Data v. Whiny Idiots (Throttle Response Edition)
#101
I have no idea of Sav's driving ability. Maybe he is excellent.
I can tell you FOR SURE, that if a driver at corner exit only ever rolls into the throttle from some lower amount held steady in a turn (as he suggests is optimal) as he gently unwinds the steering and adjusts the satellite radio then that driver is NOT getting everything out of the car. Cuz stuff happens and you have to react to it.
The fastest drivers wind up having busy feet and hands. It is not a Sunday Drive with everything lined up perfect. Many fast / little corrections must be made at times. If you slow down a bit, the need for little adjustments decreases. This was one of the strengths Senna enjoyed. His reaction time was about 1/2 of normal but excellent people. So on a test where a decent person can hit 0.2 seconds, he could be under 0.1 seconds. This allowed him to make corrections and inputs before things got too far out of kilter.
You don't have to believe me, just track down data traces of name drivers during key laps and "events". I posted a sample from F1. Or ride shotgun with one of the few "Alien" autocross drivers out there…they are quite busy indeed.
I can tell you FOR SURE, that if a driver at corner exit only ever rolls into the throttle from some lower amount held steady in a turn (as he suggests is optimal) as he gently unwinds the steering and adjusts the satellite radio then that driver is NOT getting everything out of the car. Cuz stuff happens and you have to react to it.
The fastest drivers wind up having busy feet and hands. It is not a Sunday Drive with everything lined up perfect. Many fast / little corrections must be made at times. If you slow down a bit, the need for little adjustments decreases. This was one of the strengths Senna enjoyed. His reaction time was about 1/2 of normal but excellent people. So on a test where a decent person can hit 0.2 seconds, he could be under 0.1 seconds. This allowed him to make corrections and inputs before things got too far out of kilter.
You don't have to believe me, just track down data traces of name drivers during key laps and "events". I posted a sample from F1. Or ride shotgun with one of the few "Alien" autocross drivers out there…they are quite busy indeed.
Last edited by sjmarcy; 08-26-2011 at 02:29 AM.
#103
Former Vendor
Thread Starter
iTrader: (31)
Join Date: Nov 2006
Location: Sunnyvale, CA
Posts: 15,442
Total Cats: 2,104
http://www.youtube.com/watch?v=luHGT...tailpage#t=24s
There are many different conditions and circumstances possible for such things.
When throttle is applied at a slower rate and at performance RPM ranges the NA results improve a bit.
When you are at the cornering limit…throttle can be moved as desired. It is not always a simple roll-on from some lower but steady throttle opening.
And there are also things called throttle-kicks and downshift blips.
If you'd like to show that a car is at the limit and had the throttle behaving well you don't show that using the type of graphs you have shown thus far, backed up with anecdotes.
I disagree, BTW- I think that you're the one that's doing a **** job of showing that behavior. You're using g-force in all of these charts, which inherently introduces a litany of additional variables into the experiment - you yourself admitted this by talking about the sine wave as the driveline lash plays itself out.
Why wouldn't you simply compare TPS vs. MAP traces? Since there is a very close correlation between manifold pressure and hub torque output (aka, two cars with the same manifold pressure will make virtually the same hub torque output), why wouldn't you use that data instead of the g-data you're so fond of, which is incredibly vulnerable to outside influences? Is there something about turbocharged cars that somehow alters the time-independent correlation between manifold pressure and hub torque? Jason asked you this same question in post 44 - you refused to answer it then, too.
Let's simplify things even further - we don't even need to be on a racetrack, nor at the cornering limits of the car to look at this. Rolling down the highway, 4th gear, 5000rpm, and slowly roll into the throttle. Will a turbo car display the same lag as if you had snapped the throttle open from idle? I'll freely admit that turbo cars do not make instant boost with perfect TPS correlation from any RPM - what I am arguing is that in the real world, the TPS/output correlation is much, much better than you seem to think it is. Instead of dancing around my argument by changing the conditions, why don't you just offer a rebuttal?
#106
Junior Member
Join Date: Dec 2010
Location: UK, in Cambridgeshire or wherever work takes me.
Posts: 375
Total Cats: 20
Sav is a good driver. For Sav's particular driving style (which, being his thread, is what's most relevant) his car responds absolutely fine. No, it won't respond as cleanly as a decent non-turbo car if you're snapping the throttle on and off, but that's not what Sav does, so for him, it's irrelevant.
Sjmarcy, I think it's unfair of you to cite a couple of top level drivers who go fast by getting brutal with the throttle, and forget about, just for example, drivers like Jenson Button, who go fast by being far smoother with the throttle. Granted, he gets hyperactive with the throttle every now and again (which kind of ruins my argument to be honest!) but most of the time, he is smoother on the throttle than Michael or Rubens, and yet just as fast, or faster.
For people who aren't familiar with the concept of throttle tapping/throttle schizophrenics/whatever, here's a Senna video that shows him treating the throttle like a hot potato. If you want to argue that Ayrton Senna would be faster by being smoother on the throttle, you're a braver man than I am.
