Re: Foundations for a 3-pinned arch stadium

I'm overwhelmed by the responses, which I think are very wise.  I was talking with an engineer/supplier from Ontario today, who has 30 years experience with large-span arches in wood.  He agrees that it is very dangerous to rely on passive pressure, and usually uses tie-rods.  What concerns me is that there seem to several examples of these structures which do not have tie-rods.
For a PEMB using a moment frame, there is some redundancy, but with a 3-pinned arch, there is none, and any movement of the supports has a direct  influence  on the  top pin  position.

I am considering a combination of tie-rods for the horizontal thrust, and rock anchors or helical anchors (I don't know the depth of the rock yet) to help with the overturning moment caused by the vertical separation between the pin at the base of the arch and the level of the tie-rod, which will be maybe 20 ft lower.
There is not enough land available to extend the footings far enough to eliminate the moment on the footing.
 

From: Kevin Below [mailto:kbofoz@gmail.com]
Sent: Sunday, October 07, 2007 7:28 PM
To: seaint@seaint.org
Subject: Foundations for a 3-pinned arch stadium

 

For an indoor sports stadium with 200-ft span, I have seen several examples using huge curved glu-lam beam sections about 6 ft high, at about 30 ft spacing.  These beams are assembled on-site into a 3-pinned arch configuration (a pin at each support, and a pin at mid-span.)
The horizontal reaction at the base pins is enormous, and exceeds the vertical reaction.  For example, 470 kips horizontal, and 405 kips vertical.  The horizontal reaction is provided mostly by passive soil reaction on the perimeter wall.
Is this good enough ?  Any thoughts on that ?
Is the passive reaction going to allow movement before it mobilizes ?   Enough to be noticeable ?
What about seismic effects on the soil behind the wall ?  Will it cause the soil to move under the pressure from the arches ?
What about a tie-rod (it would need to be big) under the sports surface, tying the 2 sides together ?

Kevin

kbofoz