I just wanted to make sure that I posted this data since it wasn't part of the original two sets of data that I posted. I also want everyone to make a note of the fact that the Vbox fails to have a consistent reading on the height of the track. This is one of the known sources of error in the GPS, vertical positional readings are subject to larger errors than horizontal positional readings. However, interestingly enough, it's easy to see in the data when the vertical readings are erroneous but not so much the horizontal readings. However, it's clear that there is a 10% deviation in vertical measurements from run to run at the same track. Even the dimmest bulb can understand how large that error is.

That night the track was kind of slick on the back end and I had a nail in my left rear inside sidewall but didn't know it. I found out the next day. The tire actually started deflating on the way home but I had let air out and put it back in at the track so of course I didn't notice anything that way. Anyway, on the first run you can see I backed off before the end of the quarter but I still beat the modded c6 vette that ran a 12 which he was happy with. He said it was his fastest ever time. Later I killed a heavily modded c6 z07 vette with drag radials that ran an 11.1 and had asked to race me.

You should notice that there is less error in the 1/8th mile traps and the 60ft times and ETs are all in pretty good agreement. I have explained why I think this is so in other threads. It is the final trap speed that is always slower by several miles per hour that I have an issue with. I have also explained that as well. This track ran an IHRA event about a month before I was out there so you can imagine their timing stuff is accurate. I also have data from Texas Motorplex which varies similarly and it is where NHRA events are held on a regular basis up in Ennis, Texas.

So I have to say that while I like the perfdb website and appreciate the work that Pencilgeek has put in to giving us this nice tool, the 1/4mile trap speeds are just not accurate due to the limitations of the GPS. The car is running too fast given the GPS measurement errors in order to hit the track trap points with collected data points. Pencilgeek attempts to make up for this with some mathematically unsupported calculations that appear to consistently underestimate final trap speed. As I have said, I like the tools, they are free and fun but not completely accurate.

Pencilgeek should feel free to come post other data that matches but that won't change the fact that GPS measurements are subject to random error. In some cases the errors may not be there and the samples are close enough at the end of the run to produce trap speeds that are close to the ones measured by the track timers. Whenever there is a difference though, I would expect the number in error to be the one produced by the perfdb website. I am not saying that tracks are not subject to measurement errors but certainly less so than the GPS is and they have more stake in their numbers than those from the perfdb website.

 

nice write up

 

Great data.