Guide for Heat-Straightening of Damaged Steel Bridge Members

Chapter 5: Heat Straightening of Flat Plates

5.1 Introduction

The fundamental element of any structural steel shape is the flat plate. Damage to bridge structures involves combinations of these plate elements, bent about their strong and/or weak axes. Understanding the behavior of plates during heat straightening is fundamental to the heat straightening process.

Two studies (Roeder, 1986 and Avent, et. al. 2000) helped define the factors affecting heat straightening of plates. As a result the following observations can be made.

• 650°C (1,200°F) is a practical and safe upper temperature limit for non–Q & T steel.
• Changes in material properties are relatively small when the temperature remains below the phase transition temperature of approximately 720°C (1330°F).
• Plastic rotation, defined as the change in angle of tangents located on either side of the damaged zone of a plate after the completion of a vee heat, is the basic measurement of movement during heat straightening.
• The rotation produced by a vee heat on an otherwise unrestrained plate is directly proportional to vee angle and heating temperature.
• Plastic strain during straightening occurs primarily within the vee heat region.
• Plastic strain is somewhat sensitive to geometry of the plate. However, much of this sensitivity can be attributed to differences in rate of heating and heat flow.
• Due to the difficulty in controlling the many variables associated with heat straightening, the magnitude of movements for individual heats may vary considerably.
• Varying the vee depths between 75–100 percent of the plate width has little influence on the plastic rotation of a vee heated plate.
• Plate thickness and width do not significantly influence plastic rotations, provided sufficient heat is applied to generate a specified consistent temperature within the vee.
• External restraints can significantly increase the movements per vee heat with the movement being related to the restraint force.
• The movement associated with each of the initial heat cycles is often larger than subsequent cycles due to internal restraints developed when a member is damaged severely enough to require a high number of cycles.
• Axial forces can be used as constraining forces, but bending moments are usually more efficient in producing movement.
• The influence of yield stress on plastic rotation is small for mild steel having an F_y between 230–345 MPa (33–50 ksi).

5.2 Variables Affecting the Movement of Heat–straightened Plates

5.2.1 Temperature

One of the most important and yet difficult to control parameters of heat straightening is the through–thickness temperature of the heated metal. Factors affecting the temperature include: number and size of torch orifices, temperature of the flame, speed of torch movement, and thickness of the plate. Studies have shown that knowledgeable practitioners commonly misjudged the heating temperature by 55°C (100°F) and, in some cases, as much as 110°C (200°F). Thus, there is considerable variability in temperature control, even with experienced users.

Figure 38. Influence of heating temperature on plastic rotation for 3/4 depth vee heats and a jacking ratio of 0.16.

The effect of heating temperature can be seen in Figure 38 in which the heating temperature was varied from 370–815°C (700° to 1500°F) in increments of 56°C (100°F). The results establish a regular progression of increased plastic rotation with increasing temperature.

The maximum temperature recommended by most researchers is 650°C (1200°F) for all but the quenched and tempered high strength steels. Higher temperatures may result in greater rotation; however, out–of–plane distortion becomes likely and surface damage such as pitting will occur at 760–870°C (1400–1600°F). Also, temperatures in exceeding 700°C (1300°F) may cause molecular composition changes which could detrimentally change material properties after cooling. The limiting temperature of 650°C (1200°F) allows for a safety factor in this regard. For the quenched and tempered steels, the heat–straightening process can be used but the temperature should be limited to 595°C (1100°F) for A514 and A709 (grades 100 and 100W) and 565°C (1050°F) for A709 grade 70W to ensure that the properties are not adversely affected.

5.2.2 Effect of Vee Angle

The results shown in Figure 38 and 39 also illustrate the effect of the vee angle when heat straightening. The amount of movement is approximately proportional to the vee angle.

5.2.3 Restraining Forces

The term "restraining forces" can refer to externally applied forces, self weight or internal redundancy. These forces, when properly utilized, can expedite the straightening process. However, if improperly applied, restraining forces can hinder or even prevent straightening.

The proper procedure for applying a restraining force is to create a moment tending to compress the stretched area. The ratio of the moment at the vee due to the jacking force, M_j, to the plastic moment, M_p, of the cross section, is M_j/M_p. This term is referred to as the jacking ratio. The effect of jacking ratios ranging from zero to 50 percent with four different vee angles are shown in Figure 39. It can be concluded from this data that plastic rotation is generally proportional to the jacking ratio and the proper use of external loads greatly expedites the heat–straightening process.

Figure 39. Influence of jacking ratio on average plastic rotation for 650°C (1200°F) heating temperatures (lines represent a least squares curve fit).

In summary, parameters which have an important influence on the plastic rotations produced by vee heats are: (1) vee angle, (2) steel temperature, and (3) external restraining force. In the usual range of three–quarters of the plate width or greater, the depth of the vee appears to have little effect. Likewise, the plate dimensions are of minor significance as long as the heating patterns attain the desired temperature.

5.3 Analytical Development

Two general approaches have been used to develop an analytical procedure for predicting member response during a heat–straightening of a plate damaged by bending about the major axis. One approach involves finite element/finite strip thermal and stress analyses including inelastic behavior. The stress and strain equilibrium is the summation of small steps and considers the influence of the non–uniform temperature distribution. This approach is lengthy, is only possible using computer techniques and a typical analysis for a single vee heat can require extensive set up and computer time.

The other approach considers the global action of the vee. The goal of the analytical development is to obtain an equation which can be used to predict the angle of plastic rotation produced by a vee heat. Avent, et. al. (2000) developed this type of model using the following assumptions: (1) longitudinal plastic strain occurs only in the vee heat zone (and in a reflected vee about the apex for partial depth vees); (2) at any specified distance from the neutral axis of the plate, the strains in the longitudinal direction are constant over the zone of the vee; (3) the planes defined by the sides of the vee remain planes after heating and rotate about the apex of the vee; (4) confinement during heating is not perfect single axis along the longitudinal direction (i.e., some longitudinal movement during heating is assumed): (5) the permanent strains occur within the inner two–thirds of the vee with an effective vee angle of two–thirds the actual angle, (6) the plastic rotation varies linearly with jacking ratio, (7) perfect confinement is equivalent to a 20 percent jacking ratio, (8) the zero jacking force equals 60 percent of the perfect confinement case and (9) the heating temperature is 650°C (1200°F). The resulting formula for plastic rotation, φ, (angle change due to a single vee heat) with zero jacking force is

(Eq. 5.1)

where θ is the vee angle. The jacking force is incorporated by the introduction of a jacking force factor

(Eq. 5.2)

and the plastic rotation is

(Eq. 5.3)

The formula compares well to the experimental data and is the first simple formula available that includes the parameters of heating temperature of the steel and magnitude of restraining force (jacking force). The form of this approach also lends itself to the behavior of rolled shapes, axially loaded members, and composite and non-composite girders.