I thought I'd toss out a suggestion for a possible change to the way BikeCAD handles curved seat tubes.  The way it currently stands, if the builder is capable of measuring effictive BB offset, the TT/ST miter angle comes out perfectly.  A problem I've personally been running into is that the input parameters [BB ctr to bend center (A) and Bend Radius (R)] likely can't be accurately and reliably measured by smaller builders with simple tools.  I know about how much springback I get when bending a 1.375" CWSR tube but it varies a bit from batch to batch and it's impossible (for me) to get an accurate actual bend radius in part because of springback and due to the small area of the bend vs the radius.  I can also get close on the A dimension but can't quite nail the perfect bend point and without an accurate radius it's moot anyway.

My work around is to setup a separate drawing with a straight seat tube and the measured offset.  I take the DT/ST miter angle from the drawing and then add the actual bend angle from the ST to get the miter angle and it's been working out great.  

I haven't given much consideration to the math and programming involved, but I think if we could keep the offset (B) input and add the actual bend angle to the input list it would eliminate a few steps and save time for those of us who are doing a lot of curved seat tubes.

IMO A and R are sure great for finnessing the appearance of the drawing for customer communcation but they're not doing much (for me) in terms of efficiency.

Thanks Brent.  As always, I think BikeCAD is an outstanding tool and I appreciate all the work you put into making it even better.

Sean