Personally once the strength has been maximised by the other things your are trying to do, in order to get strength in the Z direction to avoid snapping off or pulling apart, rather than including glue and a nail I would hollow out the part at the failure point, then print an inner section (either square or two semi circles) in a different orientation - then just slot it inside the part - that should hopefully stop it from being able to be bent and snapped.
To avoid it pulling apart - I would probably just run a long bolt through the centre too with is then tightened up with a not and washer to stop it pulling apart.
Sounds like you interpreted that perfectly. Look for the section the printed perfectly all the way around, and stay below that on the Max Volumetric Speed in the filament profile in the slicer.
I think the best way to improve strength on this part is to print it in the strongest orientation for your forces (green arrow horizontal on the plate). More plastic (gussets and infill) in the current orientation only buy you small gains in strength. You need big gains.
In order to get the job done I asked a friend who does this kind of stuff for a living to give me a hand. His design is massively beefed up and the part that previously broke is no longer the weak point by a long shot:
That design had some clearance issues so I grafted the key elements to my design, which ends up frankenstein-esque in comparison, but also gets the job done:
The key parts were to do the best job bracing against the front of the support board (the brown box) and to get a tall vertical gusset onto the side of the breaking section. In addition, I added an internal I-beam to get more wall loops.
I did find my friend’s design approach inspiring and realized I am relying too much on operations on volumes and not enough on sketches…
I’m attracted by the suggestions made here to break the part up and print each portion in its best orientation or to insert a slice printed in the best orientation. I made a failed attempt at something like that and realized that it’s not as easy as it sounds 'cause getting the interface between pieces to hold together well isn’t trivial. Or at least I didn’t find it trivial.
I’m still a bit stuck on the question “how do I ensure I get as good a Z-adhesion as possible”. It looks like this is going to take some printer/slicer settings fiddling and a whole bunch of test pieces. Maybe I should switch back to PETG and try the annealing too (doesn’t work with ASA)…
Thanks to everyone who contributed suggestions! (And feel free to add more!)
I print lots of Polylite ASA and have found Flow to be ~0.95 and the pressure advance with a 0.4mm nozzle to be in 0.04-0.05 range going down to 0.02-0.03 with a 0.6mm nozzle. I find the auto flow calibration to be total garbage so I always use Orca Slicer manual calibrations.
The PA differences between nozzle sizes is rarely discussed and threw me for a loop when I started using the 0.6mm nozzle. PA is really only needed for speeds >25mm/s so that is why that one test you did at 10mm/s looks considerably better in regards to the gaps.
Also, there are several Youtube videos that prove adding carbon fiber to any filament will greatly reduce the layer strength/adhesion while increasing the impact strength. ASA also has comparatively poor layer strength. Regular PETG or PLA is probably the best choice for this project.
