TAP implementation changes expected offset

Question

TAP implementation changes expected offset

Anonoei opened this issue 8 months ago · comments

Different implementations of TAP have different expected offsets.
Examples:

TAP Name	Expected Result	Offsets observed	Travel observed	Travel-to-offset
CL CNC Tap	0.03-1mm	-0.15, -0.042	0.0740, 0.025	* 2
CL CNC 2	0.5-1.5mm	-0.83	0.047	* 17.7
TAP R6	0.5-2mm	-1.04	0.045	* 23
TAP R8	0.5-2mm	-1.260, -1.320	0.0270	* 10 + 1
Mellow Tap	0.5-2mm	-1.725
Vitalli CNC	0.5-2mm	-0.865, -1.070	0.500

Thanks to the people below for providing these numbers

RedWish, Migo, Sanaa, EightFinger, Morfesto, cplmayo

Observed results

v2 Stock RC8 Results (Redwish), expected ~1.260

Samples: 5, Total Steps: 27
Probe Mean: 0.0038 / Min: 0.0013 / Max: 0.0050
Measure Mean: 0.0308 / Min: 0.0308 / Max: 0.0308
Travel Mean: 0.0270 / Min: 0.0250 / Max: 0.0300

v1 Vitalli MetalTap (EightFinger), expected ~0.865

Samples: 10, total Steps: 100
Probe Mean: -0.0235 / Min: -0.0356 / Max: -0.0069
Measure Mean: 0.0265 / Min: 0.0265 / Max: 0.0265
Travel Mean: 0.0500 / Min: 0.0500 / Max: 0.0500
Calculated STA Z-Offset: 0.052

CL CNC v2 (cplmayo), expected -0.830

Samples: 5, Total Steps: 47
Probe Mean: -0.0002 / Min: -0.0012 / Max: 0.0013
Measure Mean: 0.0468 / Min: 0.0468 / Max: 0.0468
Travel Mean: 0.0470 / Min: 0.0450 / Max: 0.0500
Calculated z-offset: 0.094

Reth · Answer 1 · Tue Oct 10 2023 11:27:43 GMT+0800 (China Standard Time)

I think what you might be measuring is the linear stickyness. What about if you 1) pulled the toolhead away quickly, and 2) added a slight pause for the toolhead to sink back down on the linear rail

Anonoei · Answer 2 · Wed Oct 11 2023 09:49:59 GMT+0800 (China Standard Time)

@The-Original-Reth I'll give this a try. I think since I switched the probe_speed to 1 I've gotten lower travel distance - and a lower calculated offset.

Rev · Answer 3 · Thu Oct 12 2023 01:26:38 GMT+0800 (China Standard Time)

Have run Auto_Tap a few times now, and the calculated Z-offset on my R8 Tap is always ~1.310.
This ends up with Z0 about .6mm too high. When I manually set the z_offset, the nozzle just touching the bed ends up at 0.611

18:19:12
Auto TAP performing 10 samples to calculate z-offset on R8 tap
Probe Min: -2.0, Stop: 2.0, Step: 0.0025
18:19:43
Auto TAP sample 1
Traveled: 0.0300 from z-0.5335 to -0.5035 on step 12
18:20:00
Auto TAP sample 2
Traveled: 0.0300 from z-0.5335 to -0.5035 on step 12
18:20:18
Auto TAP sample 3
Traveled: 0.0300 from z-0.5335 to -0.5035 on step 12
18:20:35
Auto TAP sample 4
Traveled: 0.0300 from z-0.5335 to -0.5035 on step 12
18:20:53
Auto TAP sample 5
Traveled: 0.0300 from z-0.5329 to -0.5029 on step 12
18:21:10
Auto TAP sample 6
Traveled: 0.0300 from z-0.5335 to -0.5035 on step 12
18:21:28
Auto TAP sample 7
Traveled: 0.0325 from z-0.5341 to -0.5016 on step 13
18:21:47
Auto TAP sample 8
Traveled: 0.0325 from z-0.5335 to -0.5010 on step 13
18:22:05
Auto TAP sample 9
Traveled: 0.0325 from z-0.5341 to -0.5016 on step 13
18:22:23
Auto TAP sample 10
Traveled: 0.0325 from z-0.5341 to -0.5016 on step 13
18:22:23
Auto TAP Results
Samples: 10, Total Steps: 124
Probe Mean: -0.5336 / Min: -0.5341 / Max: -0.5329
Measure Mean: -0.5026 / Min: -0.5026 / Max: -0.5026
Travel Mean: 0.0310 / Min: 0.0300 / Max: 0.0325
Calculated z-offset on R8 tap: 1.310

## [probe]
## z_offset = -0.611

Anonoei · Answer 4 · Thu Oct 12 2023 14:13:27 GMT+0800 (China Standard Time)

@ieRev Could you try running it again on the newest version? Your results vs other ones I've seen are... frightening for this being able to work on R8. Thank you for posting!

