OpenDACT(Delta Automatic Calibration Tool) - For Repetier

[KAS]

I guess that depends on how much one wants to tinker with the machine and what's considered good enough. For me, I get way better results with a dial indicator doing everything manually while taking my time. I've seen many different calibration techniques from following procedures to auto calibration routines but nobody post up examples of prints showing what to expect.

[626Pilot]

FSRs are alright, but you can get similar results from a much cheaper Z probe.

I think my theoretical favorite would be the kind of hot end with a hinge and a spring, and if it runs into the print surface it swivels up and touches a microswitch. Same benefit as FSRs (not having to swap out a probe).

[jason128]

I started using the simple HTML programme- and it worked an absolute treat- this is fantastic- absolutely amazing work in making this.
I just hot glued a dial gauge in place, its easy and quick, and i just break it off afterwards
unfortunately i realised my steps per mm was set wrong so was scaling all my prints down. i went to recalibrate, and i just cant get the html file to work now- it just freezes up - tried it on different browsers and different machines and get the same problem. Im sure im doing something stupid- i just cant figure out what.
Any ideas?

cheers

Jason

[DGBK]

I started using the simple HTML programme- and it worked an absolute treat- this is fantastic- absolutely amazing work in making this.
I just hot glued a dial gauge in place, its easy and quick, and i just break it off afterwards :)
unfortunately i realised my steps per mm was set wrong so was scaling all my prints down. i went to recalibrate, and i just cant get the html file to work now- it just freezes up - tried it on different browsers and different machines and get the same problem. Im sure im doing something stupid- i just cant figure out what.
Any ideas?

cheers

Jason

Make sure you click the button that loads the presets first. If you don't click that there will be some empty fields, and the program will not work properly.

[milp]

Thanks for making something as promising as this, but i'm going crazy about trying to utilize this tool.

1. I can't seem to fix the caliper in a way that i can do all the measurements in the g code, at one or another spot the caliper tip always leaves the glass plate. I'm using the standard hotend thing in my rostock max v2.

2. I'm running linux and only have a windows 7 vm. I have to lower the baudrate in windows to 128000 for it to work. It starts, moves down for one or two measure ment spots and then begins to home the extruder carriage at ridiculously low speed, after that i have to reset eeprom in order for the printer to work again at all.

3. Is there a way i can manually (without connecting the printer at all) type in all the eeprom info and measurements in your tool and have it process it? If so how does it work?

[featheryjam2]

I just wanted to thank you for this work! For all interested, here's my setup:

Z-probe - I used an inductive sensor (see Tom Sandladerer's video on youtube here: https://www.youtube.com/watch?v=EcGFLwj0pnA ). I originally had FSR's (but cheap ones from China ebay sellers). They were problematic for me because they were very inconsistent between sensors. 2 sensors would go off on the slightest touch but the 3rd sensor would take a ton of force before switching. This was also evident based on the location of probe measurement. At some spots on the bed the FSR's would never switch and the probe would go so far down into the bed that the bed would just release from the mounting. Once I transitioned over to the inductive sensor (and aluminum bed) I had better luck. The real challenge with these are mounting. What I did to mount (at first) was I took out the set screw from the nozzle and heater block and replaced it with a long screw. I then used both nuts on the inductive sensor to tighten down around the screw head--not ideal but was good enough for the preliminary calibration, which then allowed me to print a true mounting block.

Autocalibration version: 2.0.3PA. I had issues with this at first, but I noticed a very stupid setting I had in my firmware. I had the bed safe limit at 200mm, so basically at points far away from center and away from tower, the machine incorrectly thought that the carriage had hit the bed--this led to weird errors as the probe was measuring what I think was a different location with an offset. This caused weird calibration issues. However, once I reduced this value to a more reasonable value (5mm), the autocalibration ran well.

For the zprobe, I first had it offset from the extruder nozzle and did a preliminary calibration. My mounting for the probe wasn't solid so as a result the calibration wasn't great. However, I was able to print a bracket so that I could remove the hot end for calibration and have the probe directly on center, where the nozzle would be if the hot end wasn't detached. Using the autocalibration software, I was able to get the first "success" after 23 iterations. I then went ahead and clicked calibrate a few more times after checking the measurements in repetier-host. After about iteration 38, i was very happy with the autocalibration.

