Path: utzoo!news-server.csri.toronto.edu!rpi!zaphod.mps.ohio-state.edu!magnus.acs.ohio-state.edu!tut.cis.ohio-state.edu!pacific.mps.ohio-state.edu!linac,att!cbnews!cbnews!military From: emory!Dixie.Com!jgd@gatech.edu (John G. DeArmond) Newsgroups: sci.military Subject: Re: Radar counting turbine blades (was Re: Airborne Radar) Message-ID: <1991Mar18.131907.13138@cbnews.att.com> Date: 18 Mar 91 13:19:07 GMT References: <1991Mar14.041632.29149@cbnews.att.com> Sender: military@cbnews.att.com (william.b.thacker) Organization: Dixie Communications Services Lines: 66 Approved: military@att.att.com From: emory!Dixie.Com!jgd@gatech.edu (John G. DeArmond) TOHALL@MARS.LERC.NASA.GOV writes: > What you are actually measuring is the blade passing frequency. >This will be equal to the fan shaft speed times the number of blades. In >the case of fighter aircraft engines, the number of blades differs from >one to another, and the fan shaft speed will vary with throttle setting. >You have a real mess if you try to use blade passing freq to ID a threat: > >Was that a 10 bladed fan running at 10KRPM, or a 20 bladed fan running at >5K RPM (for example)? Additionally, the fan will produce harmonics at >multiples of the blade passing frequency: A 16 bladed fan running at 12000 >RPM produces signals at 3200 Hz, 6400 Hz, 9600 Hz ....... These signals >could be misinterpreted as well! The 6400 Hz signal which corresponds to the >2nd harmonic of the 16 bladed fan at 12000 RPM could be confused with the >blade passing fundamental tone of a 32-bladed fan running at 6000 RPM, etc. This is true but the signal analysis is not necessarily as complicated as you might think. Being the ever curious type, I have conducted a couple of experiments since this topic came up. I had the fortunate coincidence to attend a Winston Cup Dragrace points race at the Atlanta Dragway last weekend. My reason for going was that the entertainment was a contingent of jet cars and a jet-powered Kentworth semi truck (with THREE Westinghouse J-48s with afterburners!) Knowing that the jets were going to be there and in light of this discussion, I took one of my small 10 ghz Gunnplexers and added an earphone so that I could listen to the doppler. (I had meant to take my Kustom Signals HR-7 speedgun but I got lazy and did not make a battery pack in time :-) The short story is that despite the wire mesh birdcage debris trap mounted on the engines, I could easily hear the blade chop as the engines were spooled up on the startpacks. The frequency quickly exceeded the bandwidth of the earphone but I have no doubt that the components remained as the engine reached operating speed. The second experiment involves examining the signal on a scope from a gunnplexer aimed at several bladed devices in my lab including a turbocharger fan driven by compressed air. The turbocharger should be representative because the speed and balance requirements are of the same order of magnitude as jet engines. On the scope, the blade chop could be seen as would be expected. More interesting, the shaft irregularities, even with the turbo, was clearly obvious as periodic amplitude variation of the chop signals. Armed with this knowledge, it is easy to postulate some fairly simple signal analysis that could separate harmonics from the fundimental AND could determine the engine RPM which allows calculating the blades on a fan. A simple FFT can separate the harmonics. A bit more complexity, probably some form of deconvolution, would extract the RPM information. Given that the military and its contractors have resources that somewhat exceed mine (:-), Fairly reliable IF (Identify Friend) from turbine signatures should be possible. If the military assumes that any craft not identified as friendly is a foe, the problem is simplified significantly. -- John De Armond, WD4OQC | "Purveyors of speed to the Trade" (tm) Rapid Deployment System, Inc. | Home of the Nidgets (tm) Marietta, Ga | {emory,uunet}!rsiatl!jgd |"Politically InCorrect.. And damn proud of it