Bed leveling and calibration. Very frustrated.

[pronoyc]

Hello,

I have been trying to get a decent print out of my Rostock MAX V3 for almost over 3 weeks now but I am afraid I've failed miserably. I started off by calibrating my printer using the recommended steps in the assembly guide. First the simple calibration and then the advanced calibration using Mattercontrol software. I couldn't get a decent first layer with that method.

Then I resorted to this method https://www.youtube.com/watch?v=hYPXNmLXRXI

After setting my Z height and then subsequently bringing the printer down to the three points on the bed. I used the paper technique to match the heights. But even after that I haven't had any luck. I've played around with the extruder settings and configured the E steps (using a caliper).

Clearly this hasn't worked as evident from the layers that I printed on different parts of the bed (Sorry I tried to get the best shots possible but with the placement of the light a little glare did sneak in).

So at this point, what am I missing? How do I level the bed properly so that I can get a decent print. I am using Simplify3D and here are the current settings:

layer height: 0.1500 mm
Extruder Temp: 220 C
Bed temp: 65C
Infill: Fast honey comb 20%
Cooling fans: 40
Printing speed: 3000mm/min
Retraction distance: 6.50 mm
Retraction speed: 6600mm/min

[ROvermeyer]

Don't feel alone, you're not the only one with this problem.

[djarmag]

Use three alternating bed clips, having 6 can cause more warping of the bed. Also make sure it is not too tight so watch when you secure it, make sure it doesn't push the bed down too much. I would do g29 three times, then the Delta Calibration. After you save the EEPROM settings I do another g29.
Download

75mmDisk-pt2mmtall.stl

(24.89 KiB) Downloaded 257 times

and test your first layer settings using this file.

Set your first layer to .2mm or .3mm
extruder temp 210 or 215 if your hot end allows it
bed temp 50
0 infill
cooling fans 100 (wont matter until it reaches height set in parameters)
print speed 20 or 30mm/s
retract distance 2mm at 50mm/s (no more than 5mm otherwise it can cause jamming)

Try not to do multiple items until you figure out what is causing the unlevel. Print slow and do one change at a time.

I recommend using Kisslicer

I feel your frustration and I've been back and forth since December but I ended up buying upgrades and they are definitely worth it. FSR probes+mount, E3D V6 hotend, Duet Wifi are what I have installed and shown great success.

[pronoyc]

Use three alternating bed clips, having 6 can cause more warping of the bed. Also make sure it is not too tight so watch when you secure it, make sure it doesn't push the bed down too much. I would do g29 three times, then the Delta Calibration. After you save the EEPROM settings I do another g29.

Thanks for taking the time to reply. When you say Delta calibration, do you mean the youtube video I linked or are you talking about the calibration methods specified in the assembly manual?

Try not to do multiple items until you figure out what is causing the unlevel. Print slow and do one change at a time.

I only did that because I wanted to see which parts of the bed were closer to the nozzle and which were farther. I have an idea on how the bed is tilted, but it's too fine a tuning.

I feel your frustration and I've been back and forth since December but I ended up buying upgrades and they are definitely worth it. FSR probes+mount, E3D V6 hotend, Duet Wifi are what I have installed and shown great success.

I really can't afford any upgrades a.t.m. Having spend a grand on this machine, I don't think it's fair to go for upgrades when the basic stuff isn't being printed right. I got a large volume printer because I had some ideas I wanted to prototype. At this rate, I am spending more time getting the printer to work than I am on my idea. But I will give the calibration another shot. Thanks!

[iKeyborg]

Use three alternating bed clips, having 6 can cause more warping of the bed. Also make sure it is not too tight so watch when you secure it, make sure it doesn't push the bed down too much. I would do g29 three times, then the Delta Calibration. After you save the EEPROM settings I do another g29.

Thanks for taking the time to reply. When you say Delta calibration, do you mean the youtube video I linked or are you talking about the calibration methods specified in the assembly manual?

Try not to do multiple items until you figure out what is causing the unlevel. Print slow and do one change at a time.

I only did that because I wanted to see which parts of the bed were closer to the nozzle and which were farther. I have an idea on how the bed is tilted, but it's too fine a tuning.

I feel your frustration and I've been back and forth since December but I ended up buying upgrades and they are definitely worth it. FSR probes+mount, E3D V6 hotend, Duet Wifi are what I have installed and shown great success.

