Correct... concrete cover to the insert is 2" top and bottom. The steel plate is 3" deep (1/2" chamfer on the precast) with the line of fasteners in the middle, and the tensile force is from a guard post welded to the plate. It will be up to the precaster to confirm the slab edge can accommodate the moment.
I'm looking at the trapezoidal block of concrete to resist the load with the epoxy securing the threaded rod.
Yep. Sound in principle, and it’s done this way all the time. Just be careful with the geometry. I gathered from your comments that, in 3D-space, there is a concrete edge that is way too close to the anchor. I could easily have misunderstood.
The total slab thickness is 4". I didn't cause the problem... It's the proximity of the top and the bottom of the slab that's the problem. I'd normally just 'punch' this out on my calculator. There were enough variables and uncertainties that I did a little SMath program so I could change the parameters, quickly. I was presented with the problem; I usually get one or two of these each month. I'd originally looked at using 1-1/2" to the top and 2-1/2" to the bottom. In Canada, for guards, we can use 1/2 the guard forces for loads 'inwards' so I tried to minimise the force. I thought better about it and felt it best to have more concrete cover and deal with the larger force.
??? My original approach was to use a plate top and bottom with a through bolt. I decided to look at it a little more closely to see if I could get the proposed detail to work.
Powder actuated fasteners... Thanks