Testing on Turkey Slices
To define the effectiveness of the CNC and discover possible faults, we test ARTS 2.0 on turkey slices.
Since Lightburn is not made for milling with spindle, the G-code has to be post-processed through a python code as per now.
Summary
Slice | Settings | Status | Comments | Future change |
---|---|---|---|---|
1 | Speed: 10mm/s Power: 100% | Failure | Too fast. | Slow down. |
2 | Speed: 2mm/s Power: 100% | Failure | Did not cut all the way through. | Lower the spindle. |
3 | Speed: 1mm/s Power: 100% | Success | Mill bit diameter too high | Change mill bit / Raise spindle. |
4 | Speed: 1mm/s Power: 100% | Failure | Freezer too warm, slice thawed and loosened | Let the freezer be cold enough before cutting. |
5 | Speed: 1mm/s Power: 100% | Success | A lot of spooning | Make a suction. |
6 | Speed: 1mm/s Power: 100% | Failure | Clamping was not able to keep slice suspended against downward force. | Make a baseplate with bigger hole. |
7 | Speed: 1mm/s Power: 100% | - | - | - |
8 | Speed: 1mm/s Power: 100% Layer height: 0.50 | Success | The last sample did not fall down | |
9 | Speed: 1mm/s Power: 100% Layer height: 0.50 | Success | The last sample did not fall down | Increase layer height for bigger last cut |
10 | Speed: 0.6mm/s Power: 100% Layer height: 0.75 | Success | The first sample did not fall down. Had to change mill bit to Ø1.0mm | |
Slice 1: Failure
CNC cut too fast. Mill pushed the slice around slightly.
Slice 2: Failure
Samples had a difficulty coming loose.
Mill pushed the slice around slightly.
Slice 3: Success
This time the mill cut well through the samples, but the mill bit diameter was quite large, and so quite a bit of material was lost.
The mill also left a bunch of spooning, which argues for a suction.
The samples did not slice quite as preferred after having being cut loose from the slice, one small sample even jumped away.
Slice 4: Failure
Cutting went well. We tried letting the CNC take multiple passes to improve the cut quality. I suspect the first layer is cut too deep.
Also added priority for the cross in the grid samples. Might be better to treat them as separate samples and sample them one by one instead of cross plus square.
(No image)
Slice 5: Success
Cutting went well. We tried letting the CNC take multiple passes to improve the cut quality. The mill was raised slightly since last cut, this resulted in a nice quality cut, but also a lot of spooning.
Slice 6: Failure
We tried cutting out samples without a bed, to let the samples fall down when done, this did not work as hoped. The clamping was not strong enough to resist the downward force. It was however able to resist the shear force.
(No Image)
Slice 7:
We modified the cutting procedure to cut the matrix as four separate squares, instead of as a cross and a square. This will take longer to cut, but will provide one sample at a time.
Image
Slice 8:
First test slice with new setup.
Temperature data found in: arduino_data_log - Copy 130420241814
Image
Slice 9:
Temperature data found in: arduino_data_log - Copy 140420241921
Slice 10:
Temperature data found in: arduino_data_log - Copy 150420241409
Image