On 11 Maj, 22:04, eric.jacob...@ieee.org (Eric Jacobsen) wrote:> On Sat, 11 May 2013 19:45:34 +0000 (UTC), glen herrmannsfeldt > > > > > > > > > > <g...@ugcs.caltech.edu> wrote: > >Eric Jacobsen <eric.jacob...@ieee.org> wrote: > > >(snip, I wrote) > >>>I have seen emission inspection systems that want to measure the engine > >>>RPM that plug into the cigarette lighter (presumably some harmonics go > >>>into the engine electrical system) and also a vibration sensor that goes > >>>on the hood. (So, mechanically coupled, not air coupled, vibrations.) > > >> Tapping the electrical system is the most reliable "non-contact" > >> method I know of for estimating rpm reasonably accurately, but you > >> still need to know the number of cylinders or that'll get fooled > >> pretty easily. � The characteristics of the electrical noise from the > >> ignition are quite different than the acoustic response, though. > > >The emission inspection system knows the VIN, which I believe can be > >traced back to the number of cylinders. > > >The engine that I do this with, and sometimes it took the system a > >little while to figure out, is a Chrysler 3.3l V6 that uses three > >ignition coils and no distributor. No idea what that does on the > >electrical system different than a traditional one coil system. > > >The spark timing is generated directly by the computer, from rotational > >position sensors on the crankshaft. > > Most modern cars are that way, many with an independent coil directly > on each plug. � There's still enough electrical noise from the spark > to get picked up elsewhere in the system for rpm estimation, partly > because that allows a more energetic spark with less loss to the plug. > > If you have a diesel, though, you're screwed. �;) � �This is a common > problem when running a diesel on a chassis dynamometer, as there's no > easy way to instrument the rpm to get both hp and torque measurements. > > Eric Jacobsen > Anchor Hill Communicationshttp://www.anchorhill.comI thought those plug into the cigarette lighter things were really looking at the noise from the alternator so they don't care how many cylinders or what fuel just how many poles on the alternator -Lasse
Clipped before FFT
Started by ●May 10, 2013
Reply by ●May 11, 20132013-05-11
Reply by ●May 12, 20132013-05-12
On Sat, 11 May 2013 18:07:49 -0700 (PDT), "langwadt@fonz.dk" <langwadt@fonz.dk> wrote:>On 11 Maj, 22:04, eric.jacob...@ieee.org (Eric Jacobsen) wrote: >> On Sat, 11 May 2013 19:45:34 +0000 (UTC), glen herrmannsfeldt >> >> >> >> >> >> >> >> >> >> <g...@ugcs.caltech.edu> wrote: >> >Eric Jacobsen <eric.jacob...@ieee.org> wrote: >> >> >(snip, I wrote) >> >>>I have seen emission inspection systems that want to measure the engin= >e >> >>>RPM that plug into the cigarette lighter (presumably some harmonics go >> >>>into the engine electrical system) and also a vibration sensor that go= >es >> >>>on the hood. (So, mechanically coupled, not air coupled, vibrations.) >> >> >> Tapping the electrical system is the most reliable "non-contact" >> >> method I know of for estimating rpm reasonably accurately, but you >> >> still need to know the number of cylinders or that'll get fooled >> >> pretty easily. =A0 The characteristics of the electrical noise from th= >e >> >> ignition are quite different than the acoustic response, though. >> >> >The emission inspection system knows the VIN, which I believe can be >> >traced back to the number of cylinders. >> >> >The engine that I do this with, and sometimes it took the system a >> >little while to figure out, is a Chrysler 3.3l V6 that uses three >> >ignition coils and no distributor. No idea what that does on the >> >electrical system different than a traditional one coil system. >> >> >The spark timing is generated directly by the computer, from rotational >> >position sensors on the crankshaft. >> >> Most modern cars are that way, many with an independent coil directly >> on each plug. =A0 There's still enough electrical noise from the spark >> to get picked up elsewhere in the system for rpm estimation, partly >> because that allows a more energetic spark with less loss to the plug. >> >> If you have a diesel, though, you're screwed. =A0;) =A0 =A0This is a comm= >on >> problem when running a diesel on a chassis dynamometer, as there's no >> easy way to instrument the rpm to get both hp and torque measurements. >> >> Eric Jacobsen >> Anchor Hill Communicationshttp://www.anchorhill.com > >I thought those plug into the cigarette lighter things >were really looking at the noise from the alternator so >they don't care how many cylinders or what fuel >just how many poles on the alternatorThat could be the case on some of them. There's a device called a G-Tech that just asks you to calibrate it by holding the rpm at a couple of known values so that it can sort out what's going on. There are some that ask for the # of cylinders, but having a calibration routine makes it all a don't care as long as there's sufficient periodic noise correlated to rpm, regardless of where it comes from. I don't know of any that ask for the number of alternator poles, probably because nobody would know that, anyway, and you'd also need to know the ratio of the pulley sizes on the accessory belt. That's not very practical, but the alternator noise could certainly be used if it was dominant and the user calibrated the device. That'd even work on a diesel. Eric Jacobsen Anchor Hill Communications http://www.anchorhill.com
Reply by ●May 13, 20132013-05-13
Robert Scott <no-...@notreal.invalid> wrote:> What I want to know is if this method is part of some established and studied theory.Just thought I'd add: acoustic analysis has been used in Formula 1 racing: http://www.grandprix.com/ft/ft00345.html And the following is a fairly detailed analysis: http://www.f1technical.net/forum/viewtopic.php?f=4&t=2218&sid=bd0070c578d4d6e9cfe084df1e7936cb There are also a number of technical papers published on the subject, such as: http://www.vjsuper.org/journals/doc/ASALIB-home/archives/L1_106.pdf Kevin McGee
