Hi Iain,
The light blue lines in the friction circle represent 0.25g increments.
When you are pulling over 1 g you'll see 4 of them. This is just like
in the strip charts.
The initial calibration that you do on a desk or table top is all that
has to be done in terms of actual calibrations. GEEZ performs a 1/2
second calibration at the start of each run (each time you hit Start or
AutoStart). That calibration sets the current position of the cube as
the zero reference point. Therefore, you should always try to hit Start
or AutoStart only when you are on very level ground. Failing this,
you'll need to make larger adjustments...it can be done, of course, but
starting on level ground makes it all easier.
Usage over 100% doesn't necessarily mean slip. Let's assume that the
Usage parameter that you are exceeding is your implied friction circle
(which it probably is). Look under Setup|Program Options|Data and see
if you are using Absolute Peaks or Sustained Peaks. The default is
sustained peaks. This means that your peak g's are computed based upon
a the highest value that you can maintain for a certain period of time,
such as 0.5 seconds. When you spike above that sustain level, you score
over 100%. So Usage > 100% means that you are currently pulling more
G's than you were able to maintain for the sustain duration during that
run.
A "Lo-Co" surface, such as gravel will make Geez see lower limits. But
GEEZ adapts automatically to that new limit, as GEEZ sets its limits on
a per run basis. Nothing is carried over from one run to the next.
Gravel roads can certainly produce a "smoothing" effect to your driving
as the breakaway and hook-up are certainly more gentle than on asphault.
Since we have "tuned" the software to asphault, that alone may account
for the very high usage rating. However, the path you took is also way
different from a typical autocross or road race track...I suspect that
when you use GEEZ in a more normal circumstance you'll find more normal
ratings.
Likewise, I wouldn't put too much stock in what you've read on that
first test run. Specificallly, the left vs right turn tendencies. See
if they show up on an autocross course first.
Sounds like you are off to a good start! Enjoy!
--Byron
I.Mannix wrote:
>
> Just got my Geez. Neat. Pretty fun stuff; my first question, what changes
> if you use it in a car on a low traction surface?
>
> I strapped my computer into my street car this morning, drove east(of
> Denver) to an officially sanctioned empty dirt road. Road is wide,
> recently graded, loose gravel over harpacked dirt. Basically, I wanted to
> get some data, play with it, see what this stuff does, try to get on top of
> the learning curve, so when I have real data to look at, I have a shred of
> a clue what I might be looking at.
>
> 1. Using the friction circle, there are two light blue circles - is the
> outermost light blue ring indicative of the edge of the circle(and when the
> line travels past that line, it indicates a loss of traction, right?).
>
> 2. I suspect I did not calibrate the cube correctly. This particular road
> is not flat in many places - I typically started atop a hill so I could see
> a fair distance in each direction(to ensure no traffic). Somewhere in the
> 200+ pages of help docs, there must be something about calibrating the box
> in the car; I found something after the fact about turning the car around
> while calibrating it(I did the initial calibration inside on a level
> countertop). My mounting point is pretty flat(in the car). Not sure that
> matters too much for what I set out to accomplish this morning.
>
> 3. A lot of time was spent fairly sideways - I chose a deserted gravel road
> so I could exceed the limits of traction without breaking the speed
> limit/risk hurting the car. Usage over 100% = slip, correct?
>
> 4. From looking at the graph - ummm, left Gs and right Gs? - the defaults -
> it appeared that I turn left better than right - my left turns(mimicking a
> slalom) were fairly round arcs, where right turns had a "step" to them -
> almost all of them. I suspect I'm sliding the car more(on gravel) turning
> right than left.
>
> 4a. Looking at the friction circle, I wound up with "apples" as described
> in one of Byron's articles in NAP. Thing is, they were lopsided apples.
> The right side of the apple(left turns, right?) were fairly solidly covered
> at the edge of the friction circle. On the left, the line tended to stop
> about 2/3 of the way to the outer blue circle. That'd indicate that I am
> making better use of the car turning left, right? Correct. Something like
> that.
>
> A recurring pattern showed up - diagonal lines from the upper left corner
> of the "apple," traveling down to the right - accelerating - into the left
> turn(right side of the apple). Huh. Seems to me that I am more
> comfortable turning left, applying the gas and driving through left arcs -
> which makes a fair bit of sense, when I think of other things I do that
> involve left - and - right turns; skiing, windsurfing, cycling.
>
> Smoothness rating was 99% on all but one map - the first map I made was
> 96%. First map I made was a baseline map - accelerate, brake, turn,
> accelerate - that friction circle was a cross, not an apple. After that, I
> tried to drive in a more "normal" fashion.
>
> Smoothness as compared to aggressiveness - aggressiveness was 70something%
> on all 7 maps. Seems to me that this is sorta reflective of the surface -
> it is easy to be smooth on dirt; turn in, car begins to slip, add throttle,
> drive out - car is an Audi 4000 quattro, seems right at home with a bit of
> slip angle on dirt roads. "Aggressiveness" looks at the amount of time
> spent on the gas and brake - on this particular gravel road, you could
> pretty much stand on the gas anywhere and break a couple of wheels loose.
> Peak Gs were just over .5, sustained in the .46-.48 range, depending. Do
> not recall peak left vs right, will have to look at that.
>
> Don't really know what my questions are, beyond the effects of using Geez!
> on slippery surfaces. I think smoothness gets affected; I'd like to think
> I'm a reasonably smooth driver, but not *that* smooth.
>
> Thoughts?
>
> Another question - linking each map to a driver profile - I have inserted
> the values for my DSP car, called my profile the default - does the program
> automatically take this into consideration when creating the map/speed/rpm?
> Probably somewhere in the extensive docs, which I've yet to finish reading.
>
> All in all, I'm really happy with the product - even a short time spent
> screwing around on dirt roads, I think I have found a flaw. Sorry this
> post is not as techy as those I have seen since I subscribed; I'll get
> things figured out sooner than later....
>
> Iain Mannix
|