Few things to note for RollieRowland:
1) For some reason, the autocalibration height maps seem better than the probe measurements I take through repetier-host with the same radius selection. For example, I calibrated with the software and had successful calibration under .02mm. However, when I go into repetier-host and take the same measurements with the zprobe, 1 measurement is consistently off by about 0.07mm (opposite x-tower for me). This isn't really a big deal to me as I'm SUPER happy with even this.

2)A bug I've noticed is that sometimes on the current height map readings the values aren't correctly updated. I've seen this on the z-tower measurement occasionally. In the console area the values read really good (under .06 mm), but on the current height map it reads at 0.14mm. This is only occasionally so it might be hard to diagnose what the true problem is.

Anyways, I just wanted to thank you for putting time into this! I haven't found a nice conclusive way to manually calibrate this machine because everyone on youtube and blogs seem to have their own differing opinions on calibration. They usually only mention changing the homing offsets and delta radius with no mention of changing the rotations or diagonal terms. Because of this, I spent literally a month going in circles trying to calibrate the machine. Just when I was ready to call it quits I ran across this forum and it magically solved my problems. I can't wait until you post the mathematics for the calibration (if you do).

Thanks,
FJ2

[teoman]

Pnp or npn type of sensor?

[RollieRowland]

Thanks for making something as promising as this, but i'm going crazy about trying to utilize this tool.

1. I can't seem to fix the caliper in a way that i can do all the measurements in the g code, at one or another spot the caliper tip always leaves the glass plate. I'm using the standard hotend thing in my rostock max v2.

2. I'm running linux and only have a windows 7 vm. I have to lower the baudrate in windows to 128000 for it to work. It starts, moves down for one or two measure ment spots and then begins to home the extruder carriage at ridiculously low speed, after that i have to reset eeprom in order for the printer to work again at all.

3. Is there a way i can manually (without connecting the printer at all) type in all the eeprom info and measurements in your tool and have it process it? If so how does it work?

1. I am assuming you mean that the calipers do not always touch the plate when the gcode brings the effector down? If so, I used to manually push the depth gauge of the calipers down for each run - became very tiresome, decided to utilize a rubber band to keep the depth gauge low enough every time.
2. Is the baud rate consistent with the printer's baud rate? If it is, then it won't cause any issues. I would assume it is, otherwise you couldn't connect. However, the only possible sources of error would most likely be caused by running the program through a virtual machine, windows 7 has had issues, so it wouldn't surprise me.
3. If you want to manually calibrate then I would suggest running version 1.0.3. If you are using a Rostock Max V2, then use the default settings, and the rest is fairly self-explanatory.

Few things to note for RollieRowland:
1) For some reason, the autocalibration height maps seem better than the probe measurements I take through repetier-host with the same radius selection. For example, I calibrated with the software and had successful calibration under .02mm. However, when I go into repetier-host and take the same measurements with the zprobe, 1 measurement is consistently off by about 0.07mm (opposite x-tower for me). This isn't really a big deal to me as I'm SUPER happy with even this.

2)A bug I've noticed is that sometimes on the current height map readings the values aren't correctly updated. I've seen this on the z-tower measurement occasionally. In the console area the values read really good (under .06 mm), but on the current height map it reads at 0.14mm. This is only occasionally so it might be hard to diagnose what the true problem is.

Anyways, I just wanted to thank you for putting time into this! I haven't found a nice conclusive way to manually calibrate this machine because everyone on youtube and blogs seem to have their own differing opinions on calibration. They usually only mention changing the homing offsets and delta radius with no mention of changing the rotations or diagonal terms. Because of this, I spent literally a month going in circles trying to calibrate the machine. Just when I was ready to call it quits I ran across this forum and it magically solved my problems. I can't wait until you post the mathematics for the calibration (if you do).

Thanks,
FJ2

I will look into this and try to replicate it if possible, I haven't had any height-map values persist through runs (except the initial height). If this is happening, it could cause more complications than simply displaying an erroneous height-map. If the log is displaying correctly though, it most likely won't cause any issues elsewhere.

