If you have no idea what this is about, search the web for the words "cross-cut sled".

A few days after watching Jonathan Katz-Moses' introduction of his absolutely game-changing cross-cut sled fence jig [1], I woke up from anxious dreams and had the idea for this partially disposable 3D-printed system.

The rails are 12.2 mm wide to fit the miter slots on my little Metabo TS 254 M. You may have to scale or remix for your purposes.

1. Print the T-shaped part twice. Avoid the formation of Z seam on the rail-facing side of the T-shaped part. Print I-shaped rails as required. Slice using elephant-foot reduction.
2. Raise the rails in the miter slots 1-2 mm above table level. I have added a printable 5 mm x 5 mm spacer that fits the deep slots on the Metabo nicely.
3. Butt the T-bars firmly against the edge of the table.
4. Epoxy the sled base plate onto the printed rails. It should be lined up with the T-bars for neatness, but not at the risk of de-butting the T-bars from the edge of the table. Before leaving the epoxy to cure, make sure the T-bars are still firmly butted up against the table.
5. Drill the rails from the bottom, countersink and screw from the top. (FYI, I used an M2 drill and Spax M3 x 12 on an 8 mm plywood base plate.)
6. With the rails now firmly attached to the base plate, butt the user-facing fence firmly against the T-bars. Clamp, drill, countersink and screw from the bottom. (FYI, M3 drill and Spax M4 x 25.)
7. Install opposite fence.
8. Saw off the T-bars.
9. Done. No mathematical formulas were harmed in the process.

FAQ

• Q: The rails don't slide easily?

  • A: They should slide easily before assembly. After assembly, they will probably be a bit stiff. Sand them (I did), grease them (I did [2]), spend the evening moving the sled back and forth (I did) until it slides. Oiling the sled from all sides will also help with friction. A high-tolerance sled that slides easily from the beginning, will be your worst enemy if you need any sort of sub-centimeter(!) precision.

• Q: Error?

  • A: This first build has a 0.16 mm error along a 30 cm edge (not amplified by a 5-cut test). I believe that is equivalent to 0.03 degrees. The improved T-bars along with higher-tech material for the user-facing fence (I used this cut-off I kept since 20 years) should yield easy improvements.The 5-cut-method [3] may be applied retroactively, provided the fence wasn't glued.

• Q: Differences between the pictures and the STL?

  • A: The T-bar in the STL is 3 cm taller than the one I used myself in the photos and has horizontal holes for temporarily screwing the fence.

• Q: How much epoxy?

  • A: Not very much. I initially used too much and had to scrape it off. It only needs to hold until the holes from the bottom are drilled.

• Q: CA glue?

  • A: I have no experience with CA glue.

• Q: What material?

  • A: The print on the photos is in PLA. ASA might be worth a try if you can manage its warping.
• Q: May I commercially?
  • A: You may not. Don't be that person.

Links

[1] - https://kmtools.com/products/katz-moses-jig-and-sled-square-revolutionize-your-woodworking
[2] - https://www.silbergleit.de/
[3] - https://www.instructables.com/The-5-Cut-Method-Made-Easy-a-Foolproof-Process-and/