G2X datalogger
09-06-2005, 09:08 PM
#1
Registered User
Thread Starter
G2X datalogger
I've been feeling starved for objective feedback on my track days, so I bit the bullet and dropped a few dimes on a G2X datalogger from Racepak. This little unit is a GPS-based system that records your position and speed everywhere on the track. It also has a couple accels (fore/aft, lateral) and can read the car's tach signal (if you tap into the right wires). It has a display unit that shows RPM and GPS speed, as well as lap times if you choose to do so.
I got home tonight and have been figuring out how to look at the data. (Well, first I had to figure out how to find the data, because it recorded not only the six sessions I did but also the trip to the track, the trip back to Covington for gas, the jaunt over to the sub place for lunch, the trip back to the track, etc.)
I was able to select a lap and use that to create a track map. It has an automatic process for dividing the track up into corners and straights, but it didn't really work all that well for PR. Since I wanted to use the nomenclature that everyone else uses, I ended up making the track map mostly by hand. This worked OK once I figured out the sequence (divide up the segments, choose "curve" or "straight", then rename them). (If you divide anything or reclassify a straight or curve, it automagically renames everything back to the default -- which does not include A and B turn names....)
One cool feature is that you can select a session and see all the laps by segment times. So you can see where you had fast segments in slow laps and vice versa. You can plot an individual lap and see rpm and mph (or any of the available data) and then overlay that with data from another lap. It also calculates a "theoretical fast lap" based on the sum of all the fastest segments.
It shows a nifty friction circle "G-G" plot, so you can see where you used all of your tires and where you didn't quite.
And yes, it does give lap times, although it costs a heck of a lot for a lap timer if that's all you use it for.
I grabbed some information from today's sessions:
[CODE]
Fast lap
I got home tonight and have been figuring out how to look at the data. (Well, first I had to figure out how to find the data, because it recorded not only the six sessions I did but also the trip to the track, the trip back to Covington for gas, the jaunt over to the sub place for lunch, the trip back to the track, etc.)
I was able to select a lap and use that to create a track map. It has an automatic process for dividing the track up into corners and straights, but it didn't really work all that well for PR. Since I wanted to use the nomenclature that everyone else uses, I ended up making the track map mostly by hand. This worked OK once I figured out the sequence (divide up the segments, choose "curve" or "straight", then rename them). (If you divide anything or reclassify a straight or curve, it automagically renames everything back to the default -- which does not include A and B turn names....)
One cool feature is that you can select a session and see all the laps by segment times. So you can see where you had fast segments in slow laps and vice versa. You can plot an individual lap and see rpm and mph (or any of the available data) and then overlay that with data from another lap. It also calculates a "theoretical fast lap" based on the sum of all the fastest segments.
It shows a nifty friction circle "G-G" plot, so you can see where you used all of your tires and where you didn't quite.
And yes, it does give lap times, although it costs a heck of a lot for a lap timer if that's all you use it for.
I grabbed some information from today's sessions:
[CODE]
Fast lap
09-07-2005, 07:58 AM
#4
Registered User
Thread Starter
I've seen higher top speeds on my speedo before. However, I wonder just how comparable that is to these numbers. I know that when I am just cruising around (on the highway) the speedo and the GPS readout are within tenths of an MPH difference. I don't know if they diverge at 100+ for any reason, because I was looking at the track, not the readouts.
In fact, I consider that a plus. I was actually less tempted to look at my speedo for information about how I was doing than I have been in the past.
I assume these times are comparible to my best times from the day I was there with you, but I don't know. Except for one Alfa day where they gave us transponders, these are the first lap times I have ever gotten. (On that Alfa day I only got a few clean, dry laps. The best one of those was a 1:47.9.)
In fact, I consider that a plus. I was actually less tempted to look at my speedo for information about how I was doing than I have been in the past.
I assume these times are comparible to my best times from the day I was there with you, but I don't know. Except for one Alfa day where they gave us transponders, these are the first lap times I have ever gotten. (On that Alfa day I only got a few clean, dry laps. The best one of those was a 1:47.9.)
09-07-2005, 11:19 AM
#6
Originally Posted by Race Miata,Sep 7 2005, 01:00 AM
So Mike, are the numbers coincide with your speedo reading and previous lap times? ...
- Michael
Trending Topics
09-09-2005, 06:43 PM
#8
Registered User
Thread Starter
Originally Posted by Tedster,Sep 7 2005, 10:25 AM
Very cool, Mike. Post up some maps and segment times when you have a chance.
I was looking through a session in more detail, and noticed a few interesting things. My exit speeds out of most corners were VERY consistant. I had a lot more variation in the braking zones and entrance speeds. Seems like that's something to concentrate on, I guess.
Also, the one corner where I wasn't very consistant on exit speeds also seemed to be very important for overall lap times. This was T6. Looking at the data it seems like T6 is a big source of the difference between a fast lap and a slower one. (Some of this may have been due to traffic -- there were a few laps where I was stuck behind a P-car with a defective mirror, and the 5/6 complex was a place where he was holding me up.)
My initial efforts at improvement should probably focus on more consistant braking in 3A and 5A, as well as getting onto the throttle earlier and more effectively in T6. In general, my fast laps were the ones where I did best down in the back section of the track.
Next time I am also going to intentionally vary my lines a bit more, and see what happens.
