m610

I finally got to drive at Buttonwillow last weekend. Fun, and it's always interesting learning a new track.

I've got lots of video, but I'll annoy you with just this one: http://videos.streetfire.net/video/0...3c4be2e882.htm

It is of the last group "A" session of the day and I was feeling pretty confident, but was still experimenting with how to drive the track, not so much the line, but gear choice, and trying to push things in the turns a bit more, maybe a bit to hard sometimes, to see how I could handle it.

The difference with this video is that I used the BS4TV program I wrote to import G-tech Pro data and create a blue screen display of the data, which I then overlayed onto the track video. It turned out better than I expected! The original 720x480 version looks a lot crisper (but with a 4.7 GB file size), but the online version looks OK.

I'm pretty sure the G-tech's RPM calibration was off on Friday. It was probably low. My shift lights never came on that day. I'll have to recalibrate it. The G-force readings are correct, but uncorrected for body roll or the banking of the track. For instance, 1.1 lateral G's might be hard to believe, but there is at least one place on the track I would believe it and that is the banked hairpin near the end of the lap. The lateral G's might even be higher since the banking would counter the G-force reading. Otherwise, the G-forces display is a handy gauge of driving effort and technique.

I've been watching the video more to see how I handled different turns, or how I handled them differently throughout the session. One thing I'm noticing is that when I do a turn right the lateral G's are about constant. When I am struggling, there is a spike in the lat G's at the maximum screw-up point, but otherwise the G's are lower. Now I am watching the video to see how I am managing the throttle through the turns.

I'm looking forward to my next Thunderhill date so I can fix the RPM logging problem AND drive on a familair track.

I also need to drive a skid pad circle so I can work out G-force corrections for body roll.

__________________

yield

Looks great. I'm thinking about picking up a gtech or something similar eventually. WHy would you correct the g-forces though. Is that what you as a driver are experiencing.

One comment about the video - I remember that you have a bunch of cameras and have played around with the mic location some. Personally, I still think the wind noise is very annoying and I have to turn the volume way down. Have you tried mounting it someplace like in the trunk so you'd get the sound from the muffler but not the buffeting?

__________________

m610

That's true, and it certainly helps explain why my arms were so sore and my lower left leg bruised before I got my seats and 4-points. The driver is feeling 1.1 G's due to turning and body roll, even if the tires aren't actually giving 1.1 G's of traction. It's just that people don't believe the numbers, then think it is all BS. I'll put the correction in the software as an option.

I tried trunk mounting the mics and it didn't sound good, kinda like how music played in the next apartment sounds like coming through the wall. I don't think wind noise is as much a problem with my current setup as is the overall sound levels. I'm going to try putting a resistor (voltage divider) in the circuit next time, and eventually trying to come up with an automatic gain control circuit.

__________________

yield

Yeah, the clipping sucks. There has to be some trick to it as car shows usually pull it off quite well, but they obviously have very nice mics, etc.

Any other complaints or thoughts about the gtech? I'm trying to get a friend to go in on one with me. The software looks interesting.

__________________

m610

If I could drive with the windows rolled up, the sound in the cabin would be very usable. I've watched a lot of scenes on TV where they are filming a car on the highway, etc., and have decided that they probably use a different sound track altogether.

RE the G-tech -- I like mine and I think it is a great deal for what you get. It records readings from three accelerometers plus RPM. Using that info you can work out horsepower, torque, lots of other stuff, withing reason.

Most of the complaints I have heard have to do more with people expecting a lot more, and not understanding what they've got. For instance, some people can't understand why the horsepower and torque measurements (drag race mode) are lower than their dyno results. The reason is mostly aero losses, but also how the front end lifts durng acceleration. Others run into problems by not setting the accelerometers (compenstates for how you mount the unit) correctly. Then there are things like doing an adequate rollout, etc.

The G-tech doesn't work with some cars because it can't get an RPM signal from the DC power. Apparently it finds and counts noise spikes on the 12v DC lines.

Setting up the G-tech is easy. Using it is easy. Downloading the data and using the PASS sofware is easy.

Choose carefully. The SS and RR units are different, and I think only the RR unit will let you download and export the data. I've got a G-tech Pro Competition. You might be able to find a used one on ebay.

The link to the G-tech forum is: http://www.gtechproforums.com/forums/

__________________

yield

Thanks for the insight. I was only considering the RR as I don't really care about the HP readings too much. It's obviously somewhat limited in what it can do but I don't think I'd be willing to spend more on a more sophisticated unit.

When you add it to your video I think it definitely helps to analyze your drive. I also heard somewhere that there was a hack that allowed you to plot your info on a track outline. Ever heard of that?

__________________

m610

I have. It is listed on the GTech forum somewhere, but last I heard the guy writing it has quit working on the program to move onto another, bigger, project. He was offering to sell the source code to GTech. I might have an earlier verson, but I couldn't get it to work right. You have to specify your car's roll rates, etc. and I don't know what to put in for this. Also, I know from experience with my program that small errors accumulate quickly when you do something as simple as calculate your speed throughout your session.

Speed = G's * time interval * 32ft/sec^2 + previous speed

After more than 30-60 seconds (300-600 data points) the numbers are ridiculous. For 10-15 second drags, it might be OK.

Found the thread. http://www.gtechproforums.com/forums...read.php?t=969

__________________