are automatically made for slight changes in the printer due to heat expansion, etc. I've added it to the G32 macro which is also called from my start print macro. My print surface heights are rock solid consistent now with every print. I will post my g32 macro this weekend in the video notes. Also, for this to work at it's best, I would install a nozzle brush so the nozzle is cleaned automatically just prior to the auto-z offset calibration to make sure no filament has oozed that might impact the auto-z calculation. I have a video for that as well. Thank you very much for the kind words and for watching!

@@KapmansBasementWorkshop TYVM, I like the G32 idea

@@shanee7430 It's been great, I highly recommend it!

​@@shanee7430 The code: [gcode_macro G32] gcode: BED_MESH_CLEAR G28 QUAD_GANTRY_LEVEL CLEAN_NOZZLE G28 BED_MESH_PROFILE LOAD="default" CALIBRATE_Z

@@KapmansBasementWorkshop I assume it would be safe to just omit the CLEAN_NOZZLE if I don't have mine setup yet. Crazy as it seems, I can't find a brass brush around town.

Have a look at this page: github.com/protoloft/klipper_z_calibration/wiki/How-To-Test-It This should get you on your way. Once you have adjusted that offset, you should be fine.

@@KapmansBasementWorkshop Been through it a couple times. It is very close, but my problem is when I do baby step it, I am not sure if I should be applying that to Z end stop, or to probe offset, or to switch offset. and I think I just keep over tweaking it. But I have also reset and restarted tuning it 3 times. Same results every time. Ended up just going with Klicky as Z endstop, and it works a treat. Thanks for the reply though, i appreciate you taking the time.

Good questions. The answer is it depends on how you want to operate. I choose not to save it and I calibrate prior to each print which accounts for anything that might of changed position due to heat expansion, cooling, change in PEI bed thickness when changing the plate between prints that may be different thicknesses or textures. If none of that is a problem for you, save the value after calibrating when 1st turning on the printer if you wish to use the same setting for a longer period of time. Even without saving, the measured setting with stay in effect until klipper restarts or you run another calibration.

@@KapmansBasementWorkshop Could you show us your Print_start macro please?

@@EliteNervous This is a possible sequence recommended by the script creator: The print start sequence could look like this: Home all axes Heat up the bed and nozzle (and chamber) Get probe, make QGL or Z-Tilt, park probe Purge and clean the nozzle if available (Get probe), CALIBRATE_Z, (park probe) (Adjust Z offset if needed) Print intro line Start printing... As for mine, I have not added it in yet. But here it is just in case you find it useful. I might end up placing it in the G32 macro instead which is called from my print start: [gcode_macro G32] gcode: BED_MESH_CLEAR G28 QUAD_GANTRY_LEVEL G28 BED_MESH_PROFILE LOAD="default" G0 X175 Y175 Z30 F3600 #-------------------------------------------------------------------- [gcode_macro PRINT_START] gcode: G90 ; use absolute coordinates G32 ; home all axes G1 Z20 F3000 ;move nozzle away from bed G1 E5.0 F3000 ; move extruder I'll post again once I have actually added it in and tested it.

@@KapmansBasementWorkshop thank you

Neither. I just changed the switch_offset setting in the autoz.cfg file based on the paper test and that's it. The software will automatically adjust to that height no matter the temperature or even if you switch bed thickness. It just works.

@@KapmansBasementWorkshop So you are saying that I should change the switch_offset to -9.00 (which is the z offset when I have the nozzle down to where it should be for printing)? Are you sure? I am using klicky pcb with z endstop (standard voron one that you have)

It might, but I have never used a BLTouch so I have not tested it.

I would. I also home after the QGL prior to using this script to make sure z is at zero after any adjustments QGL made. I have this believe that QGL changes the locations of z to possibly be non-zero. So I execute the extra home after qgl. I'll need to run some tests at some point to see if that is true. If you or anyone watching has a better understanding of this, please correct me.

@@KapmansBasementWorkshop I also do a g28 after qgl before I print, and then load the mesh

@@TORDesign So when you do that G28, does it ever seem like 0,0,0 moved slightly from where you expected it to be? But the difference might be too hard to see....

