Bitsetter Calibration

[treno (Travis)]

I’ve recently installed a bitsetter on my Shapeoko 3.
After changing tools the z height is off by 1.75 mm…

Procedure:
Carefully zero the z height with some tool.
Use the “Load New Tool” function to put in a different tool.
Watch the bitsetter, measure the tool.
Use the “JOG” function to move to 0,0. (allow the bitsetter to measure again)
Carefully lower the new tool until it just touches the workpiece… the z height at this zero location is reported at 1.750 mm
This is repeatable…
How can I adjust the offset of the bitsetter?

Here’s the debug info just in case.
Machine: Shapeoko
GRBL Version: 1.1f

Carbide Motion 618
Machine Controller State: INIT
Grbl Cycle: Idle

{
“bitRunnerEnabled”: false,
“bitSetterEnabled”: true,
“bitSetterX”: -1.0,
“bitSetterY”: -399.5,
“bitZeroType”: 1,
“connection”: 1,
“lastConfigSent”: 0,
“lastProbeIndex”: 0,
“lastSizeOption”: 1,
“odometer”: [
{
“cleared”: false,
“minutes”: 77.33558333333328,
“startDate”: “2023-10-29”,
“travelX”: 199967.1035078466,
“travelY”: 69872.3267493248,
“travelZ”: 40323.249931275845
}
],
“showCarbideRouterRpm”: true,
“spindleType”: 0,
“travelX”: -830.0,
“travelY”: -430.0,
“travelZ”: -100.0,
“version”: 0,
“workOffsetX”: 787.2749998092652,
“workOffsetY”: 352.42499980926516,
“workOffsetZ”: 18.400000000000006
}

0 = 10, Step pulse, microseconds
1 = 255, Step idle delay, milliseconds
2 = 0, Step port invert, mask
3 = 6, Direction port invert, mask
4 = 0, Step enable invert, boolean
5 = 0, Limit pins invert, boolean
6 = 0, Probe pin invert, boolean
10 = 255, Status report, mask
11 = 0.02, Junction deviation, millimeters
12 = 0.01, Arc tolerance, millimeters
13 = 0, Report inches, boolean
20 = 0, Soft limits, boolean
21 = 0, Hard limits, boolean
22 = 1, Homing cycle, boolean
23 = 0, Homing dir invert, mask
24 = 100, Homing feed, mm/min
25 = 2000, Homing seek, mm/min
26 = 25, Homing debounce, milliseconds
27 = 3, Homing pull-off, millimeters
30 = 24000, Max spindle speed, RPM
31 = 0, Min spindle speed, RPM
32 = 0, Laser mode, boolean
100 = 40, X steps/mm
101 = 40, Y steps/mm
102 = 40, Z steps/mm
110 = 10000, X Max rate, mm/min
111 = 10000, Y Max rate, mm/min
112 = 5000, Z Max rate, mm/min
120 = 500, X Acceleration, mm/sec^2
121 = 500, Y Acceleration, mm/sec^2
122 = 400, Z Acceleration, mm/sec^2
130 = 845, X Max travel, millimeters
131 = 850, Y Max travel, millimeters
132 = 100, Z Max travel, millimeters

Maintenance:

Travel (km) 0.20, 0.07, 0.04, 1.29 Hours since 2023-10-29

Queue Empty

[WillAdams (William Adams (Carbide 3D))]

You have a belt-drive Z-axis?

That probably warrants being calibrated for belt-stretch:

Calibrating the machines over long distances

The typical technique for machine calibration: https://my.carbide3d.com/faq/belt-stretch/ depends on precision measuring tools, and is limited by the reach of the tools in question. EDIT: It also depends on access to the MDI — if using Carbide Motion v6 it will be necessary to enable this in Settings | Options: [image] It should be possible to cut strips from a suitable blank of material, one along each axis — so we set up a square file which matches an available blank of material: [ima…

but for Z you’ll need to cut a series of steps and carefully measure them.

[treno (Travis)]

I do, but it’s not clear to me how belt stretch would manifest as exacatly 1.75 mm difference between two z measures in a repeatable pattern.

[WillAdams (William Adams (Carbide 3D))]

The BitSetter should be a relative measurement.

What happens if you re-measure the tool w/o actually changing it?

Please try updating to the current Carbide Motion:

https://carbide3d.com/carbidemotion/beta

If you still have the problem with that, I’ll set up one of my belt drive machines and try to duplicate and look into this with you.

[treno (Travis)]

Just ran the exact same procedure as above except not physically changing the bit, and the machine returned to the correct z zero location exactly.

Trying the beta now

[SLCJedi (Josh Tenny)]

Being that far off on a bit change tells me it’s not a calibration issue. If it were 0.010" or under, maybe.

9/10 times it’s us humans breaching the procedure it expects us to do.

How are you initially setting your XYZ? BitZero? Paper method?

[treno (Travis)]

I just ran through this procedure 3 or 4 times with 6.22 and the machine returned perfectly to zero every time.

[GizzardLipp (Brian)]

Remember folks > no one piece is exactly the same.

[LWSIV (Lester Shaw)]

Hi Travis,
I’m running a 3XXL (May 2019) with one of the last Blue HDZ units, extruded aluminum bed, HDPE waste board, and a 1.5KW water cooled spindle. One of my typical projects starts with a 1/4" precision pin for setting zeros (Bit Setter and Bit Zero V1) then runs thru 3 clearing bits, a v-carve, then a 1/4" O flute for cutout. After 5 bit changes I’m still leaving the top surface of the blue tape intact. This lets me know the tool changes are not drifting position or I’d have onion skin or damaged tape. It does sometimes catch at the edge where there wasn’t CA present to hold the two pieces of tape together and the dust collection lifts the edge.

[GizzardLipp (Brian)]

Interesting. do you need a JOB?

[LWSIV (Lester Shaw)]

Thanks for the offer but I have a job I like most of the time .

[GizzardLipp (Brian)]

Talented for sure , always one open here if you get bored.