I am currently compiling my notes into a paper that should be easier to understand(and read). This involves adding detailed diagrams, thorough explanations, and a commentary that follows the code behind the calibration - so it will most likely not be done this week. And the more I delve into this, the more I want to write my own firmware from scratch. One that has support for autocalibration and the ability to modify the pitch of the tower - essentially calculate the angle of the tower and adjust values accordingly(variable not constant with changes in height). This would correct for scaling issues at every height of the build volume. The main reason that I want to do this is the concern with the inconsistent scaling errors that occur when the tower angles(angle from bed to tower, not rotation) are not perfect. It offsets the entire calibration, and is impossible to perfectly correct - even at 1 specific height. This problem is amplified the higher the carriages/effector are/is. Once my frame is rebuilt, I will determine if I want to dive into this even more absurd task...

Also, I should mention that the new version (auto-cal prog) wasn't fully fixed last Saturday, which is why I never uploaded it. I have time this weekend, the new version should be up probably Saturday night. I may also need some help/data from those who have Orions, Kossels, or any other printer with a printer radius(center of linear actuator to the center of the printer) smaller(or larger) than 200mm.

[featheryjam2]

I'd be happy to help / provide data. I'm still shocked at the quality of prints I'm getting! (I'm running a kossel mini)

[RollieRowland]

I'd be happy to help / provide data. I'm still shocked at the quality of prints I'm getting! (I'm running a kossel mini)

Oh, in that case... what are your default EEProm values and what were the EEProm values after the calibration? One more; would you be able to check if your X and Y scaling is accurate?

[milp]

Thanks for making something as promising as this, but i'm going crazy about trying to utilize this tool.

1. I can't seem to fix the caliper in a way that i can do all the measurements in the g code, at one or another spot the caliper tip always leaves the glass plate. I'm using the standard hotend thing in my rostock max v2.

2. I'm running linux and only have a windows 7 vm. I have to lower the baudrate in windows to 128000 for it to work. It starts, moves down for one or two measure ment spots and then begins to home the extruder carriage at ridiculously low speed, after that i have to reset eeprom in order for the printer to work again at all.

3. Is there a way i can manually (without connecting the printer at all) type in all the eeprom info and measurements in your tool and have it process it? If so how does it work?

1. I am assuming you mean that the calipers do not always touch the plate when the gcode brings the effector down?

No, what i mean is that when i put the caliper onto my hotend carriage with cable ties, at some point during calibration the side that it is attached to will leave the glass plate because the nozzle always moves to the outermost point on the glass plate. So if for example i choose to attach the caliper on the side of the hotend carriage which points towards the X/Y middle, the caliper measuring tip will not be on the glassplate at the Zopp and Y stops.
Is it possible to omit one or two measurement points?

There is no way to attach the caliper to the Stock rostock max v2 hotend and have the measuring tip be right above the bed at all times. Or has anyone else done this successfully with the stock hotend?

EDIT:

Is it the 1.0.3 the html version you mentioned?

Do all the values mentioned in it correspond with names in the eeprom settings?

[RollieRowland]

No, what i mean is that when i put the caliper onto my hotend carriage with cable ties, at some point during calibration the side that it is attached to will leave the glass plate because the nozzle always moves to the outermost point on the glass plate. So if for example i choose to attach the caliper on the side of the hotend carriage which points towards the X/Y middle, the caliper measuring tip will not be on the glassplate at the Zopp and Y stops.
Is it possible to omit one or two measurement points?

There is no way to attach the caliper to the Stock rostock max v2 hotend and have the measuring tip be right above the bed at all times. Or has anyone else done this successfully with the stock hotend?

EDIT:

Is it the 1.0.3 the html version you mentioned?

Do all the values mentioned in it correspond with names in the eeprom settings?

Oh, sorry I misunderstood what you meant. You could just modify the gcode to use a smaller radius. And actually, there was a newer version of the manual calibration.

Here is 1.0.4b - https://drive.google.com/file/d/0B06WCA ... cslist_api

The majority of the values correspond to EEProm values, though there are some that are just for the calculation, those are in the expert panel(except accuracy and measurement inversion). Since you are using an rmax2, you can just use enter defaults. Make sure your EEProm values all match the programs values though, or vice versa.

[jason128]

I started using the simple HTML programme- and it worked an absolute treat- this is fantastic- absolutely amazing work in making this.
I just hot glued a dial gauge in place, its easy and quick, and i just break it off afterwards
unfortunately i realized my steps per mm was set wrong so was scaling all my prints down. i went to recalibrate, and i just cant get the html file to work now- it just freezes up - tried it on different browsers and different machines and get the same problem. Im sure im doing something stupid- i just cant figure out what.
Any ideas?

cheers

Jason

Make sure you click the button that loads the presets first. If you don't click that there will be some empty fields, and the program will not work properly.

