Start with perpendicular holes.
1) Assume that the table is exactly the same size as the box. What pressure and how much flow is needed to float the box? If both the box and table are perfectly flat and nonporous, the flow would be quite low. In the real world, nothing is perfectly flat, so there will be a minimum flow. You can estimate that flow by assuming a worst case flatness, then calculating the flow through the gaps at sufficient pressure to float the box. It is better to run a test and measure the pressure and flow.
2) There is a pressure drop through the holes. The pressure under the table is higher than the pressure to float the boxes. How much higher is a function of the hole size.
3) When you have a hole size and pressure under the table, then you can calculate the flow through the holes that are open (not covered by the box). You will need to investigate the effect of hole size - smaller holes at higher pressure versus larger holes at lower pressure.
4) The total flow will be the sum of the flow that floats the box plus the flow through the open holes. You can use large holes (several mm diameter) with a blower or small holes (about 1 mm or less) with compressed air. I have seen air conveyors with both methods, you need to find the best solution for your specific situation.
5) NOW you can use angled holes. Angled holes will move the boxes. I once saw literature from a company that was selling air conveyors that would move the product uphill, so it can be done. If you choose the large hole design, you can explore using louvered sheet similar to the image below. The louver shape will be different, and the sheet metal needs to upside down from that shown:
If you choose drilled holes, try drilling at a 45 degree angle, then experiment from there.
6) If you get enough experience with air float conveyors, you should be able to calculate air flow, air pressure, hole size, hole angle, and hole spacing. Those calculations will start with experimental data, then use standard fluid flow calculations to extrapolate that data to your specific requirements.
7) This is a situation where you absolutely, positively need to do experiments. Then compare the experimental results to your calculations. Then do more experiments based on what you learn. Expect to do the calculate >> experiment >> calculate >> experiment cycle several times before you fully understand how to design this type of conveyor.
8) This sounds like a fun project, so enjoy.