Rev · Answer 5 · Fri Oct 13 2023 01:19:31 GMT+0800 (China Standard Time)

Updated to newest version, then set Z_offset to 0.00 in printer.cfg, homed & QGL'ed before running:

18:01:32
Auto TAP performing 10 samples to calculate z-offset on R8 tap using 'rev hop' method
Probe Min: -2.0, Stop: 2.0, Step: 0.0025
18:02:01
Auto TAP sample 1
Traveled: 0.0275 from z0.0663 to 0.0938 on step 11
18:02:18
Auto TAP sample 2
Traveled: 0.0275 from z0.0675 to 0.0950 on step 11
18:02:34
Auto TAP sample 3
Traveled: 0.0275 from z0.0656 to 0.0931 on step 11
18:02:50
Auto TAP sample 4
Traveled: 0.0275 from z0.0650 to 0.0925 on step 11
18:03:07
Auto TAP sample 5
Traveled: 0.0275 from z0.0644 to 0.0919 on step 11
18:03:23
Auto TAP sample 6
Traveled: 0.0275 from z0.0644 to 0.0919 on step 11
18:03:40
Auto TAP sample 7
Traveled: 0.0275 from z0.0650 to 0.0925 on step 11
18:03:56
Auto TAP sample 8
Traveled: 0.0275 from z0.0644 to 0.0919 on step 11
18:04:12
Auto TAP sample 9
Traveled: 0.0275 from z0.0644 to 0.0919 on step 11
18:04:29
Auto TAP sample 10
Traveled: 0.0275 from z0.0644 to 0.0919 on step 11
18:04:29
Auto TAP Results on R8 using 'rev hop' method
Samples: 10, Total Steps: 110
Probe Mean: 0.0651 / Min: 0.0644 / Max: 0.0675
Measure Mean: 0.0926 / Min: 0.0926 / Max: 0.0926
Travel Mean: 0.0275 / Min: 0.0275 / Max: 0.0275
Calculated z-offset: 1.210

After running, Saved Config, rehomed/QGL and did G0 Z0.. Nozzle tip was 0.6mm from the bed
This was pretty much exactly the same distance it was high by yesterday. I will be starting a (short) print and will update with how much offset is needed once it starts..

Printed needed -0.65mm to get an acceptable first layer

Glen · Answer 6 · Wed Nov 15 2023 11:18:16 GMT+0800 (China Standard Time)

Hope this might help :)

V2.4R2

Auto TAP Results on R8 using 'rev hop' method
Samples: 5, Total Steps: 25
Probe Mean: 0.0013 / Min: 0.0000 / Max: 0.0025
Measure Mean: 0.0263 / Min: 0.0263 / Max: 0.0263
Travel Mean: 0.0250 / Min: 0.0250 / Max: 0.0250
Calculated z-offset: 1.249
12:54
Auto TAP sample 5
Traveled: 0.0250 from z0.0025 to 0.0275 on step 5
12:54
Auto TAP sample 4
Traveled: 0.0250 from z0.0025 to 0.0275 on step 5
12:54
Auto TAP sample 3
Traveled: 0.0250 from z0.0000 to 0.0250 on step 5
12:54
Auto TAP sample 2
Traveled: 0.0250 from z0.0000 to 0.0250 on step 5
12:54
Auto TAP sample 1
Traveled: 0.0250 from z0.0013 to 0.0263 on step 5
12:53
Auto TAP performing 5 samples to calculate z-offset on R8 tap using 'rev hop' method
Probe Min: -1.0, Stop: 2.0, Step: 0.005

I ended up having to adjust -0.44 to get a good first layer.

Conor Duffy · Answer 7 · Wed Nov 29 2023 06:05:48 GMT+0800 (China Standard Time)

Adding to the info. Voron 2.4R2 350mm running StealthBurner, CW2 and Tap R8 (with 4 magnets).

Auto TAP Results on R8 using 'rev hop' method
Samples: 3, Total Steps: 12
Probe Mean: -0.0033 / Min: -0.0038 / Max: -0.0025
Measure Mean: 0.0167 / Min: 0.0167 / Max: 0.0167
Travel Mean: 0.0200 / Min: 0.0200 / Max: 0.0200
Calculated z-offset: 1.203
Auto TAP sample 3
Traveled: 0.0200 from z-0.0025 to 0.0175 on step 4
Auto TAP sample 2
Traveled: 0.0200 from z-0.0038 to 0.0162 on step 4
Auto TAP sample 1
Traveled: 0.0200 from z-0.0038 to 0.0162 on step 4
Auto TAP performing 3 samples to calculate z-offset on R8 tap using 'rev hop' method
Probe Min: -2, Stop: 2.0, Step: 0.005