Thanks for that tip- it did the trick- now dialed in to about +-02 which is pretty cool- looking forward to seeing if i can get bigger prints working now.


so for anyone else having trouble with the HTML file in the future:
1 make sure explorer is not disabling active X controls
2 hit enter defaults before you start entering your settings- even if you think you are filling all the defalut boxes.

[milp]

Thanks, i started using this yesterday but without attaching the caliper, instead i measured it manually every time. I did two rounds of calibration and it still wasn't very flat.

Are there scenarios where this won't work? Like mechanical damage or the towers not being in a 90° angle to the bed?

Oh, sorry I misunderstood what you meant. You could just modify the gcode to use a smaller radius.

Thanks, but how would i go about this?

[RollieRowland]

Thanks, i started using this yesterday but without attaching the caliper, instead i measured it manually every time. I did two rounds of calibration and it still wasn't very flat.

Are there scenarios where this won't work? Like mechanical damage or the towers not being in a 90° angle to the bed?

Oh, sorry I misunderstood what you meant. You could just modify the gcode to use a smaller radius.

Thanks, but how would i go about this?

The program should be able to calibrate for (almost) any error At bed height, even if your towers are not square. However, if your towers are square, then the accuracy will persist through all heights.

If the calibration is not working right away, either try more iterations, or use the expert mode - which takes longer but will be adjusted specifically to your printer.

In the gcode there are three recurring values in the gcode: 130, 113, and 65. These can be found by multiplying (consecutively):
130 = 0.482 * plateDiameter;
113 = 0.417 * plateDiameter;
65= 0.241 * plateDiameter;
So to modify the values, replace plate diameter with the diameter you want to use. Then in the gcode, replace every instance of 130, 113, and 65 with your calculation.

[geneb]

I'm working through the process of getting Blue MAX set up for using FSRs. Here's a couple of things I've run into.

1. If you don't want to set a jumper on the Ultibot FSR board, change Z_PROBE_ON_HIGH to be false.
2. Even if you set Z_PROBE_SPEED and Z_PROBE_HEIGHT in Configuration.h, you're still going to need to make sure you change the EEPROM values "Z-Probe Height" and "Z-Probe speed" to their correct values. The height should be 0 (for the FSR) and the probe speed should be 30. The firmware is going to give priority to the EEPROM setting over the hard-coded setting. You want to make sure you set both, that way if you have to clear the EEPROM for any reason, you won't lose your Z probe configuration.

More later!

g.

[milp]

I tried the normal manual calibration for ~5 iterations and i still have multiple millimeters in height difference on my measurement points still.

I guess ill try the expert variant tomorrow, but i really don't get why this isn't working for me. Could that mean a stepper is damaged?

I checked the tower angles and they are pretty much 90° each.

[bot]

I think the least likely scenario is a problem with a stepper motor. Make sure your belt tension is good, cheapskate tension, etc.

[milp]

I think the least likely scenario is a problem with a stepper motor. Make sure your belt tension is good, cheapskate tension, etc.

I just went through all that, i've been trying around for a week now. Got new delta arms after two had broken in the past when there were some unsupervised printer accidents. I noticed the hotend carriage still had some "free play" in it so i put cable ties around the ends of the delta arms on each side.

[geneb]

Milp, Blue MAX took *18* iterations. If you value your sanity, I'd get an FSR kit.

Below is the run I just finished. I'm going to re-run it with the hot end and bed at operating temp.

g.

Code: Select all

Connected

Setting Z-Probe Speed

Setting Z-Probe Height

Variables set

Request to read EEPROM sent

EEProm capture initiated

EEProm:  Steps:80, X Max:357.813, Y Max:357.813, Z Max:350.675, Z-Probe Offset:0, Diagonal Rod:269, Horizontal Radius:129.617, X Offset:0, Y Offset:0, Z Offset:0, Alpha A:210, Alpha B:330,  Alpha C:90, Delta A:0, Delta B:0, and Delta C:0

EEProm captured, beginning calibration.
Center:350.905, X:-0.700000000000045, XOpp:0.0799999999999841, Y:-0.0200000000000387, YOpp:0.629999999999995, Z:0.169999999999959, and ZOpp:-0.160000000000025

Setting Z Max Length

Current iteration: 1

HRad:129.617

HRad Correction:0

HRad Average before calibration:-2.8421709430404E-14

HRad Average:-2.8421709430404E-14

Horizontal Radius Calibration Success; Checking height-map.

