Right, that figure he posted, you may not have seen it, shows the OP's shoulder slope criteria is not a simple direct relationship to lane slope.
thanks for all the input.
no, the shoulder doesn't match the lane slope all the time, only in certain instances where the lane slope is greater than 6%. between -1% and 6% the shoulder stays at 6%. from -1% to -5% the shoulder varies in relation to the lane slope. if the lane slope is -3.2% then the shoulder slope is 3.8%. at -6% and greater it becomes crowned. it is a pita. like i said, i've got a profile for the shoulder but any change in alignment or profile really becomes time consuming.
I stand corrected. I had missed the *.png that was posted. Definitely an ideal situation for SAC and really a rather simple one at that. I'd still use Parameter References though and have the shoulder SAC have an input parameter of the lane slope and then the shoulder is calculated and generated from there.
Yes, that would work through most of the criteria range, except for those steep lane slopes at the top of the criteria range, where it looks like he has a crowned shoulder that is smoothed with a fillet, which SAC could easily handle.