Most of the FDM printers can get to similar resolution. Standard is 0.4mm nozzles and 0.2mm layers. 0.1mm Layer is fairly easy to go and dropping the nozzle size to 0.3 works and you can get 0.2 (tough to get it working well)
Here are some examples of the same components made in three different materials
- white: DirtyPCBs' SLA
- purple: Shapeways' strong and flexible nylon
- red: Prusa printer's PLA, 0.4mm nozzle, 0.15mm step, 100% infill
In the pictures below, the surface finish of the PLA and nylon is easy to see, but that of the SLA is so smooth the part appears to be out of focus!
The first component is a scope probe tip designed to (successfully) minimise the inductance of a ground lead. The dimensions and finish are non-critical, and the internal nibs are an interference fit over the probe.
The nylon is the best material since the outer walls deform when the probe is inserted, gripping well but allowing the probe tip to be easily removed without damage to the probe. The PLA works, but requires careful reaming out for it to be an interference fit, and its hardness an roughness damages the probe. The SLA is intermediate. Overall the PLA is sufficient to prove the concept, but it unsatisfactory otherwise.
The second component is a handle for a Tek P6013 HT probe. The key design feature is that the 2mm pitch thread which screws into the HT part of the probe.
The SLA variant works perfectly, inparticular the thread works surprisingly well. The PLA variant is a failure: the thread doesn't screw in, and the slicer automatically modified the shoulder's overhang.
The final component is a collet for a Tek 24x5 squirrel cage cooling fan. The design points are the compression of the collet inside the fan, by being pulled to the left with a screw.
Lightly compressing the PLA variant with fingertips caused it to fracture along a layer boundary, so it is unlikely it could survive the tension exerted by the screw. That's not entirely surprising, PLA (and similar materials/depositions) was never going to be a good material for this task. The SLA variant works well, and will probably be OK in the medium term: the SLA scope probe handle is fine after 4 years, and the collet won't get very warm.
You can not make any conclusions with a printer that is clearly not set up correctly let alone based on a sample size of one. The collet example with the PLA shows retraction issues (random wisps and blobs in between the collet jaws), over extraction (bulging on the sides or possibly belt issues) and I suspect the temperature of the extruder is 5-10 degrees to low (shot is a little fuzzy but the bright shiny lines show clean separation on most of the fin so poorly bonded on the sheared layer). As to the option to use FDM the jaws are to small in size with that layer orientation let alone with a badly setup printer.
The principal settings were as defined by the person that maintains the printer and has used it far more than I have.
I am perfectly happy to believe that
with some fettling better results could be achieved. I still doubt that the strength would be sufficient, and I am not interested in fettling with
somebody else's printer.
As to threads anything M12 course and above works fine in PLA, M10 Course with a tweak to the thread clearances works too but nothing under that. An example below close up of my Fluke 5616 RTD case which is an M14 picked because it was close at hand.
That thread still looks crude; it might be sufficient in some circumstances, but not for the purpose I indicated.
There is no argument about the surface quality of SLA over FDM. Putting up one off 'fdm inappropriate' designs/models and poorly setup printer to back your ongoing position with very limited use cases in particular hands on use remains an issue.
Of course PLA is an inappropriate material! Which bit of "PLA (and similar materials/depositions) was
never going to be a good material for this task" did you miss?
I did not make any global statements. I provided a small number of data points, and am content to leave others to draw their own conclusions.
You, OTOH, appear to be oversensitive when anybody points out that SLA printers have characteristics not shared with other additive manufacturing processes.
I have no intention of re-opening the previous nastiness w.r.t. 3D printers, and will not do that.