Center:350.895, X:-0.730000000000018, XOpp:0.0600000000000023, Y:-0.00999999999999091, YOpp:0.610000000000014, Z:0.159999999999968, and ZOpp:-0.139999999999986

Setting Z Max Length

Current iteration: 2

Delta Radii Offsets: 0.670000000000016, -0.600000000000023, -0.0199999999999818
Checking height-map
Center:350.895, X:-0.360000000000014, XOpp:0.160000000000025, Y:-0.350000000000023, YOpp:0.54000000000002, Z:0.139999999999986, and ZOpp:-0.149999999999977

Setting Z Max Length

Current iteration: 3

Delta Radii Offsets: 0.870000000000005, -0.79000000000002, -0.00999999999999091
Checking height-map
Center:350.895, X:-0.259999999999991, XOpp:0.180000000000007, Y:-0.470000000000027, YOpp:0.509999999999991, Z:0.149999999999977, and ZOpp:-0.139999999999986

Setting Z Max Length

Current iteration: 4

Delta Radii Offsets: 0.949999999999989, -0.829999999999984, -0.0199999999999818
Checking height-map
Center:350.905, X:-0.21999999999997, XOpp:0.180000000000007, Y:-0.5, YOpp:0.490000000000009, Z:0.129999999999995, and ZOpp:-0.129999999999995

Setting Z Max Length

Current iteration: 5

Delta Radius Calibration Success; Checking Height-Map
Center:350.895, X:-0.230000000000018, XOpp:0.189999999999998, Y:-0.490000000000009, YOpp:0.509999999999991, Z:0.120000000000005, and ZOpp:-0.139999999999986

Setting Z Max Length

Current iteration: 6

Tower Rotation calculated, check XY Panel

XYZ Offset Correction: 0
XYZ Offset Average Before Calibration: -0.200000000000008
XYZ Offset Average Afer Calibration: -0.200000000000008
Virtual heights: X:-0.230000000000018, XOpp:0.189999999999998, Y:-0.490000000000009, YOpp:0.509999999999991, Z:0.120000000000005, and ZOpp:-0.139999999999986

XYZ:112 53 178

Checking height-map

Center:349.465, X:-0.180000000000007, XOpp:0.170000000000016, Y:0.29000000000002, YOpp:0, Z:-0.680000000000007, and ZOpp:0.370000000000005

Setting Z Max Length

Current iteration: 7

Tower Rotation calculated, check XY Panel

XYZ Offset Correction: 19.999999999981
XYZ Offset: 28.5714285714015
XYZ Offset Average Before Calibration: -0.200000000000008
XYZ Offset Average Afer Calibration: -0.189999999999998
Virtual heights: X:-0.180000000000007, XOpp:0.170000000000016, Y:0.29000000000002, YOpp:0, Z:-0.680000000000007, and ZOpp:0.370000000000005

XYZ:91 86 100

Checking height-map

Center:349.745, X:-0.189999999999998, XOpp:0.180000000000007, Y:-0.42999999999995, YOpp:0.450000000000045, Z:0.0300000000000296, and ZOpp:-0.089999999999975

Setting Z Max Length

Current iteration: 8

Tower Rotation calculated, check XY Panel

XYZ Offset Correction: 59.9999999993744
XYZ Offset: 85.714285713392
XYZ Offset Average Before Calibration: -0.200000000000008
XYZ Offset Average Afer Calibration: -0.196666666666639
Virtual heights: X:-0.189999999999998, XOpp:0.180000000000007, Y:-0.42999999999995, YOpp:0.450000000000045, Z:0.0300000000000296, and ZOpp:-0.089999999999975

XYZ:69 37 103

Checking height-map

Center:350.025, X:-0.20999999999998, XOpp:0.170000000000016, Y:-0.0799999999999841, YOpp:0.25, Z:-0.299999999999955, and ZOpp:0.120000000000005

