Peak G readings - do they mean anything?

[pentoman]

I'm trying to test a number of cars round the same bit of track and get a lateral G figure for them. However, I am finding it hard to get reliable results.

Using the summary statistics I can view the "peak" G, but this is not that useful. It seems if you throw the car from left to right, you can get a very high brief peak, but this does not really indicate much. Is there some way to get a "peak sustained G" or something for the highest sustained G a car acheived?

thanks

Russell

[ewaher]

use the 90% filter or 95% filter for your G readings and see how your data looks after that.

[pentoman]

Thanks for the reply

I tried that, but it seemed to take a 95% average of the peak G (can't see how that would be useful really). I'll give it another go - perhaps I'm mistaken.

Russell

[ewaher]

Hm, unless I am misinterpreting you, you just need to set up a new reading in the summary statistics with:

Mode: 95% max and another with 95% min, for lateral G. That will give you approximately what the car is capable of doing on the track.

From the RT manual...

"95% Max Of"
"This is the maximum value of the variable between the start and endpoints after the highest 5% of values have been removed. Mathematically this is the 95th percentile of the samples. Note that parameter is calculated using the actual data, and not assuming a normal distribution."

More than likely you can use the 98% reading, since G spikes are very short and infrequent.

[pentoman]

Right OK , I think I was using 98% average (duh), will have a look for 98% max.