Sjmarcy, I think it's unfair of you to cite a couple of top level drivers who go fast by getting brutal with the throttle, and forget about, just for example, drivers like Jenson Button, who go fast by being far smoother with the throttle. Granted, he gets hyperactive with the throttle every now and again (which kind of ruins my argument to be honest!) but most of the time, he is smoother on the throttle than Michael or Rubens, and yet just as fast, or faster.
For people who aren't familiar with the concept of throttle tapping/throttle schizophrenics/whatever, here's a Senna video that shows him treating the throttle like a hot potato. If you want to argue that Ayrton Senna would be faster by being smoother on the throttle, you're a braver man than I am.
#108
Former Vendor
Thread Starter
iTrader: (31)
Join Date: Nov 2006
Location: Sunnyvale, CA
Posts: 15,442
Total Cats: 2,104
One last thing, then I'm done for good.
Jason mentioned earlier that I had commented ~2 years ago that the 2871 needed to be driven differently, and I stand by that today. The requirement is that you apply a small amount of throttle directly after turning in - this is good practice in pretty much any application (N/A, S/C, or turbo), but it's especially crucial with a "big" turbo like the 2871.
The goal with the throttle application is to increase the shaft RPM. By increasing shaft RPM, you end up with a significantly more responsive car - if you're starting from 60 or 70,000rpm on the shaft, you have that potential airflow available to you immediately. What sjmarcy doesn't understand/refuses to admit is that the sudden snap from 0% TPS is the worst-case-scenario for turbo cars - not only do you have limited flow through the engine, but you have virtually no shaft RPM. Not only is it the worst-case scenario, but nobody ever uses it - even in the F1 examples he posted, you can see that MS starts from ~15% throttle, and then quickly ramps up to 100% throttle. (My charts show the same - this is quite flattering )
The difference between 0% and 15% is huge, huge, huge. The shaft RPM increase that you gain from that initial throttle application means that every subsequent change in throttle position will not only use the normal airflow, but the pre-selected shaft RPM as well. This builds on itself, too - as you add throttle, the shaft RPM stays up, making the turbo respond better to the next input, and so on. Anyone with a turbo car can see this for themselves - go out on the highway, select 4th gear at ~4500rpm, spin the turbo up to 3psi on the brakes, and then stomp the throttle and see how quickly you can obtain full boost. It's virtually instantaneous. This is what I aim to do in the turbo car - I get to the point where I'm applying ~120-150% of stock power to the rear wheels, which is usually all it can handle, right up until the track-out point, at which time I stomp the loud pedal, increase output to ~300% of stock, and start thinking about when the next upshift is.
My point with this thread was not to say that turbos have absolutely no lag, ever - that's obviously not true. My point was that in real-world conditions, with slower throttle applications and higher RPMs, the lag is all but eliminated, especially if you have the discipline to pre-apply a small amount of throttle to get the turbo wound up before it's time to pull the trigger on corner exit. With small turbos, it matters even less - with the 2554R I could be really sloppy with my throttle pickup and the manifold pressure would sweep right up with my right foot. I had logs of that setup, 0% TPS to 100% TPS in 1.0 seconds, and the boost would hit 15psi before TPS hit 100%.
Some will never believe this, no matter how much evidence they're given - that's a shame, since they are missing out on the enjoyment of a properly-tuned turbo Miata.
Jason mentioned earlier that I had commented ~2 years ago that the 2871 needed to be driven differently, and I stand by that today. The requirement is that you apply a small amount of throttle directly after turning in - this is good practice in pretty much any application (N/A, S/C, or turbo), but it's especially crucial with a "big" turbo like the 2871.
The goal with the throttle application is to increase the shaft RPM. By increasing shaft RPM, you end up with a significantly more responsive car - if you're starting from 60 or 70,000rpm on the shaft, you have that potential airflow available to you immediately. What sjmarcy doesn't understand/refuses to admit is that the sudden snap from 0% TPS is the worst-case-scenario for turbo cars - not only do you have limited flow through the engine, but you have virtually no shaft RPM. Not only is it the worst-case scenario, but nobody ever uses it - even in the F1 examples he posted, you can see that MS starts from ~15% throttle, and then quickly ramps up to 100% throttle. (My charts show the same - this is quite flattering )
The difference between 0% and 15% is huge, huge, huge. The shaft RPM increase that you gain from that initial throttle application means that every subsequent change in throttle position will not only use the normal airflow, but the pre-selected shaft RPM as well. This builds on itself, too - as you add throttle, the shaft RPM stays up, making the turbo respond better to the next input, and so on. Anyone with a turbo car can see this for themselves - go out on the highway, select 4th gear at ~4500rpm, spin the turbo up to 3psi on the brakes, and then stomp the throttle and see how quickly you can obtain full boost. It's virtually instantaneous. This is what I aim to do in the turbo car - I get to the point where I'm applying ~120-150% of stock power to the rear wheels, which is usually all it can handle, right up until the track-out point, at which time I stomp the loud pedal, increase output to ~300% of stock, and start thinking about when the next upshift is.