@@KapmansBasementWorkshop I’ll keep an eye on that thanks.

Sure. Here are all my cfg files: github.com/Kapman1/Kapman1 Thank you!

Thanks :) it turned out that I had a problem with the probe, poor conductivity between magnets

@@MrRoldak Interesting. I would not have guessed that. Thank you for posting this!

It sounds like you might have a typo. Typically, that causes a parsing error.

That is odd. I agree, it should not matter. Do you think the bed or the gantry might not be flat or straight? Is there a loose part or screw anywhere?

I've got 3 that range in size from 0,8mm to 1.4mm. That has a substantial impact on how my prints look (or even print) if i don't re calibrate. And the paper method for me is a little too subjective. So now I get consistency. Thanks for watching!

There are at least 3 Z_Offsets (or even more) in klipper, which you use with Stock Voron 2.4 setup. There is in the [stepper_z] section position_endstop, which determines how far the nozzle is away from the bed (positive value) or below the bed (negative value) when the endstop pin switch triggers. Then there is in [probe] section z_offset which determines the difference between probe triggering and the nozzle. My probe e.g. triggers when the nozzle is 0.710mm above the bed, which you "meassure" or calibrate with probe_calibrate (whereafter it moves to the automatic section at the end). For a stock voron V2.4 this value is in prinziple not relevant, as it will use "mesh_relative" and rely on the position_endstop. And then there is the Offset Z which u can adjust during printing via the display or via the set_gcode_offset z_adjust=x.xxx commandiot to come closer to the bed or stay more far away.. And sadly the intention of voron in this design doesn't work as well as they thought. In principle the system they use should be perfect, as the nozzle tips the pin and independent what changes the switching point should have a permanent hight over or under the bed.. But there is where thermal deformation comes in.. the bed will bend with temperature, but really not much. The hight of the bed may change if u use different printplates. But mainly the temperature inside the closed chamber will destroxy the system. The x axis will bend a bit, which means as the pin is not at x center, the carriage will be on another hight, depending how warm your printer is already, relative to the position measured. And even much more and destroying the solution, the z endstop pin will grow, and it's really much it grows, enough, to destroy your first layer at about every 10K more chamber or "pin" temperature. And as the pin is so close to the bed it will even warm up higher than chamber temp.

You don't need to use this method when switching bed surfaces, only advantage is that this also takes care for different nozzle height for when you switch nozzles..

Linux tips here: "cd ~" will change to your home directory. if you want to change to a specific users home, "cd ~user" and if you have SSH, you generally have the ability to copy files with SCP... "scp user@hotsname:/file/from /file/to" or "scp local/file/from user@hostname:/remote/file/to" where the remote path starts from ~ unless you put an absolute path starting with /

@@KapmansBasementWorkshopdoes Klipper support bed profiles? Just setting up a 2.4r2 350. But on say my Prusa I have a smooth and textured profile, once I got Z dialed in on those all I have to do is switch the bed on the screen.

I don't think so. Beacon has it's own software that must be used.

Yes, You are setting the z-offset one last time manually. From that point on, the software takes over so you don't need to do it any more. Thanks!

I think it's a great question! The z-endstop switch on Vorons sits right next the the bed plate and is in a position where the printer places the actual nozzle on the z-endstop. That is how it homes the z position. Other printers I have used have placed the z endstop elsewhere and have not homed using the actual nozzle tip. It sounds like you might be in that situation. When researching this. I stumbled upon this solution where someone seems to have gotten BLTouch working right here: github.com/hawkeyexp/auto_offset_z. This is not a Voron printer and it is based on this script. This person goes about it slightly differently but to the same effect. Have a look as you might be able to use it and solve this. The other option is to possibly relocate your z-endstop or simply add an additional endstop just for this reason.

@@KapmansBasementWorkshop Thanks for the info. I'll check it out.

Do the nozzle scrubber mod as well.

I do plan on adding this.

While I plan on adding a nozzle brush, I'm reasonably sure that as long as the hotend is at proper temperature for the filament installed, it's typically accurate enough. But I do brush the hot nozzle with a wire brush before calibrating.

When print is finished I do a retract

@@everuss That seems like a great idea!

