|
ralph1
New Member
12 Posts |
 Posted - 09 Mar 2010 : 12:21:46
|
Hello
Have the smoothing concerns as expressed by threads circa 2005 been resolved?
I am curious because if the hp/torque figures reported from the on road dyno application are still noticeably spikey, then I have in mind an approach that should "in theory" reveal very smooth curves. I don't want to re invent the wheel if "smoothing" is no longer an issue but would be willing to try and put theory into practice if the problem is still real and present?
If someone could supply the native velocity data I could give a whirl in excel and have some sort of conclusion reasonbly promptly.
Essentially the problem is one of generating an accurate and smooth acceleration vs rpm curve and it is the quality or otherwise of this curve that manifests as smooth or spikey power/torque plots. If the native velocity data which, from our intuitive understanding of how subarus progress along the road, is a progressive affair, can be curve fitted using the the polynomial trendline function in excel, then the differentiation of this velocity function wrt time to get the acceleration curve is an easy and very accurate mathmatical process. (I've used a similar sort of approach to generate valve train acccleration cuvres)
What do you reckon, worth a try guys? but like I say above I don't want to re invent the wheel if you've already sorted the problem
Cheers
Ralph |
|
Data Smoothing
