01-27-2009 05:10 AM
01-27-2009 05:37 AM
There are two sorts of filters:
Normal filters working on an aquired signal as an array and PointByPoint filters working on one value. The first one you can use for offline analysis and the second one for in time filtering.
Which typ of filter is best depends on the signal how it should look without any distortion. Maybe a lowpass or bandpass filter is a good starting point. To eleminate any noise resulting from the revolution of the motor a bandstop filter is a good choise.
01-28-2009 07:40 AM
01-28-2009 08:54 AM - edited 01-28-2009 08:56 AM
60Hz noise isn't typically a problem in car.
This is true but you have a ripple on the 12 Volt Power line resulting from the electric generator which can or can not be visible in the signal.
01-28-2009 01:06 PM
In addition to generator (alternator) noise, add: fuel injectors, ABS solenoids, ignition spark, powertrain solenoids. Can be a nasty environment.
-AK2DM
01-28-2009 01:45 PM
Just as a side note, you might want to come up with a better post title next time, such as "Signal filtering" or similar.
Since this is a LabVIEW forum, all threads here are related to LabVIEW, so your title is 100% redundant and has zero information content. Imagine the mess if every single post were titled "LabVIEW". 😮
01-28-2009 11:27 PM
An often forgotten point is that the filter approach varies depending on whether the energy you are trying to remove is in-band or out-of-band. In-band energy (meaning any components with a frequency up to 1/2 the sample rate) can be removed by filtering in software. Out-of-band energy has to be done in hardware because the purpose if this filtering (sometimes called anti-alias filtering) is to remove any frequency components greater than 1/2 the sample rate so they don't corrupt the output of the A/D converter.
Having said all that: Spark plugs are amazing things - harmonically speaking... Look at them as small very-very-wideband transmitters. Their fundamental frequency is obviously a function of engine speed, but the harmonics go much much higher.
Mike...