Strange Scaling Issue with 3D Prints

[JSModels]

I've experienced an unexplainable (at least by me!) "scaling" issue with 3D printing 7mm bridge (L1) chairs. I put "scaling" in quotes, because I don't know what else to call it!

This is definitely more of a printing issue rather than a Templot issue, but I hope it's still ok to post here. Apologies too for being quite long-winded...

A bit of an explanation of the background to this. I created an STL file from Templot a few weeks ago when I got my printer set up, following @James Walters video, of 8 pairs of L1 bridge chairs (7mm scale). This was then sliced in Chitubox using the settings I had found worked best for the AnyCubic Basic Grey resin (not the water washable variety), and printed in that resin*. I was very pleased with the results; the chairs fitted the Peco code 124 rail pretty well; one or two were a little tight, but most slid on beautifully. Here's one of the rafts with the rail slid in:

The photos show up very obvious contour lines, but in reality these aren't really noticeable in the flesh, and don't bother me. Note that some of the bolt heads look a little odd, but that's down to a bug that caused the chair models to be duplicated (now resolved by @Martin Wynne ).

The only problem with these chairs was that they are quite brittle, which was fully expected with this resin. I had ordered some tougher resin (AnyCubic ABS-Like Resin Pro 2) but that hadn't arrived.

Cut to the last few days, and the time came to test the new resin. I didn't expect that the new resin would work perfectly with the same settings, but as a baseline test I printed a batch using the exact same sliced CTB file as previously used.

Somehow, and I really can't work out how, the result is compressed in the Z axis:

The original print on the left, the ABS-Like print on the right.

I did wonder if it was because the exposure time was too short (I think it's likely to need 2.5 to 3.0 seconds ideally) and perhaps some intermediate layers failed, but if that was the case then surely the print would just fail completely, wouldn't it? It was definitely the same CTB file, as there's only 1 file relating to the chairs on the USB stick.

Not only that, but it's also stretched very slightly in the X and Y axes:

I'm completely baffled by this. Maybe I'm missing something really obvious, but I really can't understand how this could happen?


*Printer is Geeetech Alkaid. Layer height 0.05mm, Bottom layer count 3, Exposure time 2.0s, bottom exposure time 20s

 

[Martin Wynne]

Printer is Geeetech Alkaid. Layer height 0.05mm, Bottom layer count 3, Exposure time 2.0s, bottom exposure time 20s

@JSModels @Steve_Cornford

Hi Jonathan,

We have had the mystery of missing lower layers before, I can't find the link at present. I don't think we ever got to the bottom of it (!).

But I suggest try increasing the bottom exposure time. For the ABS-Like resin on the Mars2P and the Alkaid, I use:

exposure time 2.0 seconds, or 2.2 seconds in cold weather.

bottom layer count 5.

bottom exposure time 35 seconds.

which have always worked well.

Also I warm the build plate under the hot tap and wipe it dry immediately before use.

cheers,

Martin.

 

[Steve_Cornford]

Was there a significant difference in ambient temperature?
Having cleaned the resin tank (due to change of resin) did you repeat the bed levelling?
Steve