You want to consider midspan deflection as well as the "aesthtics" of the installation. The maximum stress levels in the Zick design guidline is not the only criteria you should use.
Saddle deflection plays a role here too. Too much rotation can pry anchor bolts
The propane tank people have probably pushed the l/d thing to it's limits (longer tanks are cheaper on a cost/unit volume basis). I believe that they have gone to about 10 or 12:1........See this link
A horizontal vessel on two saddle supports is was treated by Mr Zick as a simply supported beam. There are two things to consider (1) the deflection of the mid-span and (2) the longitudinal compressive stress at the top of the cylinder at the mid-span. ASME limits the maximum compressive stress to the B stress value found in the external pressure charts. If you want a more detailed analysis, you could try CodeCalc, as it has the provision for analysing horizontal vessels on saddle support in accordance with the British PD 5500 code which assesses the compressive stress at the top of the mid-span more comprehensively. Most software does not consider the elastic line deflection of the vessel, so the rotation of the of the vessel at the saddle supports is not computed. The rotations usually turn out to be very small.
The deflections and rotations of the cylinder can be estimated using the usual beam formulae.
I have found that CodeCalc and PVElite accomplish the analysis comprehensively.
