New feature on ClubRegistration
#16
That very topic has caused me a great deal of joy this spring at TWS going counterclockwise into T1. T1 has always been a scary corner for me, coming off the banking at 120+ and it looks so short. I've had the benefit of VR talking about this specific to T1 at TWS during the PCA DE classroom sessions.
It used to be a heavy braking zone followed by horsing the car's nose into T2. Glancing at the speedo i'd be doing about 72 at the apex of T2 and I've managed to spin the car at that point at about that speed. It felt fast, but the data showed it wasn't. I always get on the brakes right after the oval transition - I just can't break that habit and now I don't think I have too.
I get on the brakes at the same point, but I don't get on them nearly has hard. It took a while, but just feeling a little bit of braking settled the knot in my stomach and settled the car. I just kept backing up the point where I got hard on the brakes and once there I even tapered off how hard I got on the brakes. I find it amazing just how hooked up the car can get entering T2 and I'm clipping that apex now at 80+ MPH and exiting at over 90MPH - and it's easy peasy. I don't even consider it a heavy braking zone any more and it's fun to reel in high horsepower cars at that spot . It was also quite a realization when I noticed I was in T1, hauling the mail, with the car loaded up in a turn, and my foot on the brakes - I never straighten the wheel or the car out during the process.
Best part is, this simple change has netted me about 2 seconds, going from 2:04-2:05 down to 2:02-2:03. I though my days of getting seconds worth of improvement were over. My data acquisition shows almost all of it is from maintaining entry speed into T2.
Is it worth me digging up video of both methods through T1-T2? Same car, same driver, comparable tires.
-Mike
It used to be a heavy braking zone followed by horsing the car's nose into T2. Glancing at the speedo i'd be doing about 72 at the apex of T2 and I've managed to spin the car at that point at about that speed. It felt fast, but the data showed it wasn't. I always get on the brakes right after the oval transition - I just can't break that habit and now I don't think I have too.
I get on the brakes at the same point, but I don't get on them nearly has hard. It took a while, but just feeling a little bit of braking settled the knot in my stomach and settled the car. I just kept backing up the point where I got hard on the brakes and once there I even tapered off how hard I got on the brakes. I find it amazing just how hooked up the car can get entering T2 and I'm clipping that apex now at 80+ MPH and exiting at over 90MPH - and it's easy peasy. I don't even consider it a heavy braking zone any more and it's fun to reel in high horsepower cars at that spot . It was also quite a realization when I noticed I was in T1, hauling the mail, with the car loaded up in a turn, and my foot on the brakes - I never straighten the wheel or the car out during the process.
Best part is, this simple change has netted me about 2 seconds, going from 2:04-2:05 down to 2:02-2:03. I though my days of getting seconds worth of improvement were over. My data acquisition shows almost all of it is from maintaining entry speed into T2.
Is it worth me digging up video of both methods through T1-T2? Same car, same driver, comparable tires.
-Mike
#19
Rennlist Member
Related to this braking technique is trying to hold a higher gear at turn-in as a balanced car helps you get on the gas and to full throttle sooner, making the higher gear possible and more effective. The TWS T1 example above is a great demonstration of this. With the Cup car gears in my 996, I downshift from 6th to 5th in the braking zone before T2 and carry that all the way to T3. Getting this right is as much fun as taking the Kink at RA at full tilt! One of my favorite corners anywhere.
#20
Rennlist Member
I'm going to read all of your pearls Dave.
One of these days I'm going to try to use the brake pedal.
One of these days I'm going to try to use the brake pedal.
#21
early lighter braking and G sum
congrats dave on a great article. may i ask you a question. does this technique change the way we look at a "combined g" (G sum ) channel . it seems to me that rather than seeing a smooth rise in g-sum there would be a valley just before the lat-g's pick up . in my mind this correlates to a wider but squatter traction circle ,all because during the braking phase we are developing less long-g . does this make sense
the reason i ask is that i am trying to figure out a data channel that can be used to determine how early and how much lighter the braking should be . i know that lap or sector times would be helpful but i am looking for something more immediate
thanks
the reason i ask is that i am trying to figure out a data channel that can be used to determine how early and how much lighter the braking should be . i know that lap or sector times would be helpful but i am looking for something more immediate
thanks
#22
Rennlist Member
Thread Starter
Hmm. Great question. I guess I will hedge my answer and say that it depends on how much trail braking we do. Even with less of a late & "surprising" brake spike, we still don't want to waste time. Thus, if we're able to trail brake (arguably easier with slightly less initial brake spike), and efficiently use the brake zone for braking & steering, I am not sure that valley would be there...
