Braking G's with R6's?
02-24-2011, 10:12 AM
#16
Rennlist Member
This is turning into a great discussion. Always enjoy VR's posts, as they have real-world value.
Certainly, when people get their braking systems working well and their confidence up, they get "punch drunk" with braking power and become addicted to stabbing the pedal, generating that spike and feeling it as the forward resistance of their shoulder harnesses digs into their chest and shoulders. I know I did. To my detriment. The peak spike was not what was important, it was what came after!
What is much more important, IMO, is consistency and being able to sustain that deceleration level over the course of a longer braking zone. Then, of course, measuring the efficacy of trail braking by overlapping the longitudinal g (braking force) with the rise of lateral g (cornering force) to see how well the braking is melded into the beginning of the cornering phase. A gSum (combined g) will do that too. Any precipitous drop in gSum between the end of braking and mid-corner lateral loading is wasted traction circle. So really, it's two issues. Braking efficiency (losing speed in the shortest distance) and staying on the periphery of the friction circle (a goal we should ALL have!
)
I look at data from Tom Long, Leh Keen or David Murry (as well as a lot of "normal" people I consider accomplished in the art of proper braking) and they're generating on the high side of the range, but for the shortest possible distance and MUCH closer to the cornering phase than the average joe.
Remember, the more the aero, the more you'll (or should see) a higher initial decel rate, trailing off 15-30% as the speeds come down through the longer brake zones. Also, uphill zones like T14 at VIR (and downhill with compression, like the Toe of the Boot at the Glen) can skew long g data. I've seen over -1.75 for the same car I'm seeing -1.32 on the flat.
Certainly, when people get their braking systems working well and their confidence up, they get "punch drunk" with braking power and become addicted to stabbing the pedal, generating that spike and feeling it as the forward resistance of their shoulder harnesses digs into their chest and shoulders. I know I did. To my detriment. The peak spike was not what was important, it was what came after!
What is much more important, IMO, is consistency and being able to sustain that deceleration level over the course of a longer braking zone. Then, of course, measuring the efficacy of trail braking by overlapping the longitudinal g (braking force) with the rise of lateral g (cornering force) to see how well the braking is melded into the beginning of the cornering phase. A gSum (combined g) will do that too. Any precipitous drop in gSum between the end of braking and mid-corner lateral loading is wasted traction circle. So really, it's two issues. Braking efficiency (losing speed in the shortest distance) and staying on the periphery of the friction circle (a goal we should ALL have!
)I look at data from Tom Long, Leh Keen or David Murry (as well as a lot of "normal" people I consider accomplished in the art of proper braking) and they're generating on the high side of the range, but for the shortest possible distance and MUCH closer to the cornering phase than the average joe.
Remember, the more the aero, the more you'll (or should see) a higher initial decel rate, trailing off 15-30% as the speeds come down through the longer brake zones. Also, uphill zones like T14 at VIR (and downhill with compression, like the Toe of the Boot at the Glen) can skew long g data. I've seen over -1.75 for the same car I'm seeing -1.32 on the flat.
Professional Racing and Driving Coach
02-24-2011, 10:57 AM
#17
Rennlist
Basic Site Sponsor
Basic Site Sponsor
Join Date: Jun 2008
Location: Durham, NC and Virginia International Raceway
Posts: 19,034
Received 3,167 Likes
on
1,831 Posts
This is one reason why I don't use the "Segment Analysis" spreadsheet on TraqView or TraqStudio to quantify braking g peaks anymore and focus only on the graph and graph playback, in that particular program.
The smoothing algorithm present in the Traqmate Accel/Braking g versus Distance (preferred, as opposed to g versus Time) graph is much more accurate than the peaks you see in the display of the segment values.
Having run MoTeC, AiM, Race-Keeper, Race Technology DL-1, Traqmate and Video VBox in the same cars (not quite all at one time, but up to four systems simultaneously to validate data and accuracy between them), I can assure you that mounted (and/or calibrated, if they require that step) correctly, they're all close enough to get very accurate and valuable data, better than we can drive!
__________________
-Peter Krause
www.peterkrause.net
www.gofasternow.com
"Combining the Art and Science of Driving Fast!"
Specializing in Professional, Private Driver Performance Evaluation and Optimization
Consultation Available Remotely and at VIRginia International Raceway
-Peter Krause
www.peterkrause.net
www.gofasternow.com
"Combining the Art and Science of Driving Fast!"
Specializing in Professional, Private Driver Performance Evaluation and Optimization
Consultation Available Remotely and at VIRginia International Raceway
02-24-2011, 11:12 AM
#18
Rennlist Member
One of the valuable tools we just were educated on at a Data training seminar was the ability to create math channels that graph gsum (yes, it's back in discussion) during cornering as Peter points out. When the software allows you to build your own math channels then you can quickly display to the driver where improvements can be made. Another cool data channel is time vs reference lap improvements which typically demonstrates the psychological progression when your on your game. It was interesting to look at the affects on a drivers confidence when they "blow a corner" and have difficulty getting back on their game for an entire lap. The time difference graph really displays another issue not considered by the end user.
Great discussion. I hope I am not to far off the OP subject.
Great discussion. I hope I am not to far off the OP subject.
02-24-2011, 08:04 PM
#19
Rennlist
Basic Site Sponsor
Basic Site Sponsor
Join Date: Jun 2008
Location: Durham, NC and Virginia International Raceway
Posts: 19,034
Received 3,167 Likes
on
1,831 Posts
One of the valuable tools we just were educated on at a data training seminar was the ability to create math channels that graph gsum (yes, it's back in discussion) during cornering as Peter points out.
