Dc42IR Calibration Problems on Rostock Max V2



  • So to start with I've been toying with this for several weeks, been quite busy at work so I caused myself problems early on by trying to move too fast through the write-ups and missing stuff!

    The machine is a Rostock Max V2 melamine frame. It has an E3D v6 mounted with the Trick Laser aluminum effector and mount combo. Arms are Trick Laser CF labeled 300.15mm with red ball cup ends using the supplied (rather stiff) springs for tension. Carriages are SeeMeCNC injection molded. The heated bed PCB is the original, the build surface is original glass that has been painted matte black on the back per my readings here and Dave's write-up about using his IR leveling sensor.

    Original mechanical end stop switches were flipped around so the metal lever no longer causes variance in activation. The screws now strike directly at the plunger.

    Belt tension has been set using a digital gauge measured at 12" up from the base top plate with the belt at 1" of deflection from the tower so I know they are all within 0.05oz of each other.

    The effector DOES tilt ever so slightly as noted by a cheap bullseye level. Not accurate enough to give me specific degrees, but quite clear there is a bit of movement.

    I've even swapped out the TR tension springs for lesser power springs several times downgrading until the springs were so weak they couldn't hold the effector to the arms.

    Current bed.g setup for 16 points, 6 factor.
    Here are the results of four runs of G32 w/S-1

    Bed probe heights: 0.004 -0.177 0.045 0.057 -0.374 -0.276 -0.034 -0.512 -0.465 -0.148 -0.112 -0.035 -0.238 -0.162 -0.285 -0.233, mean -0.184, deviation from mean 0.167

    Bed probe heights: -0.009 -0.189 0.045 0.044 -0.374 -0.263 -0.021 -0.500 -0.465 -0.161 -0.112 -0.035 -0.226 -0.137 -0.272 -0.220, mean -0.181, deviation from mean 0.163

    Bed probe heights: -0.009 -0.202 0.020 0.032 -0.374 -0.238 0.016 -0.475 -0.452 -0.148 -0.112 -0.035 -0.213 -0.112 -0.272 -0.220, mean -0.175, deviation from mean 0.158

    Bed probe heights: -0.021 -0.214 0.020 0.032 -0.361 -0.226 0.016 -0.475 -0.465 -0.161 -0.125 -0.048 -0.213 -0.112 -0.272 -0.220, mean -0.178, deviation from mean 0.156

    Help me see what I'm not seeing!


  • administrators

    Whatever is causing the tilt will be causing geometrical inaccuracies too, and also causing the relative heights of the IR sensor and the nozzle to change - which causes the trigger height of the IR sensor (relative to the tip of the nozzle) to change. You can compensate for the variation of trigger height using H parameters on the G30 commands in bed,g, although measuring all the trigger heights is tedious. It's better to fix the geometrical issue that is causing the tilt. See https://duet3d.com/wiki/Calibrating_a_delta_printer#Make_sure_your_build_is_accurate for common causes.



  • You don't have to save the results, but what happens when you try a 9 point factor?



  • @dc42:

    Whatever is causing the tilt will be causing geometrical inaccuracies too, and also causing the relative heights of the IR sensor and the nozzle to change - which causes the trigger height of the IRT sensor (relative to the tip of the nozzle) to change. You can compensate for the variation of trigger height using H parameters on the G30 commands in bed,g, although measuring all the trigger heights is tedious. It's better to fix the geometrical issue that is causing the tilt. See https://duet3d.com/wiki/Calibrating_a_delta_printer#Make_sure_your_build_is_accurate for common causes.

    Yup I'd prefer to fix the mechanical geometry. I've also done the H value offsets before by doing that tedious sampling. It didn't seem to help and barely reduced the error…or somehow I managed to do it wrong. And since I don't have all the high dollar tools, the machinist guys in the shop took it out and put their fancy run-out dial indicators on the towers and used some rather expensive 1m long calipers and gave it back to me and said it was square as could be with the flexing melamine frame. However, I didn't have them measure the rods. I simply trusted that TR makes them accurate since they use a jig. Bad assumption?