I believe you need to manually adjust it in the auto.cfg file 1st to get it closer than 1mm. You will need to experiment as shown in the video with the piece of paper. Once you have that done, it should take care of itself in the future.

With my eyes, it's very hard to see any change. I need to use a piece of paper or a feeler gauge to notice a difference. But 0.1mm is usually very noticeable with paper.

It doesn't. But then it's not needed.

I use one of these: mods.vorondesign.com/details/t1DBVlcUBbdEK6habEsVzg you can get a kit if you like for about 5 or 6 dollars. It mounts right next to the bed. In the description I have links to the probe that mounts to the print head.

Someone seems to have gotten BLTouch working right here: github.com/hawkeyexp/auto_offset_z But I can't use a BLTouch due to heat issues, so I have not tested this. But you can certainly give it a try.

Hmmm. Good question. I suppose the reason for a measurement each time is to take into account any changes due to temperature impact on the printer's expansion that may be different between prints or a cool printer. If that switch height were hard coded, would it impact that measurement? I want to try this now, but my printer is working so well that I'm scared to! I've also been thinking about changing the start print macro to eliminate as many steps as possible between QGL and the z-offset calibration. But now every print is rolling off this printer with exactly the same texture on the bottom with no problems or adjustments needed. I'm kind of enjoying this!

@@KapmansBasementWorkshop I guess the mesurement is even better, because one object less that can cause inaccuracy by heating /docking (Probe)

@@Micha-46456 So at some point I'm going to try it. If you beat me to it, please post back. I'm very curious.

@@KapmansBasementWorkshop I will - I recently ordered my voron kit, but it won't be available until the end of May at the earliest.

@@Micha-46456 Have a great build!

Someone seems to have gotten BLTouch working right here: github.com/hawkeyexp/auto_offset_z But I can't use a BLTouch due to heat issues, so I have not tested this. But you can certainly give it a try. If you do, please let us know if this works. Others have asked if this workis with the BL and CL touch. Thank you for watching!

@@KapmansBasementWorkshop Oh wow thank you, now I just need to figure out where to place the endstop with enough clearance for everything on my Ender 3. My Voron envy continues, one day I can have a trident with this as well never have to do any more calibration lol

@@afkafkafk I waited quite a while before I pulled the trigger on mine. I know what you are feeling!

There should be nothing else but eliminating the calls and not including the file at the top of your printer.cfg. I'll have a look and see what I can find.

I'm going to assume you are including the autoz.cfg file in your printer.cfg file. As you make the changes to the switch offset, are you restarting Klipper to be sure the cfg files are all reloaded?

​@@KapmansBasementWorkshop Thank you for your answer. yes i added the include and yes i always click save and restart. I also restarted the Pi in between.

@@ah_banny8478 Apologies for asking the most basic question there, but it happens to everyone...Can change these setting by hand and save them? I wonder if it's rights or possibly a typo? Can you post the cfg files someplace and post a link so I can have a look?

​@@KapmansBasementWorkshop ... Thanks for help, I hope it works ;-)

@@KapmansBasementWorkshop Did you ever get this resolved? I'm having the same problem.

Its really up to you and how you setup the printer. I placed it in the print_start macro so it would happen prior to every print.

@@KapmansBasementWorkshop one more question am I supposed to save the offset to the probe or z stop.

@@TheCyclingConquistador You save the offset from the nozzle to the bed at the correct height you need for the 1st layer. Use the paper test and then do some final adjustments to that value with some test prints. Once complete, you should be set.

@@KapmansBasementWorkshop thanks, I really appreciate your help and the video.

@@TheCyclingConquistador Happy to help.

Do you have a "position_endstop:" that isn't commented in the [stepper_z] section of the printer.cfg file? It should be commented. After changing that, reboot and then try to save your calibration. It should appear at the end of the printer.cfg file in the section with the comment that you should not touch that area.

It has a # next to it & all green is that correct ? It seems when ever I adjust the AutoZ.cfg switch offset figure the end stop (or something ) adds or subtracts that adjustment so I gain no closer z (Trying to close the z to bed more ) ( I know the higher the figure of the switch_offset means closer to the bed ) & my bed mesh map is 6cm up in the air off the bed position_endstop = 0.680 But it would need another +6mm on it but I know putting it here isn’t right ??

