Audio datalogging (was Re: Shock Sensor Question)

[Phil Hunter]

"Jon Snoddy" wrote:

> Not sure if this will work, but I think is should: 
> Plug the data recorder analog output signal into a
> laptop's analog audio input and record it to an
> uncompressed wave file.  View it with an audio
> editing program.  Audio editing programs provide a
> timeline view that will have very accurate time
> marks in seconds, fractions of seconds and even
> samples (44k per second). Another cool trick is that
> the input is stereo, so you can record and display
> two channels of data at once.  Using one of the
> many multi-track recorders, you can view successive
> recordings on top of one another on the same time
> line.  Record audio with a microphone on one channel
> so you can hear bumps and engine speed (easy to
> arrive at RPM from pitch) in time synch to the
> shock data. Hmmm.  I'm beginning to like this idea.
> 
> You will need to install a "pad" to reduce the
> signal from 0-4v down to about 0-1 volt (what is
> that, 6db?) to keep from overloading the input of
> the laptop.  I imagine you can buy a 6db pad at
> Radio Shack.  Many computers have low pass filters
> at the front end that make it impossible to record
> DC which means that any movement at a rate slower
> than 50 to 20 Hz would come out as a straight line.
> There are many audio A/D cards available for desktop
> units or with a firewire connector on a laptop,
> some of which have defeatable low pass filters.
> 
> You will want to make sure the voltage is limited to
> what ever your line input can tolerate.
> 
> - -jon

Dang, someone else had the same idea, or maybe "great
minds think alike", basically using audio recordings
as an oscilloscope or datalogger. My thunk was for
detecting engine knock by capacitively coupling the
cam sensor's digital signal together w/ the output of
the ESC into the microphone input, didn't think about
using the stereo Line In, but I like it. You'd see
only spikes that represent the edges of the signal,
but since it's not a symmetrical square wave, one s/b
able to figure out where each cylinder would be along
the time line, and from there figure out which
cylinder(s) are knocking from the position of the
ESC's spikes in the middle of the pattern. With a
little care, you s/b able to determine if the ping is
at the wrong time or too long, indicating "false
knock" caused by something else. Doesn't have to be
the output from the ESC either, could be directly
from the knock sensor or even a microphone. Knowing
which cylinders are knocking should help narrow down
the possible causes.

As for DC being blocked, at the cost of additional
complexity you might consider modulating the signal
on a high audio frequency carrier, perhaps using
something like an NE602 mixer IC.

Thanks for the post, Jon, you had some ideas I hadn't
thought of. Might just have to get off my dead butt
and try it out.

regards,
phil
(digest)

__________________________________________________
Do You Yahoo!?
Check out Yahoo! Shopping and Yahoo! Auctions for all of
your unique holiday gifts! Buy at http://shopping.yahoo.com
or bid at http://auctions.yahoo.com

----- End of forwarded message from owner-diy_efi at diy-efi.org -----
----------------------------------------------------------------------------
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