Conversion of Sine to Random Testing

[Frank Trott]

I am trying to specify a simulation test for an electro-mechanical device fitted to a gas turbine engine. In order to simulate the engine environment I have developed a profile based upon customer defined data (temperature and sinusoidal vibration). With reference to Mil-HDBK-781 all testing that I can find relates to the use of a random vibration input but I am not able to state that a specific profile matches my customer's criteria without being able to provide a direct comparison between sine and random. I have found a reference to a conversion in Scientific Atlanta literature, but have been unable to get this formula to reproduce an acceptable profile. Is there a recognized method of conversion between sine and random?

[Dennis Wilkins]

Random vibration is used for stressing product like it will be stressed via shipping (air, truck, etc.) and tends to excite many vibrational modes. There is no direct relation between sine and random vibration testing, and results of testing both ways will yield different effects dependent on the physical aspects of the product.

See Wayne Tustin's site at http://www.equipment-reliability.com/ This used to be the Tustin Institute, but is renamed the Equipment Reliability Institute since some other institution, also in California, already had the name. They are the premier source on vibration effects on equipment.

Dennis Wilkins - Hewlett-Packard / ReliaSoft Instructor