Calibration, Belt Tension, and Z offset results

Over the last week I added 10 foot 12ga. U channel under the hourglass wood frame to tie the left and right extensions to the same plane, tightened up the bolts, and added 1.5" risers under the belt mounts.

Unfortunately I forgot the part about bottoming out the z axis betore calibration but ran two passes with the calibration and belt retraction values set to 900. I just left the Z axis where it was after my last test run. The results were:

Flex measurement: TLBR: 6.188 TRBL: 4.075] (added this value during edit)
Calibration values:
Fitness: 1.160927543957941
Maslow_tlX: -23.1
Maslow_tlY: 2302.7
Maslow_trX: 3591.7
Maslow_trY: 2309.3
Maslow_blX: 0.0
Maslow_blY: 0.0
Maslow_brX: 3593.3
Maslow_brY: 0.0

Flex measurement: TLBR: 3.846 TRBL: 5.130]
Calibration values:
Fitness: 0.92977191021453
Maslow_tlX: -25.6
Maslow_tlY: 2297.6
Maslow_trX: 3594.0
Maslow_trY: 2303.7
Maslow_blX: 0.0
Maslow_blY: 0.0
Maslow_brX: 3593.9
Maslow_brY: 0.0

Then I bottomed the Z axis, which put one of the belts on a slight upward angle to the anchor. I didn’t know about the Z offset (100, 56, 34, 78) defaults. Using the 900 force values I couldn’t get calibration working. It would loop, belts wouldn’t retract, and the frame flex values were all over the place.

I then changed the z offsets to 66, 22, 0, 44 and increased the z axis until the lowest belt was on the same plane as the mount. Increasing belt retraction and calibration tension to 1200, I ran a clean calibration with the following results:

Flex measurement: TLBR: 2.345 TRBL: 3.304
Calibration values:
Fitness: 1.160482013167061
Maslow_tlX: -24.9
Maslow_tlY: 2276.9
Maslow_trX: 3598.8
Maslow_trY: 2292.1
Maslow_blX: 0.0
Maslow_blY: 0.0
Maslow_brX: 3603.0
Maslow_brY: 0.0

I am still having issues with the CNC not releasing belt tension at the end of calibration, and also with some of the belts not retracting once I manually release them from the anchors. I raised the retraction in increments until at 1700 the last two belts parked in the garage (fully retracted).

-Scott

2 Likes

is this a 4.0 or 4.1 machine? especially in the 4.0 arms, it’s possible for the
motor to be too close to the center and not rotate smoothly, causing the need
for high retraction limits. (getting the belt packed tightly is hard

but as long as the belts are tight as you are calibrating, a high setting for
full retraction isn’t a big problem in itself, but it could be in indication
that the arm isn’t moving well

is it always the same two arms that have trouble retracting?

note that there is calibration limit and retraction limit. the retraction limit
is when pulling all the way in, the calibration limit is what’s used to decide
the belts are tight for calibration (it can usually be a little lower as you
don’t have the over-full spools rubbing)

David Lang

This is a change in the new firmware. I need to re-enable releasing tension all the time.

1 Like

is this a 4.0 or 4.1 machine? especially in the 4.0 arms, it’s possible for the
motor to be too close to the center and not rotate smoothly, causing the need
for high retraction limits. (getting the belt packed tightly is hard

It is a 4.1. I upgraded the 4.0 with the kickstarter kit and added the new parts and bearings. So far I haven’t had a clean calibration using 900s. I need to save the clean config and experiment some more when I have time.

is it always the same two arms that have trouble retracting?

Random it seems. Sometime I can click Retract Belts and one more will retract, but I need to set a force >900 to get them all in.

note that there is calibration limit and retraction limit. the retraction limit
is when pulling all the way in, the calibration limit is what’s used to decide
the belts are tight for calibration (it can usually be a little lower as you
don’t have the over-full spools rubbing)

I will try the next test with both values starting at 900, and increase calibration if I get the loose belts during the first 4-5 calibration step.

I am also curious why when calibration starts to loop, I can power down the Maslow and the console log still increments. Isn’t this log coming from the Maslow, or is my laptop doing calculations? Thanks!

1 Like

Scott wrote:

I am also curious why when calibration starts to loop, I can power down the
Maslow and the console log still increments. Isn’t this log coming from the
Maslow, or is my laptop doing calculations? Thanks!

The laptop is doing the calculations.

David Lang

2 Likes