Setting Z Max Length

Current iteration: 9

Tower Rotation calculated, check XY Panel

XYZ Offset Correction: 59.9999999993744
XYZ Offset: 85.714285713392
XYZ Offset Average Before Calibration: -0.200000000000008
XYZ Offset Average Afer Calibration: -0.196666666666639
Virtual heights: X:-0.20999999999998, XOpp:0.170000000000016, Y:-0.0799999999999841, YOpp:0.25, Z:-0.299999999999955, and ZOpp:0.120000000000005

XYZ:45 28 69

Checking height-map

Center:350.305, X:-0.189999999999998, XOpp:0.170000000000016, Y:-0.25, YOpp:0.330000000000041, Z:-0.139999999999986, and ZOpp:0.0600000000000023

Setting Z Max Length

Current iteration: 10

Tower Rotation calculated, check XY Panel

XYZ Offset Correction: 29.9999999999431
XYZ Offset: 42.8571428570616
XYZ Offset Average Before Calibration: -0.200000000000008
XYZ Offset Average Afer Calibration: -0.193333333333328
Virtual heights: X:-0.189999999999998, XOpp:0.170000000000016, Y:-0.25, YOpp:0.330000000000041, Z:-0.139999999999986, and ZOpp:0.0600000000000023

XYZ:80 28 110

Checking height-map

Center:349.995, X:-0.300000000000011, XOpp:0.25, Y:0.0500000000000114, YOpp:0.129999999999995, Z:-0.350000000000023, and ZOpp:0.160000000000025

Setting Z Max Length

Current iteration: 11

Tower Rotation calculated, check XY Panel

XYZ Offset Correction: -Infinity
XYZ Offset: -Infinity
XYZ Offset Average Before Calibration: -0.200000000000008
XYZ Offset Average Afer Calibration: -0.200000000000008
Virtual heights: X:-0.300000000000011, XOpp:0.25, Y:0.0500000000000114, YOpp:0.129999999999995, Z:-0.350000000000023, and ZOpp:0.160000000000025

XYZ:46 34 70

Checking height-map

Center:350.255, X:-0.149999999999977, XOpp:0.180000000000007, Y:-0.279999999999973, YOpp:0.389999999999986, Z:-0.0999999999999659, and ZOpp:0.0300000000000296

Setting Z Max Length

Current iteration: 12

Tower Rotation calculated, check XY Panel

XYZ Offset Correction: 8.57142857141581
XYZ Offset: 12.2448979591654
XYZ Offset Average Before Calibration: -0.200000000000008
XYZ Offset Average Afer Calibration: -0.176666666666639
Virtual heights: X:-0.149999999999977, XOpp:0.180000000000007, Y:-0.279999999999973, YOpp:0.389999999999986, Z:-0.0999999999999659, and ZOpp:0.0300000000000296

XYZ:29 2 59

Checking height-map

Center:350.535, X:-0.21999999999997, XOpp:0.160000000000025, Y:-0.149999999999977, YOpp:0.260000000000048, Z:-0.239999999999952, and ZOpp:0.100000000000023

Setting Z Max Length

Current iteration: 13

Tower Rotation calculated, check XY Panel

XYZ Offset Correction: 8.57142857141581
XYZ Offset Average Before Calibration: -0.200000000000008
XYZ Offset Average Afer Calibration: -0.2033333333333
XYZ Offset Calibration Success; checking height-map

Center:350.525, X:-0.199999999999989, XOpp:0.199999999999989, Y:-0.150000000000034, YOpp:0.269999999999982, Z:-0.230000000000018, and ZOpp:0.0999999999999659

Setting Z Max Length

Current iteration: 14

ABC:210.143390625 329.9886796875 90.00566015625

Heights: X:-0.199999999999989, XOpp:0.190156249999979, Y:-0.150000000000034, YOpp:0.190156249999979, Z:-0.230000000000018, and ZOpp:0.189687499999978

Setting A Rotation

Setting B Rotation

Setting C Rotation

Checking height-map

Center:350.525, X:-0.20999999999998, XOpp:0.180000000000007, Y:-0.160000000000025, YOpp:0.189999999999998, Z:-0.240000000000009, and ZOpp:0.180000000000007