#23
Rennlist
Basic Site Sponsor
Basic Site Sponsor
Join Date: Jun 2008
Location: Durham, NC and Virginia International Raceway
Posts: 19,150
Received 3,331 Likes
on
1,891 Posts
does this technique change the way we look at a "combined g" (G sum ) channel? it seems to me that rather than seeing a smooth rise in g-sum there would be a valley just before the lat-g's pick up.
in my mind this correlates to a wider but squatter traction circle, all because during the braking phase we are developing less long-g . does this make sense?
the reason i ask is that i am trying to figure out a data channel that can be used to determine how early and how much lighter the braking should be .
i know that lap or sector times would be helpful but i am looking for something more immediate
thanks
in my mind this correlates to a wider but squatter traction circle, all because during the braking phase we are developing less long-g . does this make sense?
the reason i ask is that i am trying to figure out a data channel that can be used to determine how early and how much lighter the braking should be .
i know that lap or sector times would be helpful but i am looking for something more immediate
thanks
The smooth rise championed by Dave's "soft brake" approach is separate from the GSum measure at the end of braking that you have the question about.
A dip at the end of the brake application before the rise due to the building of LatG would not be a characteristic of "soft" or "hard" braking at all, but instead, an indicator that the driver likely braked too soon, let off the brakes early and did NOT smoothly transition off the brakes CLOSE enough to the turn in to allow full use of the tire. The G-G would show this dynamically, if animated.
Your estimate of what the G-G would look like with Dave's approach is correct, and the GSum would go from zero to some portion of demonstrated max during the "soft braking," then rise to max (hopefully) as throttle and latG built up.
Our job as drivers is to use most fully the available contact patch, which is best expressed by GSum. Ideally, you would see a rise from zero or slightly above driving down the fast straight, go up relatively swiftly (as swiftly and steeply as it takes to keep the car stable or near stable) to max and STAY there through the beginning of the track out phase, when latG's fall as the steering wheel is straightened. IF there was a dip, move the braking zone as a whole up to, then INTO the cornering zone.
We know that most cars should and can brake at 90%-95% of maximum SUSTAINED cornering loading as expressed in latG. The Friction Circle (G-G) should extend at least as far up or down (whatever direction your analysis software chooses to display it) as it does to the sides. Most importantly, there should be a convex, rounded pattern between full brake and full cornering, with NO concavity and hopefully, NOT a straight line between LongG max and LatG max. It's really a very simple and powerful representation of what you are doing.
In my opinion, there is no data channel that can be used to objectively measure an introduced mitigation of demonstrated capability. One person's "soft" is not another person's "soft," so how can that be duplicated every application? If the GSum rises to half the peak value, that is half the tire unused and twice the distance required to shed the same speed.
Sector times would tell the story, too, but not without spending a lot more time comparing them, and you would still need to quantify LongG and/or GSum along with it the come to any conclusion.
Andrew Davis said in his chalk talk today, "hammer those brakes" into T1, T6 and T10A, but to respect that some of the tire grip is used for turning while braking at T3 and T5, so balance the brake application while preserving some of the tire for cornering grip. It was fun to hear that.
The key is, whatever approach you use, formulate a repeatable, super consistent way of doing things, then measure. Let the data guide you, because we KNOW what it's SUPPOSED to look like!
__________________
-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
#25
Thanks, may I explore this a bit more. The radius of any corner is a constant for a given line. The idea of Earlier-Lighter-Braking(ELB)is to get more velocity into the corner and in particular to have a faster minimal velocity. This will result in a predictable or desired max lat g at that corner . I know it sounds silly to say but to me this is the “good” g and is represented by the later part of the g-sum and by the width of the traction circle.
It seems to me that you want to achieve the above (good G) with the minimal amount of braking . this would be the “bad” g. taken to its ridiculous extreme you would like to have the ideal good g with 0 bad g . this is kind of different than what is classicly taught. the classic way of teaching cornering is with late max braking, then maybe trail brake in . that is why channels such as “brake aggression” were developed . you can see just how good you were on the brakes. Max long g was built up quickly then released into turning. So you were a “hero “ if you could drive that long-g up as rapidly as possible and get that traction circle to open up . ELB isn’t like this . I think the rise of the g sum would be longer with a smaller slope ,and then the peak rise would be at max lat g.
So I was thinking that what you really want is highest maximal G with the smallest area under the G sum curve (ie,integrate G sum ). You can write a math channel or develop a plot to visualize this . why does this matter? Well if you are going to use ELB then the question of how early or how much lighter to apply the brake can be experimented with to maximize the result
Do you think any of this makes sense
It seems to me that you want to achieve the above (good G) with the minimal amount of braking . this would be the “bad” g. taken to its ridiculous extreme you would like to have the ideal good g with 0 bad g . this is kind of different than what is classicly taught. the classic way of teaching cornering is with late max braking, then maybe trail brake in . that is why channels such as “brake aggression” were developed . you can see just how good you were on the brakes. Max long g was built up quickly then released into turning. So you were a “hero “ if you could drive that long-g up as rapidly as possible and get that traction circle to open up . ELB isn’t like this . I think the rise of the g sum would be longer with a smaller slope ,and then the peak rise would be at max lat g.
