Below are some tips if you run into trouble or want to make adjustments.
Before you venture into model edits, be warned this was a frantically rushed design-build begun merely a week out from the conference (and that includes most of the component sourcing!). As a result, some my modeling best practices flew out the window. So if you’re courageous enough to fire this up in Solidworks, please accept my humble apologies for any pain that ensues. (I did start on a clean, redesign-from-scratch in Onshape but never got around to finishing it).
Parts placed directly on the build platform will suffer some compression in the first few millimeters printed. The effects diminish as the build platform ascends enough to stop “squishing down” the PDMS coating at the bottom of the tank. As a result, the overall height of the part (along the Z axis) may come out shorter than intended. The relative distance to features in the compressed portion (measured along the Z axis) will be inaccurate (generally less than expected), although further up the print accuracy is restored. This is one reason Preform-generated supports raise parts up a certain minimum height above the base.
You sometimes also get outward “bleeding” in the X and Y directions along the edges of the base, as a result of the printer overcuring the first few layers. This means your part may have a very thin “lip” along its edge where it meets the build platform, although the effect is generally a lot less prevalent than compression.
Here’s how the cross-section of a vertical beam with three equally spaced holes in it might come out. Note the effects are greatly exaggerated for the purposes of illustration:
Where possible, parts in the project have been designed so the accuracy of their overall printed height isn’t critical, and features near the build platform are crafted such that they can absorb the effects of compression without unwanted consequences to fit or aesthetics. The beveled edges on the top of the Lid, and bottom of the Base, are a great example - they’ll come out more “squat” in some printers verses others, but are large enough that you don’t really notice, and it doesn’t harm anything if the printer “eats” some of their layers. Further, their bases have been preemptively stretched a little to compensate - so while the curves look a little oblong in the design file, it helps them come out closer to round when printed.
There are a few exceptions where the negative effects couldn’t be avoided. The Glass panel is quite thin in order to get good transparency without post-processing. If you find your printer “eats” it entirely, you might need to thicken it. Similarly, the thickness of the Butterfly logo can be tuned to achieve a perfectly flush fit.
The Display panel relies on varying the cut depth around the progress circle in order to create the gradient effect. If the minimum thickness is too small you might wind up with a hole in it, but if it’s too thick it will block out too much light and the dark portion of the gradient won’t be visible.
Finally, the depth of the Tower might need to be tweaked to gain a flush fit with the rear of the Base and to clamp the battery snugly between it and the Base.
Here are some dimensions in the model you can play with if needed. Features named “ExtraBase” are generally the ones intended for compression compensation. There are a few of them scattered around (sorry!) with values based on intuition, trial and error.
- ExtraBase variable in Cover.sldprt. Controls how much extra shell thickness is added to the top surface of the lid (compared to the sides).
- D1@ExtraBase in Base.sldprt. Extra thickness added to bottom of Base.
- D1@ExtraBase in Tower.sldprt. Extra thickness added to back surface of Tower; tweak if it isn’t flush with the back of the Base when the two are snap fitted together.
- D1@ExtraBase in Glass.sldprt. Extra thickness added to Glass.
- D1@ExtraBase in Display.sldprt. Extra thickness added to Display. Decrease to make the progress circle brighter, increase if you’re getting holes in the Display.
- D6@ProfileSketch in Tray.sldprt. Height of tray feet (oversized to take into account compression).
- D1@ExtraBase-PartialSinceDesignTakesIntoAccountCompression in Tray.sldprt. Extra thickness added to bottom of Tray.
- D1@BackplateAndExtraBase in Collar.sldprt. Adjusts thickness of Collar backplate.
- D1@ExtraBase in Button.sldprt. Extra height added to Button.
- D1@ExtraBase in Wiper.sldprt. Extra thickness added to Wiper.
- D1@ExtraBase in Crystal.sldprt. Extra thickness added to base of Crystal.
- D1@ExtraBase in Butterfly.sldprt. Extra thickness added to base of Butterfly.
Keep in mind the degree of compression can fluctuate between different locations on the build platform, and after swapping in a new build platform or tank.
Adjust these if snap fits are too tight or too loose. I purposely left the Tower and Base fit a little loose since I wasn’t able to print enough of the latter before the conference, and would be doing last-minute assembly mixing in untested parts run on a different (Form 1) printer by a new friend of mine in Boston. (So there’s the explanation for you really keen recipients who inquired about the feint hex markings on the bottom of your BabyForm2; you’ve got a vintage, limited-run edition!)
