dmadison
commented
3 years ago My vote is to leave it as is.
Aside from the manufacturing headaches of trying to maintain two versions of the flaps, I think the concern for the extra 0.2 mm of the 'pins' is misplaced. In my experience the flaps don't get stuck because the pins hit the side of the enclosure, they get stuck (if they get stuck) because the flap "twists" in the spool holes along Y, so the top corners of the flap hit the side of the enclosure. Keeping the spool slop loose is far easier than trying to tighten up all of the tolernaces with the laser-cut enclosure.
With the current version on master I've had no issues with the 3.2 mm flaps getting stuck in my 3.0 mm enclosures, though I did need to back the acceleration off for some of the motors (8ab30be is a similar change).
It's also worth bearing in mind that manufacturers come and go, and with the changes of #93 the design should now be flexible to varying thicknesses (including 3.2 mm). For Yanks with access to 1/8" material and a laser-cutter the 3.2 mm flaps are still perfect.
scottbez1
The flap dimensions have always been "wrong" though they generally work ok. This issue is to track comments/considerations around fixing those dimensions, which would be a breaking change (and a potentially confusing one, given the changes will likely be around 0.5mm in aggregate).
Historically it was believed that Ponoko sold 3.2mm thick MDF (because that's what their website incorrectly stated when this project started) so the notches on the flaps were designed to be 3.2mm wide to match. Even if the MDF was actually nominally 3.2mm thick, it was a mistake to set the notches to that exact width, as any variability in thickness would mean that the flap edges and flap pins could extend past the spool where they can more easily get caught on the sides of the front window.
However, since the MDF and acrylic are actually both 3mm nominal thickness, this means that an accurately cut flap can stick out by 0.2mm on either side of the spool (or more, if the real material thickness is below nominal thickness, such as Elecrow wood which is closer to 2.8mm in my experience).
The challenge with fixing this is that it will require both a flap revision and enclosure revision, and users will have to be very careful to make sure they are using matching revisions as they will look nearly identical but will not be compatible. It also requires ordering a new tool for the custom flaps I sell (~$120 one-time fee) and maintaining inventory of both variants for a while; there is also an ongoing risk that the supplier may accidentally use the wrong tool for future orders due to the visual similarity, or that I may accidentally pack the wrong flaps when fulfilling orders.
The major question is whether it's worth dealing with the hassle of fixing this at this point considering the potential for confusion and complication with inventory and compatibility. As long as the spool is well-centered in the window I have not personally had issues with the flaps catching on the window since there is built-in tolerance/slop between the spool and window, but others have occasionally reported flaps getting caught (though it isn't 100% clear that this was the reason for that). Basically it's unclear if this theoretical problem is actually a critical enough issue in practice to be worth fixing, especially considering the practical complexity of fixing it.
An alternative "fix" would be to officially recommend using shim washers to maintain spacing if needed, and potentially increase the gap between the spool and window sides a bit, which would not be a breaking change and has no compatibility/inventory-control complications.