knock frequency characteristics
Carter Shore
clshore at yahoo.com
Thu Aug 30 12:55:18 GMT 2001
--0-1217035557-999176095=:84832
Content-Type: text/plain; charset=us-ascii
Mark Agnello wrote:
>These are just educated guesses, but I suspect they
>apply. Detonation is closely coupled to the pressure
>wave in the combustion chamber. The pressure wave is
>an acoustic event. As such, it should be influenced
>by the geometry of the combustion chamber at the
>instant of the event. The dominant mode's wavelength
>will probably be proportional to the cylinder bore.
Taylor mentions this in Vol 2, but does not go into any details, at least not in my edition.
Many (but not all) of the pressure oscillographs clearly show a decaying ringing envelope following the initial pressure spike from the detonation event. It is suggested that this may be an artifact of the sensor itself, but not explored.
>WRT the cylinder head and/or block structure
>influencing the dominant frequencies, I agree there
>will be some interaction. Unlike the 10 Hz bell
>analogy, I don't believe there will be a dominant
>vibrational mode of the engine structure. These
>structures are too complex to have a primary mode and
>I'll bet auto manufacturers spend a lot of time
>designing engine structures that don't have
>significant vibrational modes. I know aircraft engine
>manufacturers do this. If they didn't, the life of
>the engine mounted fuel pumps, actuators, sensors, and
>accessories would be severely shortened.
>What do y'all think? Remember, I'm just a lonely old
>EE that likes to read SF.;)
The standard shadetree method for testing a KS is to use a broomstick or hammer handle to whack the block. Although this will certainly create resonance due to cylinder dimensions, my gut feeling is that the structure has the major influence in this case.
In a variation of the old investigator's saw to 'follow the money', we can 'follow the energy'. When a knock event occurs, how much energy is created, and where does it go? The oscillographs mentioned above show a fairly rapid envelope decay. Of course the cylinder volume is changing, the mixture chemistry is changing, the temperature is changing, etc. but I suspect that most of the acousic energy is coupling into the surrounding structure.
I for one would love to attach a good spectrum analyzer to an engine, and then excite the assembly, sweeping the frequency to see what the resonant modes really are.
Next step would be to capture some knock events, under different conditions, and compare with the results from above.
The current DSP chips can easily detect complex acoustic events, but you have to first know what you are looking for. Prior posts about using a copper pipe to listen for knock have been intriguing, validating that a very sophisticated instrument (a trained human ear) is capable of detecting knock acoustically.
>Autospeed published an article based on an Aussie grad
>student's research into engine knock earlier this
>year. Does anyone have a copy and does it shed light
>on this issue?
I'd love to see that too.
Thanks all
Carter Shore
---------------------------------
Do You Yahoo!?
Get email alerts & NEW webcam video instant messaging with Yahoo! Messenger.
--0-1217035557-999176095=:84832
Content-Type: text/html; charset=us-ascii
<P>Mark Agnello wrote:</P>
<P><BR>>These are just educated guesses, but I suspect they<BR>>apply. Detonation is closely coupled to the pressure<BR>>wave in the combustion chamber. The pressure wave is<BR>>an acoustic event. As such, it should be influenced<BR>>by the geometry of the combustion chamber at the<BR>>instant of the event. The dominant mode's wavelength<BR>>will probably be proportional to the cylinder bore. <BR></P>
<P>Taylor mentions this in Vol 2, but does not go into any details, at least not in my edition.</P>
<P>Many (but not all) of the pressure oscillographs clearly show a decaying ringing envelope following the initial pressure spike from the detonation event. It is suggested that this may be an artifact of the sensor itself, but not explored.</P>
<P> <BR>>WRT the cylinder head and/or block structure<BR>>influencing the dominant frequencies, I agree there<BR>>will be some interaction. Unlike the 10 Hz bell<BR>>analogy, I don't believe there will be a dominant<BR>>vibrational mode of the engine structure. These<BR>>structures are too complex to have a primary mode and<BR>>I'll bet auto manufacturers spend a lot of time<BR>>designing engine structures that don't have<BR>>significant vibrational modes. I know aircraft engine<BR>>manufacturers do this. If they didn't, the life of<BR>>the engine mounted fuel pumps, actuators, sensors, and<BR>>accessories would be severely shortened.<BR><BR>>What do y'all think? Remember, I'm just a lonely old<BR>>EE that likes to read SF.;) <BR></P>
<P>The standard shadetree method for testing a KS is to use a broomstick or hammer handle to whack the block. Although this will certainly create resonance due to cylinder dimensions, my gut feeling is that the structure has the major influence in this case. </P>
<P>In a variation of the old investigator's saw to 'follow the money', we can 'follow the energy'. When a knock event occurs, how much energy is created, and where does it go? The oscillographs mentioned above show a fairly rapid envelope decay. Of course the cylinder volume is changing, the mixture chemistry is changing, the temperature is changing, etc. but I suspect that most of the acousic energy is coupling into the surrounding structure.</P>
<P>I for one would love to attach a good spectrum analyzer to an engine, and then excite the assembly, sweeping the frequency to see what the resonant modes really are.</P>
<P>Next step would be to capture some knock events, under different conditions, and compare with the results from above.</P>
<P>The current DSP chips can easily detect complex acoustic events, but you have to first know what you are looking for. Prior posts about using a copper pipe to listen for knock have been intriguing, validating that a very sophisticated instrument (a trained human ear) is capable of detecting knock acoustically.</P>
<P><BR>>Autospeed published an article based on an Aussie grad<BR>>student's research into engine knock earlier this<BR>>year. Does anyone have a copy and does it shed light<BR>>on this issue?</P>
<P> </P>
<P>I'd love to see that too.</P>
<P>Thanks all<BR></P>
<P>Carter Shore<BR></P><p><br><hr size=1><b>Do You Yahoo!?</b><br>
Get email alerts & NEW webcam video instant messaging with <a
href="http://rd.yahoo.com/mail_us/tag/?http://im.yahoo.com/">Yahoo! Messenger</a>.
--0-1217035557-999176095=:84832--
----------------------------------------------------------------------------
To unsubscribe from diy_efi, send "unsubscribe diy_efi" (without the quotes)
in the body of a message (not the subject) to majordomo at lists.diy-efi.org
More information about the Diy_efi
mailing list