@@adrianstealth340 Did you create a file called "auto_z" and set those parameters and include it at the top of your printer.cfg? Have a look here at my files at the time: github.com/Kapman1/Kapman1 Look at the "switch_offset" setting in the auto_z.cfg (or whatever you might have called that file.

@@KapmansBasementWorkshop hi yes I have that file in place , I’ve overcome few problems after updating Klipper recently , the switch_offset in AutoZ.cfg doesn’t adjust anything , I’ve tried a very low figure to hopefully witness the nozzle go way to high but it didn’t do the adjustment , the AutoZ routine works well but it’s having zero effect & not being applied

@@adrianstealth340 My paologies, but I don't really know what is wrong. This has worked for me and worked well.

I have a "euclid.cfg" file setup that is included in my printer.cfg file. In the Euclid.cfg file you should have something like this: [gcode_macro BED_MESH_CALIBRATE] rename_existing: BED_MESH_CALIBRATE_ORIGINAL gcode: M401 ; deploy Euclid Probe if needed BED_MESH_CALIBRATE_ORIGINAL ; check bed level M402 ; dock Euclid Probe So yes, there are 2 of them and the original needs to be renamed and it's called from the new one.

​@@KapmansBasementWorkshop Thank you so much for your help and quick reply. I have Klicky_Bed_Mesh_Calibration.cfg In this file it has [gcode_macro BED_MESH_CALIBRATE] rename_existing: _BED_MESH_CALIBRATE ... I removed the line "rename_existing: _BED_MESH_CALIBRATE" then saved and restarted and the error message went away. I don't know if I fixed it or just created another issue down the line. haha Thank you again for your help.

@@Hopeinformer I believe you created an issue for yourself unless you just don't use a bed mesh. If you don't, you should be ok.

@@KapmansBasementWorkshop Crap. I do want to use bed mesh. But at this point I just want to start printing something/anything haha

@@Hopeinformer Then this should be fine.

For the auto calibration, you do need the second switch on the print head. It should work for you. If you have springs under the bed, you might need to firm those up.

@@KapmansBasementWorkshop can I use the inductive probe on my print head as second switch?

@D3ltoroxp no, you can’t. The inductive probe doesn’t measure the surface of the bed, it measures the steel plate under the surface. You need the micro switch as it actually touches the surface which is what makes this consistent. Even if you switch bed plates.

@@KapmansBasementWorkshop So when i dont use this switch on head, i cant use the plugin for auto z offset calibration ? I will only when i push z offset calibration, that he push the switch in the back of the bed.

@@D3ltoroxp , Yes, you need both switchs. The nozzle switch and the z-offset to check the surface of the bed.

Yes. Also, if it helps, here are all my cfg files: github.com/Kapman1/Kapman1

@@KapmansBasementWorkshop Thanks Sir. I will download them.

@@JohanFasth Please don't use them "as-is". Others have done that and in one case damaged their printer. Look them over for ideas and do what's right for your printer!

@@KapmansBasementWorkshop No, I ment for use them for comparison. There are some things that isn't so clear for a novice. Probably obvious for a long time user of Klipper. I have struggled with bl touch since new ear and it doesn't work. This is not my primary solution.

@@JohanFasth I have not ever used a BLTouch. I probably should spend some time with it. Soon.

My apologies, but I'm sure I understand the problem you are having.

Since you reinstalled Klipper, have you reinstalled Z calibration? My assumption is that it’s no longer installed based on the error above.

The final result should not be 0 as that means the nozzle would be touching the bed.

I have not tested this, but it's worth a try: github.com/hawkeyexp/auto_offset_z

@@KapmansBasementWorkshop Nice one! i'll take a look at that.

It works great with Klicky probe!

So funny thing. I'm setting up a BLTouch on another corexy printer as I write this. This is my 1st time using one. I don't honestly know, but I suspect not. Mainly due to the shape and construction of the BLTouch. But as I gain more experience with BLTouch (it's not fully operation yet), I may look into this next.

