The PIPO-FE and VIBIC finite-element computer codes, developed and updated over the past 30 years, are used to calculate heat exchanger tube flow-induced vibration (FIV) response. PIPO-FE includes a linear forced-vibration analysis of heat exchanger tubes subjected to all major flow-induced excitation mechanisms, namely fluidelastic instability, random turbulence-induced excitation and periodic wake shedding. VIBIC is for both linear and non-linear transient dynamic simulations of heat exchanger tubes. When used to simulate a tube with clearance supports (non-linear case), VIBIC calculates tube wear work-rates to aid in the prediction of tube fretting-wear damage. All the excitation mechanisms included in PIPO-FE analyses can be simulated in VIBIC. In addition, VIBIC can model friction forces between a tube and its supports, squeeze film forces produced by the resistance of the fluid opposing the relative motion of the tube and supports, and constant loads. An important application of these codes is the analysis of the susceptibility of a heat exchanger tube to vibration damage. These codes may be used at the design stage to assess a new heat exchanger, or during the operational stage to investigate a tube failure and determine if the damage was caused by vibration. If a vibration problem exists, then the codes can be used to assess the effectiveness of any proposed design modifications. To properly assess tube vibration damage, the codes must predict vibration response accurately. This paper documents the validation process of code predictions against measurements from three flow-induced vibration experiments conducted at Chalk River Laboratories: 1. A single-span cantilever tube bundle subjected to two-phase air-water cross flow; 2. A single-span cantilever tube bundle subjected to single- and two-phase Freon cross flow; and 3. A single-span U-bend tube bundle subjected to single-phase water and two-phase air-water partial cross flow. PIPO-FE and VIBIC code predictions for fluidelastic instability ratio and the response to random turbulence-induced excitation are compared to each other for each of these three experiments. The predictions from the two codes are in good agreement. In addition, the predictions for frequency, damping ratio, fluidelastic instability ratio and the response to random turbulence-induced excitation from both codes are in reasonable agreement with the experimental results.

This content is only available via PDF.
You do not currently have access to this content.