Setting Z Max Length

Current iteration: 15

Alpha Rotation Calibration Success; checking height-map

Center:350.525, X:-0.199999999999989, XOpp:0.160000000000025, Y:-0.149999999999977, YOpp:0.199999999999989, Z:-0.220000000000027, and ZOpp:0.160000000000025

Setting Z Max Length

Current iteration: 16

Diagonal Rod:273.529538755536

Heights: X:-0.00999999999999091, XOpp:-0.0123631004428714, Y:0.0400000000000205, YOpp:0.0276368995570922, Z:-0.0300000000000296, and ZOpp:-0.0123631004428714

Setting diagonal rod

Setting Horizontal Radius

Checking height-map

Center:350.525, X:0.0299999999999727, XOpp:0.0400000000000205, Y:0.0400000000000205, YOpp:0.089999999999975, Z:-0.0199999999999818, and ZOpp:0.0699999999999932

Setting Z Max Length

Current iteration: 17

Diagonal Rod:274.143597000027

Heights: X:0.0133333333333023, XOpp:-0.0257913262259097, Y:0.02333333333335, YOpp:0.0242086737740449, Z:-0.0366666666666523, and ZOpp:0.00420867377406304

Setting diagonal rod

Setting Horizontal Radius

Checking height-map

Center:350.535, X:0, XOpp:-0.0199999999999818, Y:0.0400000000000205, YOpp:0.0300000000000296, Z:-0.0500000000000114, and ZOpp:0

Setting Z Max Length

Current iteration: 18

Diagonal Rod Calibration Success; checking height-map

Calibration Complete; Homing Printer

[geneb]

Here's the end-result of the 2nd cal at temp (8 iterations):
[img]http://www.geneb.org/images/blue-max-cal-object.jpg[/img]

I'm happy with it. it's a little thin near the Z tower, but that's not a deal breaker.

The layer thickness is excellent - I made no adjustments to the Z height after the calibration process was finished, so the first layer actually was .25mm.

I figured I'd make this a REAL test and print it on cleaned, totally bare glass. No glue, spray, PEI, etc. PLA @ 190C and the bed at 55C. \o/

Now to FSR-ify White MAX and Orange Menace. Poor Red Sonja and Orion Lannister will have to wait until I can get FSR kits for them.

g.

[bvandiepenbos]

@geneb
how did you mount FSR's under your bed?
with mHackneys printed mounts?
with 3 of the Onyx mount screws, right?

I am concerned about XY bed jiggle if you have the pucks sized to freely move in Z to trigger FSR's correctly.

[dkaustin]

I am concerned about XY bed jiggle if you have the pucks sized to freely move in Z to trigger FSR's correctly.

Yes, indeed. Building a three-point floating support system that has no perceptible slop yet doesn't bind is quite a trick. If you get it right on the first try, you must be living right. It's the key to a good FSR installation, though.

[RollieRowland]

Version 2.0.4PA
Ok, I've just corrected the issue that was causing the XYZ offset to send absurd (20,000+ steps) values back to the printer. This version will include scaling error minimization, 3 different calibration methods, a working UI, damage prevention(no values get sent if they will break your printer), and a still not yet completed tower analysis. The tower analysis is not pertinent to the calibration, which is why I am going to upload the version anyway.

Tests(all EEPROM values reset prior to testing and at 270mm build diameter):
Test 1: Basic Calibration - Tolerance: +/-0.01mm - 6 iterations.
Test 2: Basic Calibration - Tolerance: +/-0.01mm - 8 iterations.
Test 3: A.I. Calibration - Tolerance: +/-0mm - 17 iterations
Test 4: H.A.I. Calibration - Tolerance: +/-0.01mm - 30 iterations

I recommend that you run the tests in this order(only continue if the previous failed):
1. Basic - estimates: fastest, if you know your default settings or the programs defaults work for your printer
2. A.I. - learns offsets and estimates: slower, higher chance of calibration completing if the basic gives you problems
3. H.A.I. - learns offsets and will correct its mistakes: if this doesn't work, then make sure that you are not missing a tower...

Download - https://drive.google.com/file/d/0B06WCA ... sp=sharing

[geneb]

Brian, I am using mhackney's mounts. They have zero horizontal motion to them at all. I set them all down, wicked a bit of CA under the printer-side brackets and then pilot drilled and installed #4-1/2" screws. Works fantastic.

g.