I really can't afford any upgrades a.t.m. Having spend a grand on this machine, I don't think it's fair to go for upgrades when the basic stuff isn't being printed right. I got a large volume printer because I had some ideas I wanted to prototype. At this rate, I am spending more time getting the printer to work than I am on my idea. But I will give the calibration another shot. Thanks!

I spent first week fighting calibration. It was exactly same problem. Then I loosen all towers, sat them with quite a bit of force, tighten towers and belts and since this time problem gone. I have all stock DIY R3. I have printed few 10 inches (250mm) in diameter models with always perfect first layer. I recently added two filaments upgrade. Everything works great.

[djarmag]

I understand, didn't mean to say that you can't get good results stock, I was just speaking out of what I was experiencing and wanted more. There are plenty of other users that have stock and shown better results to make me eat my words
https://www.seemecnc.com/pages/delta-calibration-wizard is the delta cal I was speaking of (in the assembly manual). The wizard you linked can work, I only used my method and set the parameters differently in MatterControl using Cura as the slicer and set the z-height in there. I use .2 or .25 for the first layer.

Use

200mm_Bed_Level_Aid.STL

(89.93 KiB) Downloaded 268 times

instead to minimize filament usage and it will tell you your high/low spots quicker.
Your image looks more like an extrusion issue rather than bed level. The bottom right of all your skirts are skinny but the squares do not correspond, what is your extrusion multiplier set to and have you calibrated your extruder? https://www.matterhackers.com/articles/ ... r-extruder

I was having inconsistencies which is why I upgraded. The nozzle is larger so printing is slightly faster than the e3d but you can swap the bore size out (something I haven't tried). My nozzle and heater block always accumulated crud and those led to irregularities on my prints and saw the e3d came with a silicone boot to cover. If the rostock had their own cover, I would of gotten that instead of a whole new hot end. But reading forums and e3d was just a clear winner as an upgrade for me. I use a couple exotics like wood and tpu so I definitely benefit from e3d.

Keep posting on forums of your results and hopefully we can help you figure out what is wrong. Loose or angled towers, loose/too tight belts, use only 3 alternating bed clamps, are what I check first as others suggested.

[Jimustanguitar]

[quote="iKeyborg"I spent first week fighting calibration. It was exactly same problem. Then I loosen all towers, sat them with quite a bit of force, tighten towers and belts and since this time problem gone. I have all stock DIY R3. I have printed few 10 inches (250mm) in diameter models with always perfect first layer. I recently added two filaments upgrade. Everything works great.[/quote]

So you're saying that the aluminum towers were at different heights/lengths because of how it was assembled? Interesting... I know a guy with .5m calipers, I'll have to measure the ID wood to wood distance on mine.

[pronoyc]

I understand, didn't mean to say that you can't get good results stock, I was just speaking out of what I was experiencing and wanted more. There are plenty of other users that have stock and shown better results to make me eat my words
https://www.seemecnc.com/pages/delta-calibration-wizard is the delta cal I was speaking of (in the assembly manual). The wizard you linked can work, I only used my method and set the parameters differently in MatterControl using Cura as the slicer and set the z-height in there. I use .2 or .25 for the first layer.

Use 200mm_Bed_Level_Aid.STL instead to minimize filament usage and it will tell you your high/low spots quicker.
Your image looks more like an extrusion issue rather than bed level. The bottom right of all your skirts are skinny but the squares do not correspond, what is your extrusion multiplier set to and have you calibrated your extruder? https://www.matterhackers.com/articles/ ... r-extruder

I was having inconsistencies which is why I upgraded. The nozzle is larger so printing is slightly faster than the e3d but you can swap the bore size out (something I haven't tried). My nozzle and heater block always accumulated crud and those led to irregularities on my prints and saw the e3d came with a silicone boot to cover. If the rostock had their own cover, I would of gotten that instead of a whole new hot end. But reading forums and e3d was just a clear winner as an upgrade for me. I use a couple exotics like wood and tpu so I definitely benefit from e3d.

Keep posting on forums of your results and hopefully we can help you figure out what is wrong. Loose or angled towers, loose/too tight belts, use only 3 alternating bed clamps, are what I check first as others suggested.

