MATERIAL JOINT THICKNESSES FOR Fasteners in Shear

American Fastener Journal, Jul/Aug 2007 by Barrett, Richard T

Q How do I size the sheet thickness for fasteners that are carrying mostly shear loads?

A This depends on both fastener and sheet strength. The balancing act is to make sure that the fasteners are stronger than the sheet. I'll give you some illustrations and guidelines.

Background!

When the design involves shear as the main load, as in a riveted panel, or bolts holding a sheet assembly together, the fasteners need to be selected more on their strength than on their tension strength. If the fasteners are not stronger in shear than the joint materials, they will fail before distributing the shear loads. Combined shear and tension loads are then compared to the total fastener strength.

Interference fit fastenersi

For interference fit fasteners, such as rivets, one of the joint sheets has pilot holes for the hole pattern. The sheets are clamped in an assembly fixture, and the pilot holes are used to match-drill through holes for the rivet installation. The rivets are then installed with interference fits. The key element here is to make sure that the rivets are strong enough in shear to fail the sheets in bearing.

HOW ARE BEARING STRESS ALLOWABLES DETERMINED?

Bearing stress allowables are empirical, and are determined per M1L-STD-1312 by measuring a sheet bearing yield and failure for a single shear lap specimen or a double shear specimen. The load is then divided by Dt to get a bearing stress allowable. This distribution is shown as P2 in Figure 2. From the figure, D is the hole diameter and t is the sheet thickness. In reality, the actual distribution is closer to that shown in Pl. However, P2 is the one used (for convenience) to determine the bearing stress allowables. The bearing allowables are usually 1.5 to 2.0 times the tensile yield and tensile ultimates for the sheet material. In fact, you can conservatively use 1.5 times tensile yield or tensile ultimate fora material where no bearing allowables are given.

Bolt hole tolerancesi

Bolt holes have tolerances both on location and size, in order that two mating pieces can be drilled on separate fixtures and still have the hole patterns match up well enough to bolt together. The hole locations have a tolerance of at least plus or minus 0.010 inch, with cast holes having much higher tolerances on both location and size. Drilled holes up to 0.375 inch diameter are oversize by at least 0.031 inch, and sizes above that are oversize by 0.063 inch. If the fastener pattern is in shear, the gaps around the fasteners must close before loading can take place. This scenario is illustrated in Figure 1. Some holes have to yield in bearing before others can pick up any load. If the sheet is stronger in bearing than the fastener is in shear, the fasteners will fail individually as each one is loaded up.

Calculating shear loads on a pattern of fastenersi

Assuming that the fasteners are stronger than the sheet, we can now calculate the total shear loads on a pattern of fasteners. Referring to Figures 3 and 4, we have a bracket with an eccentric shear load. The fasteners are in a symmetrical pattern, so we don't need to calculate a centroid. Divide R by n (number of fasteners) to get the direct shear component (P^sub c^). The moment about the centroid is M=Re. Now we measure the radial distance from the centroid to each fastener (r^sub n^). Now we square each r^sub n^ and add them to get Σr^sub n^^sup 2^. This is analogous to a moment of inertia. Then we can calculate the load on each fastener with the formula: P^sub e^= Mr/ΣSr^sub n^^sup 2^, where P^sub e^ is in lbs. The fasteners farthest away from the centroid get the highest loads. The two components can now be added vectorically, as shown in Figure 4(c) to get the total shear load.

Note that all of these fasteners have equal areas. Although having different sizes is not good, the sizes can be ratioed to the most common one, if need be, and the method of analysis will still work.

DO WE CONSIDER SLIDING FRICTION?

Since friction forces between clamped parts are very difficult to determine, we usually do not consider friction loads. They will subtract an unknown amount from the shear loads.

CONCLUSIONS

* The assembly torque tension loads must be combined with the shear loads in an interaction formula to get the total fastener loads.

* Fasteners in shear must be critical in bearing (yield some holes) to distribute shear loads.

* Sliding surface friction is a helper of unknown value.

* Match-drilling of parts makes shear distribution much easier.

* Bearing stress allowables are empirical, and little study has ever been done to improve the process.

REFERENCES

1 NASA RP-1228 Fastener Design Manual by RT. Barrett, 1990

2 Aircra/t Structures by D.I. Peery, McGraw-Hill, 1950

3 Handbook of Bolts and Bolted ]oints, edited by I.H. Bickford, Marcel-Dekker, 1998

Richard T. Barrett retired in June 1997 from NASA Lewis Research Center in Cleveland, Ohio after 34 years service and six years in the industry prior to joining NASA. At NASA, he was a consultant in fasteners, materials, corrosion, welding, and mechanical design. He has authored NASA RP-1228 "Fastener Design Manual" and written a chapter for the Bolting Handbook by Marcel-Dekker Publishing Co. Barrett has formed Barrett Engineering Consulting and now does consulting work and lectures on fastener design. He may be contacted at 440.235.1499.