I’ve finally got calibration above 5, and here’s a little table of what I went through getting there.
Flex anywhere, whether in the frame or the anchor points (or both) really impacts calibration.
| Frame and Anchors | 3x3 | 5x5 | 7x7 | 9x9 |
|---|---|---|---|---|
| Flexy Frame and Anchors | 1.70 | 0.98 | 0.65 | 0.48 |
| Rigid Frame, Flexy Anchors | 1.53 | 0.91 | 0.64 | 0.48 |
| Rigid Frame, Flexy Anchors | 1.65 | 0.89 | 0.66 | 0.49 |
| Rigid Frame, 3xRigid Anchors, 1xFlexy Anchor | 1.73 | 1.01 | 0.75 | 0.55 |
| Rigid Frame and Anchors | 1.99 | 1.03 | 0.79 | 0.58 |
Can you shed more light on what flexy anchors looked like versus your rigid anchors?
That’s excellent data. Its really cool to see how you were able to change the frame and see the results reflected in the calibration values.
Flexy anchor
It’s basically a spacer. The real support comes from the bolt going through the frame.
The inside of the aluminium box section is reinforced with some timber.
So in effect the bolt that runs through the cable end is only supported on one side. This would be similar to floor mounts for some people.
Rigid anchor (before being mounted onto the frame)
It’s screwed in place on three sides. Although not every screw hole was used.
Now the bolt is supported above and below the cable end, and it just doesn’t move.
I am experiencing same sort of thing-- I had a 4x12 table, so I put 8foot 2x4’s at ends… with 2feet hanging over… however, I can see them flex during calibration. Going to change them to 2x8’s. The anchor points are ones I designed and at least those seem quite solid.
So to add to your table of values/contribute to the data:
2x4’s with 2feet hanging off table: 1.26 (3x3); 0.76 (5x5); 0.54 (7x7); 0.42 (9x9)
Finally finished a calibration with the new 2x8’s-- so clearly better, but not quite as much as I’d hoped. Can’t see any flex at all in the 2x8’s but here are the results:
2x4’s with 2feet hanging off table: 1.26 (3x3); 0.76 (5x5); 0.54 (7x7); 0.42 (9x9)
2x8’s with 2feet hanging off the table: 1.72(3x3); 0.81 (5x5); 0.61 (7x7); 0.48 (9x9)
because the 2x8’s were wider I was able to utilize every screw in the anchor brackets, which I did.(so 6 screws per bracket).
It also shows that a 4x12 foot table(table bult on 2x6’s with 3/4" birch plywood top. So it is quite solid.) with 2x8’s is not even as good as the original poster’s flexy frame and flexy anchors!!
I guess I can run two 2x4x12’ from anchor to anchor-- or maybe some sort of triangle support to make the hour-glass style table. Open to suggestions.
That’s what I did
I didn’t like the amount of noise the belt to the top left anchor point was making. So when the chance came to ‘raise’ the anchor point relative to the frame I did that. It’s been ‘raised’ about 54mm. And thanks to the way that it has been raised, it has also been moved ‘up’ and ‘left’ further than before. So recalibration was required.
Here’s the results, not as good as last time. However, I have not yet made the relevant Z offset adjustment for the TL arm. Once I’ve done that and recalibrated, I’ll amend this table.
| Rigid Frame and Anchors | 3x3 | 5x5 | 7x7 | 9x9 |
|---|---|---|---|---|
| ‘Raised’ Top Left Anchor 1300 | 1.60 | 0.85 | 0.63 | 0.47 |
| ‘Raised’ Top Left Anchor 1500 | 1.61 | 0.89 | 0.64 | 0.49 |
And yes, there’s significantly less ‘belt noise’ now. AKA “you can hear the wear and tear”.