So I was thinking that what you really want is highest maximal G with the smallest area under the G sum curve (ie,integrate G sum ). You can write a math channel or develop a plot to visualize this . why does this matter? Well if you are going to use ELB then the question of how early or how much lighter to apply the brake can be experimented with to maximize the result
Do you think any of this makes sense
#26
Rennlist Hoonigan
which cost no drachmas
Lifetime Rennlist
Member
Rennlist
Site Sponsor
which cost no drachmas
Lifetime Rennlist
Member
Rennlist
Site Sponsor
Thanks, may I explore this a bit more. The radius of any corner is a constant for a given line. The idea of Earlier-Lighter-Braking(ELB)is to get more velocity into the corner and in particular to have a faster minimal velocity. This will result in a predictable or desired max lat g at that corner . I know it sounds silly to say but to me this is the “good” g and is represented by the later part of the g-sum and by the width of the traction circle.
It seems to me that you want to achieve the above (good G) with the minimal amount of braking . this would be the “bad” g. taken to its ridiculous extreme you would like to have the ideal good g with 0 bad g . this is kind of different than what is classicly taught. the classic way of teaching cornering is with late max braking, then maybe trail brake in . that is why channels such as “brake aggression” were developed . you can see just how good you were on the brakes. Max long g was built up quickly then released into turning. So you were a “hero “ if you could drive that long-g up as rapidly as possible and get that traction circle to open up . ELB isn’t like this . I think the rise of the g sum would be longer with a smaller slope ,and then the peak rise would be at max lat g.
So I was thinking that what you really want is highest maximal G with the smallest area under the G sum curve (ie,integrate G sum ). You can write a math channel or develop a plot to visualize this . why does this matter? Well if you are going to use ELB then the question of how early or how much lighter to apply the brake can be experimented with to maximize the result
Do you think any of this makes sense
It seems to me that you want to achieve the above (good G) with the minimal amount of braking . this would be the “bad” g. taken to its ridiculous extreme you would like to have the ideal good g with 0 bad g . this is kind of different than what is classicly taught. the classic way of teaching cornering is with late max braking, then maybe trail brake in . that is why channels such as “brake aggression” were developed . you can see just how good you were on the brakes. Max long g was built up quickly then released into turning. So you were a “hero “ if you could drive that long-g up as rapidly as possible and get that traction circle to open up . ELB isn’t like this . I think the rise of the g sum would be longer with a smaller slope ,and then the peak rise would be at max lat g.
So I was thinking that what you really want is highest maximal G with the smallest area under the G sum curve (ie,integrate G sum ). You can write a math channel or develop a plot to visualize this . why does this matter? Well if you are going to use ELB then the question of how early or how much lighter to apply the brake can be experimented with to maximize the result
Do you think any of this makes sense
So, when entering a corner, it should rise with the long G rise in braking. Then you want to start adding in lat G as the long G braking decreases while mainting the total so that the line stays as high as possible. I've put two screen shots showing this as well. To see it in the G-G graph, check out http://www.trailbrake.net/2/post/201...g-diagram.html
#27
Rennlist Member
Look forward to reading more.
#28
In my experience, this is the only way to shave those last few seconds off a lap time. It makes sense, as the brakes are so powerful compared to other driver controls.
-Mike
#30
matt, i very much appreciate the discussion and far from being argumentative i do want to challenge what you said above . i absolutely agree that the classic description of "ideal "g sum and traction curve are as you describe above . but this is predicated on the classic late max brake . what dave is describing is different , and rather than getting the traction curve to approximate a circle it more approximates a disc . i think (but am not sure ) that we are sacrificing long-g (and the inherent load transfer ) in order to get maximal lat -g of the combined 4 tires. in a sense i am really asking if the concept of the traction circle is really valid. the traction circle says to always be at the ragged edge of adhesion . always max long -g ,lat-g and the combination of the 2 . but is this really what we want . is max negative long-g (braking) really good or are we better off not being at max long -g so that we can get more lat-g .
this really isn't just a theoretic discussion . i have seen dave's description of earlier lighter braking described a number of times and my coach strongly advocates it .this description is usually accompanied by speed graphs showing the earlier braking with higher min speed at a corner (and a s a bonus earlier throttle out ) . i am just trying to figure out how to measure this
this discussion has already helped me a lot . going into it i thought that the g-sum curve using daves technique would necessarily result in a valley before the peak. now i think the ramp up to the peak will be more gradual , no valley(if done right) and maybe even a higher peak
this really isn't just a theoretic discussion . i have seen dave's description of earlier lighter braking described a number of times and my coach strongly advocates it .this description is usually accompanied by speed graphs showing the earlier braking with higher min speed at a corner (and a s a bonus earlier throttle out ) . i am just trying to figure out how to measure this
this discussion has already helped me a lot . going into it i thought that the g-sum curve using daves technique would necessarily result in a valley before the peak. now i think the ramp up to the peak will be more gradual , no valley(if done right) and maybe even a higher peak