Actual offset after babystepping in-print and saving using Z_APPLY_OFFSET_PROBE

##
## [probe]
#*# z_offset = -0.950

Arkadiusz Miśkiewicz · Answer 8 · Sat Dec 02 2023 02:30:11 GMT+0800 (China Standard Time)

Voron 2.4R2 350mm, CW2 and SB R8 with 4 magnets, umbilical here:

19:24
Offset does not match expected result. Expected between 0.70-2.00, Got: 3.274
19:24
Auto TAP Results on R8 using 'rev hop' method
Samples: 3, Total Steps: 54
Probe Mean: -1.3742 / Min: -1.3750 / Max: -1.3737
Measure Mean: -1.2842 / Min: -1.2842 / Max: -1.2842
Travel Mean: 0.0900 / Min: 0.0900 / Max: 0.0900
Calculated z-offset: 3.274
19:24
Auto TAP sample 3
Traveled: 0.0900 from z-1.3737 to -1.2837 on step 18
19:24
Auto TAP sample 2
Traveled: 0.0900 from z-1.3750 to -1.2850 on step 18
19:24
Auto TAP sample 1
Traveled: 0.0900 from z-1.3737 to -1.2837 on step 18
19:23
Auto TAP performing 3 samples to calculate z-offset on R8 tap using 'rev hop' method
Probe Min: -2, Stop: 2.0, Step: 0.005

when setting manually I use z_offset = -1.380 (or close to that value)

Sami Haahtinen · Answer 9 · Sun Dec 10 2023 06:46:41 GMT+0800 (China Standard Time)

Voron 2.4R2 300mm, CW2 & SB with Tap R8 using 4 magnets

22:42:03 
$ AUTO_TAP FORCE=1
22:42:03 
// Auto TAP performing 3 samples to calculate z-offset on R8 tap using 'rev hop' method
// Probe Min: -2, Stop: 2.0, Step: 0.005
22:42:16 
// Auto TAP sample 1
// Traveled: 0.0300 from z0.0050 to 0.0350 on step 6
22:42:22 
// Auto TAP sample 2
// Traveled: 0.0300 from z0.0038 to 0.0338 on step 6
22:42:29 
// Auto TAP sample 3
// Traveled: 0.0300 from z0.0038 to 0.0338 on step 6
22:42:29 
// Auto TAP Results on R8 using 'rev hop' method
// Samples: 3, Total Steps: 18
// Probe Mean: 0.0042 / Min: 0.0038 / Max: 0.0050
// Measure Mean: 0.0342 / Min: 0.0342 / Max: 0.0342
// Travel Mean: 0.0300 / Min: 0.0300 / Max: 0.0300
// Calculated z-offset: 1.296

Normal offset is around -1.09 so about 0.2mm difference

Joe Calvert · Answer 10 · Sun Dec 17 2023 23:04:46 GMT+0800 (China Standard Time)

Voron 2.4R2 350mm; CW2 & Stealthburner. Using Tap R8 with 4 magnets, Whoppingpochard's Maxwell Kinematic Mounts, and a textured bed.

Probe location at the corner of the bed (310, 310)

$ AUTO_TAP
16:07:09 
// Auto TAP performing 3 samples to calculate z-offset on R8 tap using 'rev hop' method
// Probe Min: -2, Stop: 2.0, Step: 0.005
16:07:09 
// Homing Z
16:07:30 
// Auto TAP sample 1
// Traveled: 0.0350 from z0.0188 to 0.0538 on step 7
16:07:40 
// Auto TAP sample 2
// Traveled: 0.0350 from z0.0200 to 0.0550 on step 7
16:07:51 
// Auto TAP sample 3
// Traveled: 0.0350 from z0.0188 to 0.0538 on step 7
16:07:51 
// Auto TAP Results on R8 using 'rev hop' method
// Samples: 3, Total Steps: 21
// Probe Mean: 0.0192 / Min: 0.0188 / Max: 0.0200
// Measure Mean: 0.0542 / Min: 0.0542 / Max: 0.0542
// Travel Mean: 0.0350 / Min: 0.0350 / Max: 0.0350
// Calculated z-offset: 1.331

Had to babystep -0.305 for a good first layer

Kept for reference: Bad readings due to a loose magnet. Fixed by adding more superglue to the magnet.

