Without having model/dimensions I cannot give exact answers but here are some comments:
- my last bellows project was very similar. Started fabric and landed in steel redesign after redesign
- 66" compensators are not cheap. If this line is large portion of the total price you better get the design right.
- In worse scenario you may have to change plant layout
- CAESAR II is quite good with compensators if you model them correctly. There are mistakes in the documentation and the technical recommendation from COADE has had couple misunderstandings. 2 rod universal ends can rotate around one axis quite easily and if this is prevented in CAESAR II model your results are far out.
- Universal compensator is mainly intended for lateral movement. If you force it to take axial movement then the rods do not carry the pressure trust at all. They are limit rods in case of anchor failure
- Vertical force Fy in the blower nozzles is from mass from the duct up to some point in the horizontal section and mainly from the lateral spring force from the universal compensator.
- Lateral spring rate of a universal can be anything. It depends on what elements have been used and how long is the nipple (centre pipe). 66" is normally designed case by case and you do not have them in literature. I made a quick calc. This has 2 layers 0.8 mm, 4 plus 4 convolutions 1000 mm long nipple and the lateral spring rate is only 230 N/mm
- CAESAR calculates the pressure trust if you enter the diameter. In this case the diametre is around 1710 mm and the pressure trust around 160 kN (I used duct OD 1650 mm). Pressure trust is carried by the rods if elements are not compressed and rods loose)
- Instead of universal compensator, which you force to take axial also I would consider 3 hinged compensators in 3 pin system. You should have EJMA code. It has good examples.
- If you go to http://www.jat.co.za/engineeringgallery.htm the second last image is for "universal" with axial compression and the one before that is 3 pin hinge compensator arrangement.
- On this size your rods are about 1 1/4" for stability reasons. Less than 1" is sufficient from stress point of view (2 rods). If you let the rods loose by compressing the unit then it doesn't really matter. If you need to start worrying about rod spring rates then you have wrong design. I have done one design where the strain was ruling (Image is on my site). Problem is the rods connect brackets, which connect to duct. Everything is deforming. How much? You do not know without doing full FEA.
- If your rods are in compression your CAESAR II model is wrong
- Don't forget compensator mass. Hinges, brackets and reinforcing can be heavy.
- For me you need to calculate using design parameters.
- I use metric system. My software calculates also imperial but not my brain.
- It would be best if you had somebody to help who knows bellows design and how CAESAR II works. Not easy to find. There is address below

I hope above helps.