Now, let it be known right now, that the data I am going to share is really just for fun, although, I suspect there might be some merrits as will be mentioned later. As I have pulled the trigger on the UMW Stage 2, and currently await delivery, I had been thinking of how to measure the relative gain in performance. Bearing in mind all that has been said about the foibles of dyno testing, and the utility of indicies like 100-200kph times, I decided to "invest" some small change in a new G-Tech ProRR Accelerometer. For US$299, I thought it was a small risk - even if it was a complete dud! So today, I found a quite road and opened my currently stock TT up, and collected some data under WOT.

Firstly, here is a plot of three runs - building courage as I went, unfortunately not quite to the 200kph as I had hoped (will need to find a slightly longer strip of road - no spare airfields around here!)
Attachment 275234
The fastest run (in green) was also the slowest, as I had a fair bit of clutch slip on launch due to poor technique), but in general, the runs all look reasonably similar and consistent. The launch/clutch issue is nicely shown in the next plot of RPM. You can see the uncontrolled RPM surge in 1st in the green run (aughhh - my poor clutch! - upgrade coming
Attachment 275235
Here is a plot of distance travelled, again with good inter run corelation.
Attachment 275236
This is the Long G plot, and allowing for the variables of launch and 1st gear perfomance, the rest looks reproducible with little run to run variability.
Attachment 275237
Finally, here is the plot of HP during the runs. The absolute number is not really the issue, as variables like slope, wind speed and direction, road surface, and weigh (I just entered the DIN weight in the vehicles Specifications), all will have an influence. Still, the issue for me is the reproducibility and the potential to make some guess as to the relative gains in performance when it comes time to give the UMW Stage 2 a run (hopefully in a month or so...)
Attachment 275238
So, the next job is to get some more data, after I fine a longer and safe location. I'll give up on the standing start and will focus on the 100-200kph times. Once established at a given location, in a given direction, at a given temperature, I might just have a cheap and useful tool for comparison within the one vehicle. I am well aware that the AX-22 has the capacity to be more accurate, but I baulked at the price (particularly down here where demand is low and the price high as a consequence). The Perfomance Box was about two to three times the price with some published data suggesting there was little difference in accuracy between the two. Jean has mentioned that he has seen problems with the Perfomance Box (and no doubt accelerometer based technology). I suspect that the comparison of one vehicel to another really does demand very tight accuracy, but when one is really looking at the one vehicle in terms of relative change, this might be different.....?
Has anyone else had experience with a pure accelerometer based tool?

Cheers.

Malcolm

 

Malcolm,

Nice data. I would definitely buy a Performance Box if it is within your budget. It is a great tool for the fun it provides, the long Gs are not very reliable for pure track usage IMO, but if used with an external antenna and well calibrated, it has showed me quite consistent results across all parameters ( I have both). Speed vs. time will be ok.

 

So this G-tech is NON GPS, just accelerometer right ?

If it is then yes some of us were using the early G-techs before the GPS stuff came out. They didn't come with all the super graph capability which you have shared above - thanks

There is some chat about the accelerometers on the 60-130 thread on here I believe ? We used them for doing the 60-130 measurement in the early days but there were problems with accuracy IIRC it was to do with some of the parameters which you enter can have a dramatic effect on the numbers, I think it was the "tilt" or "squat" (can't remember the name but the one which allows for the squat of the car under accleration) feature, depending on which value you entered here you could change the outcome of the data and the problem we concluded was that even using the tool on the same car (993tt) we will all have slightly different suspension setups therefore different amounts of squat so accuracy is out of the window.....

The Performance box is the one to buy, very cheap for what you get and dead simple to use.....