09-10-2005, 01:28 PM
#9
Registered User
Thread Starter
OK here are some screenshots (in PDF format).
http://www.eskimo.com/~mikeg/racepak/1.pdf
This shows a single lap (fast lap of session 5). I have plotted MPH from the GPS, RPM from the tach signal, and the data from the two accelerometers. The data is plotted in "distance mode" and segments are shown. The segments I defined for this are T2, T3A, T3B, T5A, T5B, T6, T8, and T9, as well as the straights between T2/T3A, T3A/T3B, T6/T8, and the front and back straights. I decided that this is how I think of the track while I drive it, so this would be a useful track mapping. (You can see that the lateral G loads in T9 are quite a bit less than in the more severe turns -- T9 is not a turn that requires you to be on the limits of adhesion.)
http://www.eskimo.com/~mikeg/racepak/2.pdf
This shows MPH from two fast laps, one just a bit faster than the other. It shows the "track zoom" feature. I can compare the lines between the two laps. The grey lap was the fast lap, but notice how T3B is pinched off. I think I must have been passing somebody slow right there.
http://www.eskimo.com/~mikeg/racepak/3.pdf
This shows the segment times for all the laps. Segments that were quicker than those of the active lap are highlighted. Notice how the active lap was pretty much the fastest one through the whole back stretch, but was slow through the T3 section? Again, I think I must have been following someone who let me by after T3B. If I had put all my best segments together in one lap, I would have dropped another second off the total.
http://www.eskimo.com/~mikeg/racepak/4.pdf
Here I am comparing segment speeds. Lots of data, but probably the two most useful columns are exit speeds (for the turns) and average speeds (for the turns and straights). (I don't know why the exit speed of some segments is higher than the "max speed". Must be something in how those are defined, or else a fencepost error.)
http://www.eskimo.com/~mikeg/racepak/5.pdf
This is a "G-G" plot (aka friction circle). As I move the pointer along the track, I can see the indicator for both the active lap and the reference lap. Generally they tracked pretty well, but you can see a little bulge here in the active lap at the turn-in for 3B. I wonder what that means, if anything? Notice that I do a lot of braking in corners, trying to "fill in" that friction circle. If I were only braking in straight lines, it would look more like a "T" than a diamond shape. Notice also that I have way more braking and cornering than I do accelerating. This pretty much defines a "momentum car"!
http://www.eskimo.com/~mikeg/racepak/6.pdf
Here's a trace of the MPH over all the laps in the session. You can see where I was help up by traffic. You can also see that some places (like track out from T2) are exteremely tight. I pretty much hit the same speed every time. Other places (like the crest of the hill at T7) show a lot of scatter. This is partly from traffic but also because I am not as consistant in T6 and T7. And sometimes I seem to brake too soon in T3A and especially in T5A. Looking at the segment times, these seem to be more important than differences in the braking zone at T2 or T8 (for me, anyway).
http://www.eskimo.com/~mikeg/racepak/1.pdf
This shows a single lap (fast lap of session 5). I have plotted MPH from the GPS, RPM from the tach signal, and the data from the two accelerometers. The data is plotted in "distance mode" and segments are shown. The segments I defined for this are T2, T3A, T3B, T5A, T5B, T6, T8, and T9, as well as the straights between T2/T3A, T3A/T3B, T6/T8, and the front and back straights. I decided that this is how I think of the track while I drive it, so this would be a useful track mapping. (You can see that the lateral G loads in T9 are quite a bit less than in the more severe turns -- T9 is not a turn that requires you to be on the limits of adhesion.)
http://www.eskimo.com/~mikeg/racepak/2.pdf
This shows MPH from two fast laps, one just a bit faster than the other. It shows the "track zoom" feature. I can compare the lines between the two laps. The grey lap was the fast lap, but notice how T3B is pinched off. I think I must have been passing somebody slow right there.
http://www.eskimo.com/~mikeg/racepak/3.pdf
This shows the segment times for all the laps. Segments that were quicker than those of the active lap are highlighted. Notice how the active lap was pretty much the fastest one through the whole back stretch, but was slow through the T3 section? Again, I think I must have been following someone who let me by after T3B. If I had put all my best segments together in one lap, I would have dropped another second off the total.
http://www.eskimo.com/~mikeg/racepak/4.pdf
Here I am comparing segment speeds. Lots of data, but probably the two most useful columns are exit speeds (for the turns) and average speeds (for the turns and straights). (I don't know why the exit speed of some segments is higher than the "max speed". Must be something in how those are defined, or else a fencepost error.)
http://www.eskimo.com/~mikeg/racepak/5.pdf
This is a "G-G" plot (aka friction circle). As I move the pointer along the track, I can see the indicator for both the active lap and the reference lap. Generally they tracked pretty well, but you can see a little bulge here in the active lap at the turn-in for 3B. I wonder what that means, if anything? Notice that I do a lot of braking in corners, trying to "fill in" that friction circle. If I were only braking in straight lines, it would look more like a "T" than a diamond shape. Notice also that I have way more braking and cornering than I do accelerating. This pretty much defines a "momentum car"!
http://www.eskimo.com/~mikeg/racepak/6.pdf
Here's a trace of the MPH over all the laps in the session. You can see where I was help up by traffic. You can also see that some places (like track out from T2) are exteremely tight. I pretty much hit the same speed every time. Other places (like the crest of the hill at T7) show a lot of scatter. This is partly from traffic but also because I am not as consistant in T6 and T7. And sometimes I seem to brake too soon in T3A and especially in T5A. Looking at the segment times, these seem to be more important than differences in the braking zone at T2 or T8 (for me, anyway).
09-11-2005, 09:24 AM
#10
Registered User
Thread Starter
BTW, at 10:25am on Sunday, my ISP's entire webserver is down. So if you are trying to see those PDFs and getting nothing, it's not you....