Influence of the hot dip galvanized coating on design

The presence of coatings on high strength bolts, and any coatings on structural members will need to be considered in the design phase.

The characteristics of any bolt system that should be considered in design include the slip factor of the mating surfaces, the fatigue behaviour of the joint, any bolt relaxation, the effect of the coating on the nut stripping strength, and the torque/induced tension relationship in bolt tightening.

Slip factors affecting mating surfaces

Bearing type joints are not affected by the presence of applied coatings on the joint faces, so galvanizing may be used without affecting design strength considerations.

In a friction type bolted joint all loads in the plane of the joint are transferred by the friction developed between the mating surfaces. The load which can be transmitted by a friction type joint is dependent on the clamping force applied by the bolts and the slip factor of the mating surfaces.

Sweep, whip, or flash blasting

Sweep, whip, or flash blasting are terms for a common technique for roughening a hot dip galvanized surface.

The aim is to expose the zinc iron alloy layer on the structural steel surface in the area of the connection without removing too much zinc.

This can be done after galvanizing by following the techniques described in AS/NZS 2312.2 Clause, AS/NZS 4680 Appendix I, or SSPC-SP 16.

  • Blast pressure 275 kPa (40 psi)
  • Abrasive Grade 0.2 – 0.5 mm (clean garnet)
  • Angle of blasting to surface no greater than 45°
  • Distance from surface 350 – 400 mm
  • Nozzle orifice diameter 10 – 13 mm of venturi type

Slip factors of galvanized coatings

Australian Standard AS 4100, Steel structures, assumes a slip factor of 0.35 for clean as-rolled steel surfaces with tight mill scale and a surface free from oil, paint, marking inks and other applied finishes. It allows the use of hot dip galvanized surfaces in friction type joints and requires the slip factor used in design calculations be based on test evidence in accordance with the procedures specified in Appendix J of the Standard. When using the test procedures in AS 4100, tests on at least three specimens are required, but five is preferred as the practical minimum. Research conducted at the University of Newcastle showed that galvanized steel that has been blasted in the bolt locations prior to galvanizing to expose the zinc-iron alloy layer will achieve a slip factor of at least 0.35. Further, the follow-up research showed the slip factor was dependent on the coating structure rather than the way the coating was produced.

In recent years, the Australian research on galvanized surfaces has been repeated and expanded in Europe and the USA. This has confirmed the Australian work and resulted in changes to several international Standards and specifications for design of bolted connections. The new standardised slip factors are consistent with the University of Newcastle research results and EN 1090-2 now assumes a slip factor for sweep blasted hot dip galvanized surfaces of 0.35. A higher slip factor value of 0.4 is available for surfaces that have a layer of inorganic zinc silicate applied after blasting. These changes should allow engineers to assume slip factors in design and remove the extra cost of testing. 

There are minor differences in slip factor calculations between Europe and Australia and the slip factors shown in Table 23 (Europe) should be checked for compatibility. Appendix 1: Comparison of Australian & European methods for calculation of slip factors provides a comparison which shows the European data for hot dip galvanized surfaces that are sweep blasted should be able to be used in the Australian context without modification and deliver a slip factor of 0.35.  

European Research

Work by Prof. Dr.-Ing. Natalie Stranghöner and others for the EU SIROCO Project using the test methods in EN 1090-2 Appendix G, shows that the slip factors of galvanized and non-galvanized surfaces can be substantially improved by controlled blasting. The blasting of galvanized surfaces must be performed in a manner that provides the required roughening to expose the alloy layers of the galvanized coating although care must be taken to ensure that excessive coating is not removed.

The EU SIROCO Project provided recommendations for the surface treatment that may be assumed to provide the minimum slip factor according to the specified class of friction surface (Table 23).

Note: the European research did not address the slip factor of ‘as rolled’ surfaces and the slip factor is shown as 0.2 for these surfaces which differs from the Australian assumption of 0.35. Simple blasting of as rolled steel can deliver slip factors in excess of the Australian slip factors by using the European methods.

Table 23: Slip coefficient values from EU SIROCO Project

Table 23: Slip coefficient values from EU SIROCO Project

Both the European and US research has shown wire brushing of the galvanized surfaces does not significantly increase slip properties with the European research only recommending sweep (whip) blasting to increase the slip factor.

In recent years, the Australian research on galvanized surfaces has been repeated and expanded in Europe and the USA. This has confirmed the Australian work and resulted in changes to several international Standards and specifications for design of bolted connections.

Figure 24: Cyclic stress v slip for fatigue samples with hot dip galvanized plates and bolts
Figure 24: Cyclic stress v slip for fatigue samples with hot dip galvanized plates and bolts

Fatigue behaviour of bolted galvanized joints

While a hot dip galvanized coating behaves initially as a lubricant, it has been shown in fatigue tests carried out by Professor WH Munse that after the first few cycles galvanized mating surfaces tend to ‘lock up’, and further slip under alternating stress is negligible (Figure 24). He found that the amount of slip rapidly decreases from first to second, and then the fifth stress cycle. Munse noted further indications of ‘lock up’ behaviour when the joints were disassembled and observed galling of the galvanized coating in regions where there had been high contact pressure. 

Where no initial slip can be tolerated, a reduced slip factor must be used in design or the slip factor of the galvanized coating may be improved by sweep blasting.

Bolt relaxation

The possible effect of bolt relaxation, caused by the relatively soft outer zinc layer of the galvanized coating on the member, must also be considered. If the zinc coating has flowed under high clamping pressure, it could allow loss of bolt extension and hence tension. This factor was also studied by Munse and re-affirmed in later studies by Heistermann of Lulea University of Technology. Munse found a loss of bolt load of 6.5 percent for galvanized plates and bolts due to relaxation, compared to a 2.5 percent loss for uncoated bolts and members. This loss of bolt load occurred in 5 days and little further loss was recorded. This loss can be allowed for in design and is readily accommodated. 

Research in Europe and the USA is continuing on this subject.

Torque and induced tension relationship in tightening

The relationship between torque and induced tension in tightening is dependent on bolt and nut thread surface finish, thread surface coatings, and conditions of lubrication.

Galvanized coatings on threads both increase friction between the bolt and nut threads and make the torque/induced tension relationship much more variable.

The effect of lubricants on galvanized threads is significant. The torque/tension relationship shows much reduced variability, and it becomes possible to tighten in excess of the minimum tension without danger of bolt fracture (Figure 25).

Figure 25 shows the torque/induced bolt tension relationship for galvanized, and lubricant coated galvanized grade 8.8 M20 high strength structural bolts. With as-galvanized fastener assemblies, there is a wide scatter in induced tension at any one torque level, and torque cannot be used to provide a reliable method for gauging the required minimum bolt tension, as specified in AS 4100, before bolt fracture occurs. Bolt failures in torsion could result from the high friction between the as-galvanized bolt and nut threads. Accordingly, prior editions of AS 4100 did not recognise use of the torque control method for tensioning galvanized or zinc plated bolts, and the current edition prefers the part-turn tightening or direct tension indicator tightening methods.

Figure 25: Torque/induced tension-relationship for M20 high strength structural bolts, galvanized and lubricant coated, and as galvanized