Be creative. e.g. If the battery fit is too loose and your LED’s don’t light up, try sliding a small fold of paper between the battery and the retaining post in front of it. That’s a lot quicker than a model edit.
- D1@TopCutSketch in Cover.sldprt. Extent of the rear cutout in the top of the Lid. Increase for more clearance to top of Tower and cartridge cap when Lid is opened. You can also tweak D1@FilletTowerClear. After adjustment, you might also need to increase D4@ColumnSketch in Tower.sldprt and fix the engraving lines on top of Tower.
- FilletBackCornersForOpenClearance in Cover.sldprt. The fillet dimensions control the amount of rounding on the back corners of the lid; increase if the cover scrapes against the top of the base when opening / closing. (See also, TestMotion.sldasm).
- D1@DisplaySlotSketch in Base.sldprt. Size of the Display panel retention slot.
- Offsets in these sketches / features impact how tightly the Tower fits into the Base. Due to rushed modeling, they can be tricky to adjust:
- RearCutSketch in Base.sldprt
- Indent1 feature in Tower.sldprt.
- D4@ColumnSketch in Tower.sldprt (I’d start here).
- D2@BattLockSketch in Base.sldprt. Adjusts how tightly the Base pushes the battery into the Tower. Decrease if your LED’s keep going out.
- D19@BattRetainingRightPost in Tower.sldprt. Controls offset from battery retention post to approximate top surface of battery. Increase if the battery is too hard to insert. See also, D1@BattLedge which controls the (thickness of the ledges which grip the battery from below.
- RearClearance variable in Tower.sldprt. Gap between back panel of Tower and Lid.
- CoverClearance variable in Tower.sldprt. Clearance between Tower column and Lid.
- D1@Led2HoleAndLocatorSketch in Tower.sldprt. Size of LED opening.
- D1@MainSketch in Button.sldprt. Diameter of button. Increase if it’s too loose in the Collar.
- D1@Led1BracketSketch in Tower.sldprt. Size of LED opening. Increase by 0.1mm if it’s too hard to insert the blue LED.
- ButterflySketch in Butterfly.sldprt. Offsets adjust snap fit tightness of Butterfly.
If you make adjustments to the STL’s and need to perform layout in Preform, I suggest you rotate parts around the Z axis so they are not parallel with the wiper edge of the platform. This reduces stress during the peel process. It’s especially important for the Lid; if its walls are parallel to the wiper edge you’ll get a “crusty” buildup where they meet the build platform (unfortunately I made this mistake on the batch I brought to the conference).
Preform might prompt you to confirm units for some of the smaller parts are in millimeters (Collar and Button). Also be sure to select the correct base surfaces when orienting your parts on the build platform (zoom in for the small parts to be sure). The Wiper in particular is easy to get wrong - it’s deceptively asymmetrical, and the side furthest from the guide rib should go down.
Most parts have one or more release chamfers along their perimeter to ease removal from the platform; you should try and point these outward to make them easy to access with the scraper.
For those curious, here’s how the built-in jig alternative works: Stick paperclip through hole, up column, and seat end in slot. Bend up toward front of tower, slip over built-in jig, bend down, then rotate the whole clip left over the locking post. You can tell I’m still experimenting to figure out the quickest method, and in practice I’m actually starting to prefer this one.
LED Viewing Angle
For keeners: It’s about 13° from the LED centerline to the outer edge of the progress bar cutout, and 25° to the illuminated button ring
No LED’s? No problem, you can fake it:
If you need to get inside your BabyForm2 (e.g. to change the battery), you can use a paperclip to open it up. Hold the Tower in one hand (gripping the sides of the vertical column between your fingers) and the base in the other hand. Insert the paperclip into the hole on the side of the unit, and press hard until you feel the snap tab start to release. You can sort of lever the tab backwards a little to push one side of the Tower backward. Continue holding the Tower and Base apart so the side you released doesn’t slip back in, then repeat the process using the hole on the other side of the Base to pop out the Tower.
When modeling, try to avoid introducing circular references. Here’s a rough dependency graph:
A few of the sketch entities are drawn freehand without firm relations. When you resize certain components (e.g. height of the lid) these sketches may need to be manually corrected. Dangling references might also be introduced after such operations (e.g. where sketch entities have been converted) - the offset entities on the feet of the Tower can be particularly troublesome. If the Onshape redesign is ever completed those issues will go away.