    I've also read that guide and looked at the carriages, the ball ends..etc. At this point IF there is something out of alignment I'm sure not seeing it or have no way to verify it. I'm leaning toward the ball ends being the issue, but can't say exactly how or why.



  • @number40fan:

    You don't have to save the results, but what happens when you try a 9 point factor?

    Haven't tried that yet. Will likely have to wait until Monday when I get back to the office.


  • administrators

    You could also try 8 factor calibration in case there is some bed tilt.

    Is there a particular pattern to the height errors? For example, high spots between pairs of towers?

    Did you run auto calibration a couple of times before doing those 4 runs in report-only mode?



  • @dc42:

    You could also try 8 factor calibration in case there is some bed tilt.

    Is there a particular pattern to the height errors? For example, high spots between pairs of towers?

    Did you run auto calibration a couple of times before doing those 4 runs in report-only mode?

    Ok, so I finally got to doing it remotely today, so I haven't/can't do anything physical to adjust anything until Monday. But I was able to run a 9 factor:
    Calibrated 9 factors using 16 points, deviation before 0.047 after 0.043
    Calibrated 9 factors using 16 points, deviation before 0.043 after 0.043

    Followed by S-1 report:
    Bed probe heights: 0.028 -0.083 0.088 -0.024 -0.042 0.059 -0.005 -0.039 0.061 0.026 -0.002 0.052 -0.036 0.034 0.010 0.003, mean 0.008, deviation from mean 0.044

    9 again:
    Calibrated 9 factors using 16 points, deviation before 0.044 after 0.040
    Calibrated 9 factors using 16 points, deviation before 0.048 after 0.043

    8 factor:
    Calibrated 8 factors using 16 points, deviation before 0.047 after 0.045
    Calibrated 8 factors using 16 points, deviation before 0.047 after 0.044

    And then I started getting "G32
    Error: G-Code buffer '$s' length overflow"

    So that's where I'm at now. Looks a load better than it did!



  • Can you run M665 and M666 to see what the results were?



  • @number40fan:

    Can you run M665 and M666 to see what the results were?

    M665
    Diagonal 311.755, delta radius 145.957, homed height 360.711, bed radius 140.0, X 0.295°, Y -0.288°, Z 0.000°
    M666
    Endstop adjustments X-0.53 Y0.27 Z0.26, tilt X0.17% Y0.23%



  • Increased the rod length quite a bit. I know David (DC42) says not to let it calibrate rod length if you know the exact length, but if it helps with calibration, I can't see why not. Your 9 factor proved that. You can save the results with M500 if you wish and when you get back to the machine, see what happens with a print. When I had massive effector tilt, it had to do with rod spacing. It was a cheap DIY delta that had a 3mm difference between the top and bottom widths of the diagonal arms.



  • @number40fan:

    Increased the rod length quite a bit. I know David (DC42) says not to let it calibrate rod length if you know the exact length, but if it helps with calibration, I can't see why not. Your 9 factor proved that. You can save the results with M500 if you wish and when you get back to the machine, see what happens with a print. When I had massive effector tilt, it had to do with rod spacing. It was a cheap DIY delta that had a 3mm difference between the top and bottom widths of the diagonal arms.

    Yup, I haven't measured the rods myself other than a quick guess with a decent ruler as I don't have a jig for proper center to center right now. And everything I read basically said "DON'T DO IT" to using the automatic rod adjustments. So that's the first time I've used the autocal with rod adjustment. Each of the rods came to me with 300.15 written on them and I'd assume that since TR uses a jig for manufacturing them…they might actually be exact. I'm not sure what to believe on that front now.



  • Grumble. It seems calibrating a delta is never complete.
    My effector tilts a little bit too. It sucks but I think the bigger the radius, the worse the tilt on the outter edge. So, smaller radius means no tilt. There is some math out there to calculate the radius based on many angles and variables but no doubt it is a labor of love it seems. My Prusa clone printer is so darn reliable, thank god I got that to use while I toy with my Delta.

    I'd LOVE to see 3 IR probes, one for each tower! I think that is the only way…

    How is your dimensional accuracy?


