Arcticdesert
Mechanical
- Sep 15, 2003
- 9
I am working on calculations for an air cooler box header to appendix 13 FIG 13-2(a)(2). I am having a hard time understanding which calculations to use for the ligaments and how they relate to what I am seeing on this job.
Both the tube and plug sheet have 96 holes in 4 rows of tube holes in columns 2 inches apart. The 2nd and 4th rows are staggered over from the 1st and 3rd columns by 1”. The rows are 1.5 inches apart.
For the plug sheet the holes are threaded and uniform 0.875” diameter. The plug sheet is 0.75” thick. Per 13.4 (g) the rules of UG-53 apply. It would seem these are on a diagonal line as shown in Fig UG53.4. UG-53(d) states efficiencies are determined using Fig UG-53.5 and Fig UG-53.6. I used Fig UG-53.6. With angle Theta of 56.3º, P` of 1.803, P of 2, d of 0.875. This gives an equivalent longitudinal efficiency of 65%.
The tube sheet is 1.00” thick and has 0.766 dia. holes. The tube sheet holes have two cuts 1/8” wide and 3/32” deep separate by 1/8” space in between. The space and cuts are centered in the tube sheet. The tube will be rolled in. Since the tube sheet holes have two different diameters, the ligament was calculated by appendix 13.6 rules for multidiameter ligaments using the diagonal spacing. P` of 1.803.
An equivalent dia. Of 0.789 leads to 0.562 ligament efficiency for membrane stress.
For bending stress the neutral axis is found to be the same at 0.5t. Moment of inertia for the composite areas 0.086 and the equivalent diameter of the hole is found to be 0.771. Ligament efficiency for bending comes in at 0.573.
I have calculated the gross area membrane and bending primary stresses using formulas of appendix 13.7(b). The rectangular header box is made of A240\316L plate, and the top, bottom, endplates are all 0.500” thick. L=50” Allowable stress 16700 @ 300º. E=1 no butt welded joints.
Calculations are exceeding both the tube sheet and the plug sheet net area membrane stress by about 40%. Also the net membrane and long side bending at both the midpoint and corner for the plug and tube sheet are exceeding the higher allowable 1.5SE for the tube sheet and plug sheet.
On the job we are preparing to fabricate new replacing an existing exchanger with all the same dimensions but have increased the thickness of the tube sheet.
I am now going back through my calculations looking for errors.
Does my approach to the ligament efficiency seam correct? What part do plugs installed in a plug sheet contribute to ligament efficiency? Are there any tricks or tips to increasing the efficiency or reducing the membrane stress without changing the wall thicknesses? Stiffeners?
Thanks in advance
ArcticDesert