I've been struggling a lot this season. The move from Stock to Street Prepared
has brought with it a steep learning curve, competing in Pro2 (and ESP in
Detroit) means the level of competition is very high, and driving a DSM (a
relative newcomer to autocross) means there is no Daddio Setup you can just
clone and use. You get to do the lion's share of your setup research yourself.
Couple all this to the unforgiving nature of the G-Force tire when pushed over
the limit, and you get a very frustrated driver looking back over the results of
the season to date wondering just why it is he sucks so badly.
However, self-pity may have a certain entertainment factor, but it doesn't win
races nor does it fix problems. What I needed to do was make a positive step
towards fixing my season.
The first step was buying Kuhmos. The G-Forces finally corded at the DSM
Shootout (after about 80 runs) and I decided that I needed a tire that was a
little more forgiving. A trip back through my logbook revealed a spin or an
off-course incident in almost every single event I ran at - a great big glowing
neon sign. And hey, the K tires are a little cheaper too. So Kuhmos it is.
The second was getting some real, hard data on exactly what the car was really
doing - and that meant a datalogger of some sort.
I had actually tried to log some data the previous season. My car is an OBDII
car, and that means it has a data port and the facilities to log on that port. I
bought some software (OBDTool) with the intent of logging speed and throttle
position (OBDTool can log 2 channels at 3 samples/second from whatever sensors
are connected to the car's ECU) The attempt was a complete bust. The laptop must
be connected to the car and running to log data, and the logistics of mounting
the laptop in the car, and of seeing the LCD screen in broad daylight, and of
successfully clicking the appropriate on-screen buttons to get to the logging
screen and start logging, all proved to be near-impossible in a competition
environment. For test sessions it was OK, but for real races, it sucked.
So it was with great interest that I examined GEEZ! GEEZ! was designed by an
autocrosser, it pays contingency money, and it was fairly cheap. Should have
been a done deal.
...except that I had a chance to play with it in the field at the Virginia Pro,
and it suffered from the same sort of laptop-related problems that OBDTool did.
Logging was a little better with the AutoStart feature, but you still needed to
navigate program screens to get to the logging section, the laptop had to be in
the car and on to log, and battery life was an issue - not to mention laptop
fragility issues.
What's more, the software was... clunky. Not _bad_, or "wrong" but... it was
harder to do the things I wanted to do than it should have been. I couldn't find
a way to overlay runs for comparision purposes. The obvious keyboard shortcuts
for data navigation (the arrow keys) weren't mapped. All little things to be
sure, but annoying little things. My first impression was "Wow, look at all the
neat things this can do", but my second impression was "Wow, it's way tougher to
get the information I want out of this than it should be"
GEEZ! had a new Palm Pilot version for data logging that seemed to be a much
better solution, eliminating the issues with having to have a laptop in the car
- but I didn't own a Palm, which meant tagging somewheres between $150-$350 to
the price of entry (which means $300 to $600 in real money).
So I went looking for alternatives.
For a while, it looked like the closest thing to what I wanted was the Corsa
system from Corsa Instrements - at an entry price of $1500 (eek!) But then
Edelbrock released the QwikData system, a near-clone of the Corsa system for
roughly half the price - and, incidently, for damn near the price of GEEZ! plus
a new Palm Pilot.
In other words, for the same money as a GEEZ! system, I could get the same data,
in a hardened box, plus 14 more data channels. It was a no-brainer; I called up
Sam at Strano's (That's 1-888-6-STRANO) and ordered the Basic system.
Installing the system was straightforward. Edelbrock supplies a pre-made wiring
harness that is intended for use in dragsters. You have to open the box and flip
a jumper to enable the second accelerometer, and the wiring harness needed a bit
of editing (instructions for adding or removing wires to the harness are
included, as are extra connector pins) Once I sorted out which wires went were,
I used 3M "vampire" taps to tap into the OEM sensor wires as they entered the
ECU. When done, I had lateral G, long G, vehicle speed, throttle position, O2
sensor, injector duty cycle, airflow sensor, air inlet temperature, engine RPM,
coolant temperature, brake on/off and battery voltage, plus 2 wires reserved for
steering angle and boost pressure for when I actually buy the sensors.
The unit can auto-start and auto-stop on user-definable criteria, just like
GEEZ!, but I chose to hook up a start/stop switch - press the button, it logs.
Press it again, it stops. Driver proof.
The software is much more basic than GEEZ! No track maps, no sexy gauges, just
strip charts. But you can overlay different runs, and turn on/off display of the
separate channels, and the user interface is more intuative (IMHO). The only
thing I found I missed was some of the calculated channels Buddy Fey uses in his
Data Power book - but the software exports to spreadsheets just fine for doing
that, and I've talked to the programmer at Edelbrock, and a lot of those
features are coming soon.
I got to use the system in anger this past weekend at the Toledo CENDIV, and it
works perfectly. I got all 4 runs logged with no errors or glitches, and when it
came to do the post-race analysis, I found the software gave me the information
I wanted with much less work than my efforts with GEEZ! YMMV.
I also got the suprise of my life. Firstly, I was getting repeated peaks of 1.7G
lateral, with averages in the sweepers of 1.4G. Braking was 1.4G sustained, and
acceleration was 0.6G sustained. Those are fantastic numbers, if they are
accurate.
But most suprising was the revalation that I am a crappy braker. It seems that
in the slow left bordered by the wall of cones, I was braking WAY too early,
then letting up, then braking, then letting up etc. - slowly easing myself into
the corner entrance. If instead I had held off, then done my braking all in one
shot at a full 1.4G, I would have cut 1.5 seconds off my time - in this ONE
CORNER. That 1.5 seconds is signifigant - it's almost exactly the margin of
victory at this event.
It turns out that the biggest problem I have now is not raw cornering potential
compared to the V8s, nor raw power, nor raw torque - it's driver braking
technique, completely NOT what I had thought.
But now I know. :) And now I can do some braking exercises to fix that.
And I also discovered that it's time for a new clutch too, by dividing RPM by
speed, and noticing that the resulting graph wasn't flat w/steps (for gear
changes) but instead all bumpy and wavy.
This datalogger is great. Color me one very happy camper. I think my season just
turned around.
DG
|