<<snip>>
Another cool data channel is time vs reference lap improvements which typically demonstrates the psychological progression when your on your game.
<<snip>>
The time difference graph really displays another issue not considered by the end user.
<<snip>>
Another cool data channel is time vs reference lap improvements which typically demonstrates the psychological progression when your on your game.
<<snip>>
The time difference graph really displays another issue not considered by the end user.
"Time Gap," "Time Slip" or "Delta-T" are all the same way, in real time and at particular points on the track, to compare performance differential between two laps. Again, Race-Keeper, Traqmate, Race Technology, AiM, MoTeC and Video VBox ALL have this pre-loaded. No math channel experience necessary!
It all fits into the idea that most folks do eighty percent of it right (or close to right) and if they could find the twenty percent that holds the key to improvement, they could progress MUCH faster...
02-24-2011, 08:53 PM
#20
Lifetime Rennlist Member
To add a bit to Peter and g sum...
Combined g has been around forever but, IMO, is limited in it's value. You can draw the g-g graph but it really doesn't tell you that much. If it looks like a cross, then clearly you are not using all the available traction. But when it looks like a squished circle, all you can really say is that you are using more of the available traction. The reason is that we just don't know (in most cases) where the outer edge of that circle should be. Each corner has it's own edges and they change lap after lap. If you have a ton of data on identical cars on the same track, then it would be a useful tool.
The issues I had with g sum were:
1) Why did someone find the need to take a standard term (combined g) and rename it?
2) ColorChange, who was big on g sum, asserted that the fastest laps came from max g sum around the track. I created a math channel and integrated combined g for each lap of one of my run offs races. There was no correlation between that and lap time and I posted that data somewhere.
3) The same guy who advocated #2 also asserted that the fastest way around the track was to threshold brake every corner all the way to the apex. He proceeded to launch a bunch of F-Bombs at one of the folks around here who disagreed with that theory.
Bottom line is that there are far better things to look at in your data than combined g's to go faster.
Combined g has been around forever but, IMO, is limited in it's value. You can draw the g-g graph but it really doesn't tell you that much. If it looks like a cross, then clearly you are not using all the available traction. But when it looks like a squished circle, all you can really say is that you are using more of the available traction. The reason is that we just don't know (in most cases) where the outer edge of that circle should be. Each corner has it's own edges and they change lap after lap. If you have a ton of data on identical cars on the same track, then it would be a useful tool.
The issues I had with g sum were:
1) Why did someone find the need to take a standard term (combined g) and rename it?
2) ColorChange, who was big on g sum, asserted that the fastest laps came from max g sum around the track. I created a math channel and integrated combined g for each lap of one of my run offs races. There was no correlation between that and lap time and I posted that data somewhere.
3) The same guy who advocated #2 also asserted that the fastest way around the track was to threshold brake every corner all the way to the apex. He proceeded to launch a bunch of F-Bombs at one of the folks around here who disagreed with that theory.
Bottom line is that there are far better things to look at in your data than combined g's to go faster.
02-25-2011, 11:16 AM
#21
Three Wheelin'
I agree. Don't use the peak value your sensor puts out. There is a lot of variation and noise coming out of an accel because the braking event is not a 100% smooth event. What you should do is use the average for the entire brake event if you're looking for the overall decel (at least for straight line braking). For an R6 on a car without significant aero somewhere in the 1.2-1.4 range sounds reasonable.
02-25-2011, 08:28 PM
#22
Moderator
Rennlist Member
Rennlist Member
Curiosity got me to check some TraqMate data from Watkins Glen. Maximum with 245/305 RA1s with Yellow 29s on a 997S (otherwise stock) was 1.47 g braking. Laps in the 2:13-2:14 range. Usually in the 1.2-1.3 range, so that 1.47 must have been an emergency stop (there was one that event, when I had somebody spin in front of me at 150 going into the Bus Stop. Left and Right Lateral Gs in the 1.5s. Have to love the old RA1 (which is why I have 2 extra sets in storage).
Wow, that was back in 2008 and I haven't bothered to hook the unit up to my 09 C2S Sport Pasm yet. Will have to think to do it this season and see how it compares with the 07.
Wow, that was back in 2008 and I haven't bothered to hook the unit up to my 09 C2S Sport Pasm yet. Will have to think to do it this season and see how it compares with the 07.
02-25-2011, 09:26 PM
#23
Rennlist
Basic Site Sponsor
Basic Site Sponsor
Join Date: Jun 2008
Location: Durham, NC and Virginia International Raceway
Posts: 19,034
Received 3,167 Likes
on
1,831 Posts
09-28-2013, 10:26 PM
#24
Rennlist Member
Joe:
At MO at the back straight I am at about 1.24 G. I brake for 200' and about 1 sec.
At MO at the back straight I am at about 1.24 G. I brake for 200' and about 1 sec.
09-30-2013, 09:38 PM
#25
Rennlist
Basic Site Sponsor
Basic Site Sponsor
Join Date: Jun 2008
Location: Durham, NC and Virginia International Raceway
Posts: 19,034
Received 3,167 Likes
on
1,831 Posts
And go from what speed to what speed?
09-30-2013, 11:51 PM
#26
Rennlist Member
The Traqmate in my '11 Boxster S with Michelin Pilot Sport Cups showed 1.31G going into the toe of the boot. But I also agree that the one number isn't the whole picture, as other posters have said.
