Can ElanScan distinguish between traction and clutch problem

[ksgrimsr]

Hi guys,

I finally found a use for my NetBook and took it for a ride today with ElanScan to record a couple acceleration passes.

On my first pull something clearly slipped during the pull, but I cann't tell if I'm loosing traction or if my clutch is slipping. Can ElanScan distinguish the difference by correlating the engine RPM's with the vehicle speed? Is the vehicle speed recorded from the rotation rate of the wheels or is it computed from the RPM's and selected gear?

A picture of the run is below and I attached my ElanScan file. If there are experts out there that know enough details about what is actually sensed vs what is computed, I'd appreciate some interpretaiton of my results. It's the first pull that had the noticeable slipping and in the graph you can see a notch in vehicle speed that I thought might indicate a traction issue. The solid line on the graph is the vehicle speed while the dotted line is the engine speed.

Knut

PullWithSlipping.jpg

[GeoffSmith]

Engine speed comes from the rev counter signal. Road speed comes from the vehicle speed sensor (VSS). Selected gear is calculated, so it's probably not possible to differentiate between wheel slip and clutch slip.

[ksgrimsr]

Engine speed comes from the rev counter signal. Road speed comes from the vehicle speed sensor (VSS). Selected gear is calculated, so it's probably not possible to differentiate between wheel slip and clutch slip.

If both signals are measured (not computed) and they are measured on opposite sides of the clutch (RPM at engine and speed at wheels), then I should be able to see what I'm looking for in the data depending on where the VSS actually sits. Plotting a scatterplot of speed vs RPM should result in the data falling along a couple straight lines whose slope corresponds to the gearing ratios (provided the RPM and speed are coupled due to the clutch not slipping).

For the VSS, I presume it is measured from the rotation of the front wheels (not the back)? Is it measured off one of the wheels (in which case depending on which wheel might break traction I might or might not see it) or is it measured before the differential?

In taking a closer look at the data I had another question about ElanScan behavior. In the posted ElanScan file, if you look at datapoints 105 to 107 this is clearly the shift from first to second (throttle is closed during the shift and the RPM's drop as I shift to the 2nd gear). However, the vehicle speed during this time is shown as increasing substantially. A little later I'm back on the throttle and the vehicle speed is shown as not increasing. The data looks like the vehicle speed measurements are delayed a bit relative to the other data measured. I can eyeball the delay and shift the data to adjust for it, but I thought I'd first ask whether there is indeed a known delay in the measured vehicle speed.

I took a quick look at an initial Excel scatterplot and I'll post a picture after I see if I can clean up the offset of the vehicle speed data (the graphing tool in ElanScan makes the X-Y scatterplot look strange, but in Excel it looks as expected but with the speed delay muddying the picture).

Knut

[Specky]

Out of curiosty have you done the usaul forth gear test on the clutch?

If it helps....

A slipping clutch is just annoying and you can smell it....losing traction at any speed makes your 4rse twitch and you can smell it...

TTFN

[GeoffSmith]

VSS is taken from the speedo cable. When you've plotted road speed vs engine speed in ES, right mouse on the graph and select the curves option - remove lines and replace with points and all will become a lot clearer - I agree, your scatter plot is crap! As it is happening at the top end, I suspect it is clutch slip.

The gear algorithm in ES is a bit weak and with a bit of tweaking, I think the number of false zeros could be substantially reduced - check out my investigations on a non-slipping run a few years ago:

ElanScanGearsReview.gif