A Novel Approach for Improved Longitudinal Stability of Multi-Stage
Launch Vehicles, 18-R8108

Printer Friendly Version

Principal Investigators
David L. Ransom
Steven T. Green
Benjamin A. White

Inclusive Dates:  10/01/09 – Current

Background - A record number of launch vehicles (rockets) are in development at this time. This includes NASA's Ares I and V launch vehicles, as well as the privately funded Taurus II, Falcon 1 and 9, and the New Shepard vehicles being developed by Orbital Sciences Corporation, Space-X, and Blue Origin, respectively. At the same time, expertise in some of the more specialized areas of vehicle development is lacking in the industry, as engineers from the previous development programs leave the workforce. In the area of feedline and structural dynamics, only two experts remain, and they are both currently retired and working as consultants. This provides SwRI a unique opportunity to expand involvement in the launch vehicle business by re-developing the skills and institutional knowledge required to provide the expert guidance and support each of these development programs will need in the coming years.

This project is focused on the subject of "pogo," which is a nickname for a fluid/structure interaction phenomenon that can lead to excessive vibrations in the payload/crew compartment of any liquid-fueled launch vehicle. Pogo vibrations occur when the launch vehicle structural mode is coupled to a feedline mode through the engine thrust feedback. This coupling occurs only at periods of flight, when the frequencies of these two modes coincide. A pogo accumulator is often used to prevent the occurrence of such vibrations by modifying the modal characteristics of the feedline system. The compliance of the accumulator is controlled by varying the pressure of the gas volume in the device, such that the desired system frequency is maintained. For this project, researchers intend to investigate the use of such an accumulator to also provide launch vehicle structural damping for solid-fueled first-stage thrust oscillations unrelated to the pogo phenomenon.

Prior to developing this concept, it is necessary to develop a more accurate flow-resistance model for the pogo accumulator. Past experience on several development projects has demonstrated a significant disconnect between the amount of resistance designed into the accumulator and the actual amount achieved in flight. Therefore, another objective of this research project is to develop an experimentally validated flow resistance model.

Approach - This project is divided into three major tasks. As described above, the ability to accurately design the flow resistance of an accumulator to achieve a desired system damping is in need of significant improvement. This is considered critical to the success of the proposed concept and is therefore the first step in this development project. An experimental facility similar to a rocket propellant feedline system has been designed and constructed for the purpose of testing various combinations of mean flow turbulence and accumulator flow resistance.

The system consists of a three-inch diameter flow loop with a circulation pump, large water tank, dynamic test section and a hydraulic pulser. The dynamic test section consists of two accumulators and a five-foot straight section of pipe. The dynamic characteristics of this section can be tuned by adjusting the gas pressure/volume in the accumulators, and by adjusting the flow resistance in one of the accumulators, which represents the pogo accumulator. The forcing function is generated by the hydraulic pulser, which can generate pressure pulsations in the range of 2 to 50 Hz with dynamic mass flow rates approximately 5 percent of the mean flow. The test data developed from this test rig is then used to validate an updated flow-resistance model. This model accounts for the frequency characteristics of the turbulent flow, thus better representing the true flow passing into and out of the pogo accumulator.

The second major task is to develop a prototype of the second-stage accumulator for first-stage structural damping. In this experimental project, the flow loop is disconnected and the dynamic test section is capped at the ends. The pulser is replaced with a shaker attached to the end of the test section. The test section is supported on a flexible support, which is then tuned to achieve a longitudinal resonance within the frequency range of the test program (0 to 50 Hz). A computer simulation of the test article is used to determine the accumulator resistance required to achieve a prescribed amount of structural stability. Based on the validated flow resistance model from the first task, the accumulator resistance will be modified. A series of shake tests will then be used to determine the amount of actual flow resistance obtained. This work has yet to be completed.

The final task is to design a pogo accumulator that will meet both first- and second-stage damping and frequency requirements. As described above, the pogo accumulator must provide the appropriate frequency separation between structural and feedline modes during second stage flight, as well as the appropriate feedline damping required to prevent new modes from becoming unstable. This is the primary function of the pogo accumulator. In support of first-stage flight, the accumulator must be tuned such that it actually resonates at the same frequency as the unstable acoustic mode in the SRM, and the resistive features must be designed to provide enough structural damping to actually stabilizing the vehicle during this stage of flight.

Accomplishments - The closed-loop flow testing is complete, and the results have proven to be very useful in the validation of the improved flow resistance model. The flow resistance model development is complete as well, although further research is continuing to better understand how the new model correlates with published work for other applications. Making this connection will allow the research to be applied in a more general fashion, extending beyond the original intended use. The modification of the experimental facility for the structural damping test is complete, and this test series is expected to take place in the coming weeks. This data is necessary for validating the same flow resistance model for the condition of zero through flow.

The last step is to apply what has been learned from the combined experimental/analytical program to develop a design methodology that allows for the use of the pogo suppressor as a structural damper during first-stage flight. This methodology includes establishing suppressor requirements for second-stage flight (traditional use of pogo suppressor) and establishing requirements for first-stage flight with the solid propellant engine. To support development of the final design concept, a finite element model of the Ares 1 launch vehicle has been generated and will be used to calculate the longitudinal modes of the structure. In addition, a feedline model of the second stage propulsion system is being developed to allow for sizing an appropriate pogo suppressor based on actual flight vehicle parameters. The improved flow-resistance model will then be included in the design methodology to identify a single resistance configuration that will work for both configurations without changing the hardware. The only variable that can reliably be actively controlled between configurations is the compliance of the suppressor, which establishes the frequency at which it provides the most damping benefit.

A closed-loop test rig developed by SwRI engineers is used to study the damping characteristics of various pogo suppressor design concepts. Pulsations in the closed-loop system are generated by a mechanically driven piston connected upstream of the test section.

Figure 1. A closed-loop test rig developed by SwRI engineers is used to study the damping characteristics of various pogo suppressor design concepts. Pulsations in the closed-loop system are generated by a mechanically driven piston connected upstream of the test section.

Figure 2. Closed-loop test results demonstrate trends similar to those predicted with the improved flow-resistance model. Most importantly, the improved flow-resistance model demonstrates that the resistance of the suppressor is dependent on the feedline flow rate, contrary to previous model predictions.

2010 Program Home