Accurizing your Y-Wing build starts with reading all of the reference material, then reverse-engineering from the confirmed donor nurnies, then getting it right down to the millimeter...
Two of the reference books confirm that the ILM studio prop Y-Wings were 38cm wide by 71cm long. As many have discovered, the LENGTH problem is easy enough to solve -- just add more spacers, cut longer T-bars, or modify (as Scott Peterson and DaveG both have so kindly done) the Open Source Files to stretch things out.
But the WIDTH of the Y-Wing is a different problem. To accurize your build using DaveG's open source code and armature, you'll need a different Tee Plate.
Old Tee-Plate (32.4cm) below, New Tee Plate (34.0cm) above
After six iterations, I am of the following opinion:
A 33.8 cm wide Tee Plate will give you an exact width of 38cm after build-up (from tip of Left outer engine nurnie to tip of Right outer engine nurnie), IF you're using the Roco 1/87 anti-tank obstacle trapezoids as your outer engine nurnies.
A 34.0 cm wide Tee Plate will give you an exact width of 38cm after build-up (from tip of Left outer engine nurnie to tip of Right outer engine nurnie), IF you're using DaveG's super-accurized trapezoids (which is what I recommend, and what I am using). His trapezoids are 1mm shorter than the Roco trapezoid height.
If you want one of these, I have a limited number of 34.0cm Tee Plates available in
the Project Run thread. I may later produce 33.8cm Tee Plates if there is interest. If you have original Tee Plates that were 32.4cm wide, and you haven't started building yet, now is the time to get the right one. Other than a trip to the Lucasfilm archive itself with a metric ruler, I don't think I can get more accurate than this.
One question remains: did the reference book photographer/measurer use measurements that were "rounded up" to the nearest cm, or even possibly "round down"? If the Y-Wing is not precisely 38.0 cm wide from outer nurnie to outer nurnie, then I'm guessing they rounded up by 4mm or less. This estimation is based off my reverse-engineering of the nurnies, and of the gap between the end of the Hummel wing plant-on and the Kettenkrad track link at wings edge. This means if my measurements are off, then the tee-plate width is too long by either 2mm or 4mm, depending on trapezoids used. Still, to my mind, that's closer than we were a year ago, which was too short by 1.6cm
Many Bothans died to bring you this information...