Essentially, my project here is a 3x5x9 inch block with the long axis in the x-direction, 5 inch in the y-direction. The stock is the whole final dimension of the block, as nothing is removed from the faces of the block, except for the slots and fillets. The zero is done with Bitzero on the upper front left corner. It is all cut with a ball-end cutter. When cut, the project is accurate in all dimensions, with two exceptions, resulting from one extra offset coming from somewhere I can’t find.
The CAD is with Fusion using a 2D adaptive for roughing and 2D contour for finishing. The Carbide 3D “machine” is chosen for the machine in Fusion manufacturing.
The whole job is produced about 1 mm in y-direction away from the y-zero. That is, everything is being cut as though the stock block were positioned 1 mm further back in y than it was. (And yes, the cutter goes exactly to correct x, y and z (over that corner) when sent there after zeroing, so it is zeroed correctly. As a result the project is not symmetrical, as viewed from the x-end. The block gets 3 slots milled into it and the top edges of the block and slots are filleted in the x-direction.
There are two problems. First problem is in the case of the two fillets on the front and back edges of the block, respectively. The fillet in the front is cut 1 mm too far back, so it is not tangent to the front face of the block, but cut into the top of the block. This produces a small negative fillet there. The fillet in the back, since the cutting is all taking place 1 mm too far back (higher in y) gets its initial roughing passes too far back to even touch the block.
The second issue is that the width of the block left between the front face of the block and the front slot is wider than it should be and its opposite in the back too narrow. All of that logically results from this cutting taking place too far back in y. The relative dimensions within the block are all correct, so it’s not related to accuracy, which I’ve checked and is very accurate.
BTW, I’ve been trying to figure this out for 6 weeks and have test cut is 5 times with the same result every time. I went through every setting in Fusion in the tool paths, and Settings and the NC. I could find to try and find any sort of offset. Can’t think why there would be. Tried many different experiments with Fusion settings to see if a difference would occur. Nope. It produces exactly in Simulation and the tool paths look great. It also looks correct in Carbide Motion (latest version), I even tried my best to read through the Carbide 3d Fusion “machine” file code. There’s a tough read.
I did an accuracy check project to see if that would have this problem, cutting a square array of square slot patterns in a board and measuring and the spacings and dimension and there was no offset added; all accurate to within 0.2%. That one uses only 2D Contour.
Any ideas as to why a ~1 mm offset in y is being added to my project? Thanks.