14:59:52 
$ AUTO_TAP
14:59:52 
// Auto TAP performing 3 samples to calculate z-offset on R8 tap using 'rev hop' method
// Probe Min: -2, Stop: 2.0, Step: 0.005
14:59:52 
// Homing Z
15:00:41 
// Auto TAP sample 1
// Traveled: 0.2150 from z0.0075 to 0.2225 on step 43
15:01:18 
// Auto TAP sample 2
// Traveled: 0.2050 from z0.0075 to 0.2125 on step 41
15:01:55 
// Auto TAP sample 3
// Traveled: 0.2000 from z0.0075 to 0.2075 on step 40
15:01:55 
// Auto TAP Results on R8 using 'rev hop' method
// Samples: 3, Total Steps: 124
// Probe Mean: 0.0075 / Min: 0.0075 / Max: 0.0075
// Measure Mean: 0.2142 / Min: 0.2142 / Max: 0.2142
// Travel Mean: 0.2067 / Min: 0.2000 / Max: 0.2150
// Calculated z-offset: 3.059
15:01:55 
!! Offset does not match expected result. Expected between 0.70-2.00, Got: 3.059

Also using a kinematic bed mount with probe location at 310, 310. Could be incompatible?

My usual offset is -0.9645

EDIT (update):

Tried again with the probe location in the center of the bed. Still no luck.

15:13:39 
$ AUTO_TAP
15:13:39 
// Auto TAP performing 3 samples to calculate z-offset on R8 tap using 'rev hop' method
// Probe Min: -2, Stop: 2.0, Step: 0.005
15:13:40 
// Homing Z
15:14:16 
// Auto TAP sample 1
// Traveled: 0.1500 from z0.0200 to 0.1700 on step 30
15:14:43 
// Auto TAP sample 2
// Traveled: 0.1450 from z0.0213 to 0.1663 on step 29
15:15:09 
// Auto TAP sample 3
// Traveled: 0.1450 from z0.0213 to 0.1663 on step 29
15:15:09 
// Auto TAP Results on R8 using 'rev hop' method
// Samples: 3, Total Steps: 88
// Probe Mean: 0.0208 / Min: 0.0200 / Max: 0.0213
// Measure Mean: 0.1675 / Min: 0.1675 / Max: 0.1675
// Travel Mean: 0.1467 / Min: 0.1450 / Max: 0.1500
// Calculated z-offset: 2.446
15:15:09 
!! Offset does not match expected result. Expected between 0.70-2.00, Got: 2.446

EDIT (Update 2):

Discovered a loose magnet. Fixing and will update again

Tom S · Answer 11 · Tue Dec 19 2023 02:10:26 GMT+0800 (China Standard Time)

Voron 2.4R2 250mm; CW2 + Stealthburner on Clone Vitalli CNC TAP. z_offset set to 0

AUTO_TAP
Auto TAP performing 3 samples to calculate z-offset on VITALII_CNC tap using 'rev hop' method
- Probe Min: -2, Stop: 2.0, Step: 0.005
Auto TAP sample 1
- Traveled: 0.0200 from z-0.0013 to 0.0187 on step 4
Auto TAP sample 2
- Traveled: 0.0200 from z-0.0013 to 0.0187 on step 4
Auto TAP sample 3
- Traveled: 0.0200 from z-0.0025 to 0.0175 on step 4
Auto TAP Results on VITALII_CNC using 'rev hop' method
- Samples: 3, Total Steps: 12
- Probe Mean: -0.0017 / Min: -0.0025 / Max: -0.0013
- Measure Mean: 0.0183 / Min: 0.0183 / Max: 0.0183
- Travel Mean: 0.0200 / Min: 0.0200 / Max: 0.0200
- Calculated z-offset: 0.432
Offset does not match expected result. Expected between 0.50-1.50, Got: 0.432

The correct z offset is more like 1.4

mizz141 · Answer 12 · Thu Dec 28 2023 17:54:41 GMT+0800 (China Standard Time)

Auto TAP Results on CL_CNC using 'rev hop' method
Samples: 5, Total Steps: 168
Probe Mean: -0.3150 / Min: -0.3212 / Max: -0.3087
Measure Mean: -0.1470 / Min: -0.1470 / Max: -0.1470
Travel Mean: 0.1680 / Min: 0.1650 / Max: 0.1700
Calculated z-offset: 0.651
Auto TAP sample 5
Traveled: 0.1700 from z-0.3212 to -0.1512 on step 34
Auto TAP sample 4
Traveled: 0.1700 from z-0.3200 to -0.1500 on step 34
Auto TAP sample 3
Traveled: 0.1650 from z-0.3150 to -0.1500 on step 33
Auto TAP sample 2
Traveled: 0.1700 from z-0.3100 to -0.1400 on step 34
Auto TAP sample 1
Traveled: 0.1650 from z-0.3087 to -0.1437 on step 33
Auto TAP performing 5 samples to calculate z-offset on CL_CNC tap using 'rev hop' method
Probe Min: -2, Stop: 2.0, Step: 0.005

Chaotic Labs Tap (v1), Stealthburner and Galileo 2 Extruder, Voron 2.4, probed after Homing and QGL, the offset was way off, it's more like, -0.310 and not 0.651 or I just don't know how that offset should be used...