Okay, so running G29 3 times and then calibrating using the wizard did wonders. I was able to get a good print for the 75mm single layer print. It's better than anything I have had before. I also made sure to tighten the belts again. But I still see uneven thickness. I have attached the new prints. I also reduced the speed and set parameters to the suggested ones in your previous post.

This is at .2mm height and 1.0 Extruder multiplier 0.55mm extrusion width on a 0.50mm nozzle dia. I have the stock hotend.

Searching the forums I see talk of tower tilt when the thickness is uneven. I don't know if that's the case here. How do I go about fine tuning my printer in that case?

Edit: It seems that the nozzle may be close to the bed between X/Z towers. I see that a little bit of the thin-ness between Z/Y as well. Should I play with the end stop? Or is there a better way to do it?

[djarmag]

I had extrusion width set to 100% last time I used Matter Control using MatterSlice engine.
Honestly you are very close to having it calibrated and these results look great, grab some digital calipers and do a first layer test. Check if the thickness printed matches close your setting. The thickness of the skirt will vary because it is still priming, so your first pic with the bed level aid doesn't indicate a bed level issue in my opinion.

What filament are you using? Check the width of the filament as well and verify it is 1.75mm.

The tilt can be checked using a carpenters square of some sort, the longer the better. First check if the aluminum extrusion is square to the melamine frame and then check the bed to the aluminum extrusion. If anything has a gap, you'll have to loosen the t-nuts, reseat the extrusion until fully seated and then tighten again. Also, I have my endstops fully screwed in so all heights match so check yours and recal if needed.

[pronoyc]

I had extrusion width set to 100% last time I used Matter Control using MatterSlice engine.
Honestly you are very close to having it calibrated and these results look great, grab some digital calipers and do a first layer test. Check if the thickness printed matches close your setting. The thickness of the skirt will vary because it is still priming, so your first pic with the bed level aid doesn't indicate a bed level issue in my opinion.

The thing is, on large diameter prints, I can see that the layer is uneven on the x/z side. If it would've been just the X side, I would've easily changed the X length by playing with the end screws. But I don't know how to work with 2 endstops simultaneously.

What filament are you using? Check the width of the filament as well and verify it is 1.75mm.

I did, they are 1.75mm average over 6 measurements. I also calibrated the E steps. Right now I am having some trouble with scarring on the second layer, I think it's the retraction settings (according to the Simplify3D build guide https://www.simplify3d.com/support/prin ... nd-ringing). I will try and and get it right, unless there are other ways to avoid that.

The tilt can be checked using a carpenters square of some sort, the longer the better. First check if the aluminum extrusion is square to the melamine frame and then check the bed to the aluminum extrusion. If anything has a gap, you'll have to loosen the t-nuts, reseat the extrusion until fully seated and then tighten again. Also, I have my endstops fully screwed in so all heights match so check yours and recal if needed.

I'll have to buy a set and check. But as far as I can see they are very well seated. But I'll check using the square.

[djarmag]

Scarring can be avoided using a 4mm z-hop or lift in your slicer setting. I use MC, Cura, or Kisslicer but havent paid for simplify3d so not sure where that setting is. I am curious if you tried running the delta cal wizard? I would suggest g29 three times, delta cal wizard, then g29 and print something to check bed level (at least thats how I do it with duet setup) I did this procedure with a RAMBo and found that running the wizard made it more accurate but it was too close to bed, but a final g29 would fix it.

[pronoyc]

Scarring can be avoided using a 4mm z-hop or lift in your slicer setting. I use MC, Cura, or Kisslicer but havent paid for simplify3d so not sure where that setting is. I am curious if you tried running the delta cal wizard? I would suggest g29 three times, delta cal wizard, then g29 and print something to check bed level (at least thats how I do it with duet setup) I did this procedure with a RAMBo and found that running the wizard made it more accurate but it was too close to bed, but a final g29 would fix it.

I have and it worked great. But I find that I have to calibrate quite often. I am going to try the z-hop settings and see if the scarring goes away.

[djarmag]

I was bored this weekend and I actually just redid my endstops to be within 4mm finally. I had my endstops fully in at first but what I ended up doing was unscrew it halfway so I have room to add or subtract thread length, recal to set the endstops, and adjust from there. Only work one axis at a time because it offsets the others. Now whenever it homes, it parks exactly center where it homes.

I still calibrate a bit even with FSR, mainly so I can adjust for good first layer heights.

Show a pic of what you are experiencing if you can.