@@KapmansBasementWorkshop also the github is , well how to say, Bare now. I do not see the why this, and the detailed how to or any of this stuff (like on your video at 3:13). So Is this even a thing anymore? I would love to have the details the creator made. I see png files but Yes all the setup and explanation etc on the github seem to be gone.

I don't think it does. But when you need to replace it, consider the microswitch option such as Euclid or Klippy. They are cheap and work well. They also work with this software.

Thanks!

If yours works well and you have no issues with it, no need to change it. However, in my case the induction probe took a long time in the heat to stabilize it's readings. Also, it can only measure ferrous metals. So it never can actually measure the actual height of your flexible build plate because it cannot "see" the pei on top, only the metal underneath. And then I have had them melt on me as well. Finally, I could not use this software to auto calibrate the z-offset. For me, this was the single biggest improvement in my day to day printing. I can change bed plates, they can be different thicknesses and the 1st layer height is always perfect. Automatically. And before I needed to wait an hour to 90 minutes for the induction probe to give stable readings in the heat. Now it re calibrates prior to each and every print. It's super nice. Thanks for watching!

@@KapmansBasementWorkshop thanks for the explanation. I do not yet have a bed sensor. I plan on adding one and wanted to know why you did not like the inductive. Thanks for the information.

@@f1hotrod527 It just was not very reliable for me. Thanks very much for the question, comments and watching!

So I always home 1st. Then pick up the klicky probe and use it to level the gantry. Drop off the probe and home again since home probably changed slightly after bed leveling. I've never successfully left the probe on. Thanks for the kind words and watching!

Klipper isn't recognizing 'clearance'. What cfg file is this error being reported?

Waiting on parts....

@@KapmansBasementWorkshop Sounds good.

I had to adjust this several times after installing. Since then, I'm not having any trouble. But, I'll try this weekend with 2 build plates stacked and see what happens. One other thing to note, I have had people tell be that mounting Euclid on the gantry is actually better than the mount I used in the video as z-height does not come into play...

@@KapmansBasementWorkshop I will try the gantry mount.

@@req00 Please let us know how it goes.

@@KapmansBasementWorkshop I installed the gantry mount and it works very well with no issues on docking etc even with autoz. I used the v4 mount from Euclid which has adjustable height but I had to remix the stl using the short gantry as I needed it to be fully docked about 5mm further over on the gantry. Works great now though and any Z height change the gantry being skewed has no effect on docking.

@@req00 This sounds great. I'm glad you were able to fix/improve this. I'm not having trouble, but I think it may be a good idea to replace mine as well sometime over the next few weeks. I've been thinking about this since your report and it just makes sense. When I get some time later today, I'll add a note to the description about this. Thank you for reporting back on this!

I have one that looks like this: [gcode_macro G32] gcode: BED_MESH_CLEAR G28 QUAD_GANTRY_LEVEL CLEAN_NOZZLE G28 BED_MESH_PROFILE LOAD="default" CALIBRATE_Z It worked well. You can see all the CFG files I had here: github.com/Kapman1/Kapman1/tree/main

Have a look on the about tab and follow my link to the git hub.

@@KapmansBasementWorkshop Thank you for the reply I did notice that afterwords. 1 more question. Is there a reason you set the xy position to 170,170 because half of 350 is 175? Changing my switch_offset does not change how high or low my nozzle prints any ideas why?

@@chucktaylor5878 There is no important reason I choose 170,170 vs175,175. I probably was a little too focused on the camera. I'm not sure why switch_offset does not change the height of your prints. I have some other changes to make over the next few days, I will test mine to make sure it still works and it's not a bug or something. I'll get back with my results.

@@KapmansBasementWorkshop I appreciate your response. I just wanted to make sure I wasn't missing something I get a little over technical on things at times.

I struggle with the same issue. I try for lite friction. No matter how much friction you choose, there will be some test print calibration anyway. Howwever, with this software change to Klipper, you should really almost never need to do it again after the 1st time. I had to replace the microswithc a few times, both actually. I ran the paper test again after that change.

I have not tried it. But it might work if you install a nozzle endstop like the Voron 2.4 has which would not be hard to do. The tricky part will be getting the values correct so the flat surface of the CR touch will touch the nozzle endstop and not the CRtouch pin. But it could be possible.