Reply by ●May 13, 20132013-05-13
On Fri, 10 May 2013 18:49:06 -0500, Vladimir Vassilevsky <nospam@nowhere.com> wrote:>However, those methods are largely out of fashion; as normalized >autocorrelation generally makes for better results.Well, I did some experimentation with autocorrelation. Correct me if I am wrong, but the autocorrelation array can be calculated as: F(f) = FFT( F(t) ) S(f) = F(f) x F*(f) A(t) = iFFT( S(f) ) I started with a real-valued 8192-point time series. I took the FFT to get 8192 complex values. Then I formed the real-valued S-array by taking the magnitude squared of each element of each element of the complex FFT array. (That is the product of each element of the FFT with its complex conjugate.) Then I took the FFT of the S array. The only difference between the FFT and the iFFT is that the sign of the imaginary part is flipped and there is a overall scaling factor change. But since I am dealing the a real-valued input, and the output is being auto-scaled for examination and analysis, those two differences are irrelevant. Anyway, after doing this and graphing the magnitude of the resulting autocorrelation array, the graph did not look much better than the initial time-series array. It look a little more sinusoidal than the initial input, but it does fade out completely at various places, even with a tuning fork supplying the audio. I assume that to use this array for pitch detection I would have to start from the beginning of the array and count peaks until I get far enough away from the beginning of the array so that the delay resolution is sufficient for my needs. Is that how the autocorrelation array is supposed to be used for pitch measurement? Robert Scott Hopkins, MN
Reply by ●May 13, 20132013-05-13
On Sat, 11 May 2013 15:23:37 GMT, eric.jacobsen@ieee.org (Eric Jacobsen) wrote:>The propeller noise dominates with an airplane and so I'd expect that >to be easier to estimate than an automotive engine. Have you used it >with airplanes with gear reduction (e.g., Cessna 421)?I don't think that is the case for small single-engine planes with minimal muffling. The exhaust note dominates. Of couse the problem with the Cessna 421 is that it is a twin, so unless the engines have been trimmed to the same speed, the mixutre of sounds will be hard to analyze. Robert Scott Hopkins, MN
Reply by ●May 13, 20132013-05-13
On 5/13/2013 8:17 AM, Robert Scott wrote: What you are trying to do is essentually pitch (or beat) detection. There is numerous amount of pitch detection stuff over the internet. Don't you bother to look? Or, if you seek professional development, contact me, or Robert, or Dmitry Teres, or some other expert privately. Vladimir Vassilevsky DSP and Mixed Signal Designs www.abvolt.com
Reply by ●May 13, 20132013-05-13
On Mon, 13 May 2013 13:26:44 GMT, no-one@notreal.invalid (Robert Scott) wrote:>On Sat, 11 May 2013 15:23:37 GMT, eric.jacobsen@ieee.org (Eric >Jacobsen) wrote: > > >>The propeller noise dominates with an airplane and so I'd expect that >>to be easier to estimate than an automotive engine. Have you used it >>with airplanes with gear reduction (e.g., Cessna 421)? > >I don't think that is the case for small single-engine planes with >minimal muffling. The exhaust note dominates. Of couse the problem >with the Cessna 421 is that it is a twin, so unless the engines have >been trimmed to the same speed, the mixutre of sounds will be hard to >analyze. > >Robert Scott >Hopkins, MNThe 421 was just an example of a fairly common airplane with gear reduction. There are many single-engine airplanes that also have gear reduction (e.g., most things with a Rotax), and plenty of airplanes with more muffling than "minimal" (especially these days). So my question was how you deal with such things, where there is an expected disparity between the acoustic signature and the actual engine rpm. Eric Jacobsen Anchor Hill Communications http://www.anchorhill.com
Reply by ●May 14, 20132013-05-14
On Mon, 13 May 2013 17:12:48 GMT, eric.jacobsen@ieee.org (Eric Jacobsen) wrote:>>>The propeller noise dominates with an airplane and so I'd expect that >>>to be easier to estimate than an automotive engine. Have you used it >>>with airplanes with gear reduction (e.g., Cessna 421)?If the propellor noise does dominate then the user would have to configure our app for the number of blades in the prop. The gear ratio would not figure into it because the instrument panel tach reads prop RPM anyway (at least it does in the Diamond Katana which uses a geared-down Rotax). But if the engine note dominates then the gear ratio would be a factor in the app setup because we still want the RPM to read the prop to be comparable to the instrument panel tach. Robert Scott Hopkins, MN
Reply by ●May 14, 20132013-05-14
On 14.5.13 5:51 , Robert Scott wrote:> On Mon, 13 May 2013 17:12:48 GMT, eric.jacobsen@ieee.org (Eric > Jacobsen) wrote: > >>>> The propeller noise dominates with an airplane and so I'd expect that >>>> to be easier to estimate than an automotive engine. Have you used it >>>> with airplanes with gear reduction (e.g., Cessna 421)? > > If the propellor noise does dominate then the user would have to > configure our app for the number of blades in the prop. The gear > ratio would not figure into it because the instrument panel tach reads > prop RPM anyway (at least it does in the Diamond Katana which uses arr > geared-down Rotax). But if the engine note dominates then the gear > ratio would be a factor in the app setup because we still want the RPM > to read the prop to be comparable to the instrument panel tach. > > Robert Scott > Hopkins, MN >The main reason for tach is to monitor the propeller. The engine speed comes as a side effect in direct-drive engines. -- Tauno Voipio
