Space launch vehicles can exhibit self-excited longitudinal oscillations, also known as “Pogo” — so named because the phenomenon vibrates the rocket up and down in a manner similar to bouncing on a pogo stick. The vibrations severely impair the astronauts’ ability to pilot or respond to emergencies and can cause structural failure of the vehicle. NASA first became aware of the disastrous consequences from Pogo during the Gemini-Titan program. The issue continued to plague the agency through the Saturn V Moon launch missions.

 

Pogo is caused when random oscillations in the rocket cause thrust oscillations by interaction with the propellant feed system — in particular, the pumps. The vehicle oscillations cause pressure oscillations in the propellant system, leading to an oscillating flow rate in the pumps that results in thrust oscillations from the engine that further excite the structure. A key factor in the occurrence of Pogo is how the pumps respond to pressure oscillations in the feed lines.

 

To understand this phenomenon better, a unique cavitating pump dynamics test facility was built at Concepts NREC’s White River Junction, VT facility in cooperation with engineers at NASA Marshall Space Flight Center (Huntsville, AL) back in 2006. Its purpose was to determine pump dynamic transfer functions for use in vehicle system stability modeling and analysis. At the time it was built, the facility at Concepts NREC was the only facility of its kind in the United States, and only one of two in the world.

 

Beginning with the space shuttle development, NASA has mandated that Pogo cannot be present on any launch system. Extensive research and development efforts were put into the space shuttle engine design, which included pump dynamics testing at Caltech in the 1970’s. The results were very successful, and the shuttle flew for 30 years without any Pogo events.

When developing new launch assets, it is important to ensure that the new vehicles and engines will also be Pogo-free.

 

Determining Pogo

 

The transfer function is a four-element matrix that relates the changes in the dynamic mass flow and pressure across the pump to the dynamic mass flow and pressure at the pump inlet. Each of the terms is a complex number that includes both amplitude and phase. The transfer functions are determined by imposing at least two independent sets of small amplitude mass flow and pressure perturbations at a given frequency on the mean inducer flow.

 

The results of the perturbations are then measured upstream and downstream of the pump to determine the four complex numbers that make up the transfer function matrix. The facility consisted of a magnetic bearing-supported pump flow loop with an inlet flow pulser, inlet flow conditioner, inlet bandwidth-enhanced electromagnetic flow (EMF) meter, test piece, discharge collector, exit flow conditioner, exit bandwidth-enhanced EMF meter, exit flow pulser, flow conditioner, loop flow meter, and a throttle valve.

 

The flow pulsers at the inlet and exit of the test piece provided the perturbations on top of the mean flow. The test rig loop was stiff and structurally robust to minimize structural vibration coupling with the flow pulsations. High-frequency pressure and flow measurements at the inlet and exit were the key measured parameters. One particular concern was accurately measuring the small flow perturbations with the desired frequency response. Typical flow meters are low frequency or averaging meters that cannot capture the frequencies of the pulsing flow in the loop. A special bandwidth-enhanced EMF meter was developed by Concepts NREC for this application and allowed accurate measurements of the pulsing flow.

 

The new meter started from a regular EMF meter with enhancements to drive it to a higher excitation frequency to cover the intended measurement bandwidth. In addition, special signal processing extracted the time-varying flow component from the signal to capture the flow rate changes with the pulsing. The new meter was a great success and has been subsequently purchased by NASA for their own transfer function flow loop.  This flow meter is available to others seeking high-frequency flow measurements.

 

A series of tests was conducted on the Space Shuttle Main Engine (SSME) low-pressure oxidizer turbopump (LPOTP) inducer to replicate what was done in the 1970’s at Caltech.  The results were similar to the early work at Caltech and validated the Concepts NREC dynamic transfer function test platform.