  • administrators

    You can use a nozzle contact Z probe, such as the one we've just started selling (https://www.duet3d.com/DeltaSmartEffector) or a piezo probe, to avoid the problem of trigger height changes caused by effector tilt. But whatever causes effector tilt causes other geometrical issues too.



  • Thank nozzle z probe is cool. Too bad not for Diamond hotends, which is the source of all my delta problems. 1 color was easy.



  • What kind of effector do you have, Eddie? Is it a printed version?



  • Ah, diamond hotend effectors, I been through about 3 different designs and no less than a total of 20 printed effectors.
    I come to think it is not the effectors but the silly want the diamond hotend is mounted to the effectors which is by zipties. Granted they are tight but whose to say it is perfectly level after pulling on the ties one at a time … Such a pain to assemble a diamond hotend on a delta effector and getting a good seal.

    Maybe I should change to a metal effector, I see them on ebay, like this - http://www.ebay.com/itm/272746591437

    The China diamoned hotend nozzles are about 2x the size of the official one, not sure how that would work out.

    It seems to be good enough for now, just need to dial things in to the .. microns.. Zip ties and microns just doesnt seem right, like duct tape and precision. 😛



  • Blah, that eBay listing shows zip ties too

    I am using this effector -
    https://www.thingiverse.com/thing:854622



  • @Eddiie:

    Grumble. It seems calibrating a delta is never complete.
    My effector tilts a little bit too. It sucks but I think the bigger the radius, the worse the tilt on the outter edge. So, smaller radius means no tilt. There is some math out there to calculate the radius based on many angles and variables but no doubt it is a labor of love it seems. My Prusa clone printer is so darn reliable, thank god I got that to use while I toy with my Delta.

    I'd LOVE to see 3 IR probes, one for each tower! I think that is the only way…

    How is your dimensional accuracy?

    Ha no kidding. One of my Rostocks at home is still in the same pile of parts it was almost 2 years ago because I basically gave up on it. I was never able to reliably print anything larger than 100mm circle. And even that was not great. Anything under that and it was flawless beautiful prints. Also note that a Rostock was my very first 3D printer….over four years ago!

    Dimensional accuracy on the machine in question is undetermined...I haven't even been able to get a solid print out of it since I started this maddening upgrade process last month. Tomorrow when I get back to the office I can load up filament and test with the results I was able to get yesterday operating it remotely.



  • @dc42:

    You can use a nozzle contact Z probe, such as the one we've just started selling (https://www.duet3d.com/DeltaSmartEffector) or a piezo probe, to avoid the problem of trigger height changes caused by effector tilt. But whatever causes effector tilt causes other geometrical issues too.

    I'm frankly not sure I'm willing to spend MORE money on the same machine. Every time there's a new fangled thing that is supposed to make it better and awesome all I ever get is a headache and marginal improvement if any. $500+ extra into it and I'm still no better off than the original machine with manual calibration mixed with some formula and guesswork. If the readings I've posted above are finally right and I can get some prints I will be truly amazed!

    What say you about the automatic arm length calculations here? Everywhere I read that I shouldn't, but that 8 and 9 factor FINALLY shows decent results.



  • This morning I run G32 and here's the results, bed @60C, nozzle@ 200C :
    Calibrated 8 factors using 16 points, deviation before 0.089 after 0.043
    Calibrated 8 factors using 16 points, deviation before 0.047 after 0.043
    Calibrated 8 factors using 16 points, deviation before 0.048 after 0.039

    So I'm thinking that's looking pretty good now. I fire off an XYZ calibration cube print and it looks terrible!
    First layer is still doing exactly the same thing I've been seeing for weeks now. Toward the right side of the machine (between Z and Y tower and around to the front) the first layer prints SUPER thin, just measured 0.07 and the left side between X and Z towers is 0.19 or 0.17 depending on how far forward or aft it is measured.

    The 'cube' is supposed to be 40x40x20 and I'm measuring ~39.16x39.14x19.91.

    This is still with NO H offsets in the bed.g. Just the 8 factor calibration


Locked
 

Looks like your connection to Duet3D was lost, please wait while we try to reconnect.