My point with this thread was not to say that turbos have absolutely no lag, ever - that's obviously not true. My point was that in real-world conditions, with slower throttle applications and higher RPMs, the lag is all but eliminated, especially if you have the discipline to pre-apply a small amount of throttle to get the turbo wound up before it's time to pull the trigger on corner exit. With small turbos, it matters even less - with the 2554R I could be really sloppy with my throttle pickup and the manifold pressure would sweep right up with my right foot. I had logs of that setup, 0% TPS to 100% TPS in 1.0 seconds, and the boost would hit 15psi before TPS hit 100%.
Some will never believe this, no matter how much evidence they're given - that's a shame, since they are missing out on the enjoyment of a properly-tuned turbo Miata.
#110
Here are a couple autocross runs with excellent PAX and event results. Throttle position versus time. Note that many corrections were needed with the throttle. Not shown are the steering and brake corrections also being made. Not much sluggish rolling on of the throttle. No extended time periods sitting there floored. Other than a 1-2 shortshift, these are all in 2nd gear.
#111
Elite Member
iTrader: (8)
Join Date: Dec 2008
Location: Kingston, Ontario
Posts: 2,910
Total Cats: 51
Also not shown is MAP.
I can sit in my driveway with the car off and make those logs.
Edit:
#1. the top log there looks like the driver is very confused and drives like a woman. (100% throttle or 0% throttle) PWM with foot on gas style driving.
#2. This isnt showing the sampling rate at a reasonable rate >10hz.
I can sit in my driveway with the car off and make those logs.
Edit:
#1. the top log there looks like the driver is very confused and drives like a woman. (100% throttle or 0% throttle) PWM with foot on gas style driving.
#2. This isnt showing the sampling rate at a reasonable rate >10hz.
#112
Also not shown is MAP.
I can sit in my driveway with the car off and make those logs.
Edit:
#1. the top log there looks like the driver is very confused and drives like a woman. (100% throttle or 0% throttle) PWM with foot on gas style driving.
#2. This isnt showing the sampling rate at a reasonable rate >10hz.
I can sit in my driveway with the car off and make those logs.
Edit:
#1. the top log there looks like the driver is very confused and drives like a woman. (100% throttle or 0% throttle) PWM with foot on gas style driving.
#2. This isnt showing the sampling rate at a reasonable rate >10hz.
Here is the same data from the upper trace to which you referred. I added a track map whose trace color corresponds to throttle position. MAF is also shown. I could convert to a YouTube if desired, the car goes CCW through the autocross course and all the traces track together.
Both traces shown were from winning autocross runs. They are logged at 100 Hz, not < 10Hz as you claim. They are not from a driveway. They are from an autocross held at an airport. Two runways plus a crossover connector were utilized.
Autocross is mostly about the driver controlling the car. Not some noob holding the gas pedal down - hey look at me I can floor it! Or Ahm betta than yall cuz mah motor eez bigger. Lol. It's about driving.
Last edited by sjmarcy; 08-26-2011 at 10:50 AM.
#113
Here is a different event using a mini-road-course arrangement. Same throttle activity (and steering and LFB not shown)…just look at the color of the traces carefully.
I overlaid and offset a portion of a faster run for comparison. I was having trouble with the exit of this turn (arrow) which was dusty, off camber, and downhill. Very hard to put down power. Lots of steering corrections. So I did a throttle kick earlier in the turn to cancel understeer to a degree. This allowed powerdown to start earlier and with a straighter steering wheel. This made around a 1/2 second improvement and was fun. See how much sooner the red (floored) trace begins?
I overlaid and offset a portion of a faster run for comparison. I was having trouble with the exit of this turn (arrow) which was dusty, off camber, and downhill. Very hard to put down power. Lots of steering corrections. So I did a throttle kick earlier in the turn to cancel understeer to a degree. This allowed powerdown to start earlier and with a straighter steering wheel. This made around a 1/2 second improvement and was fun. See how much sooner the red (floored) trace begins?
Last edited by sjmarcy; 08-26-2011 at 11:01 AM.
#114
Sav is a good driver. For Sav's particular driving style (which, being his thread, is what's most relevant) his car responds absolutely fine. No, it won't respond as cleanly as a decent non-turbo car if you're snapping the throttle on and off, but that's not what Sav does, so for him, it's irrelevant.
Sjmarcy, I think it's unfair of you to cite a couple of top level drivers who go fast by getting brutal with the throttle, and forget about, just for example, drivers like Jenson Button, who go fast by being far smoother with the throttle. Granted, he gets hyperactive with the throttle every now and again (which kind of ruins my argument to be honest!) but most of the time, he is smoother on the throttle than Michael or Rubens, and yet just as fast, or faster.
For people who aren't familiar with the concept of throttle tapping/throttle schizophrenics/whatever, here's a Senna video that shows him treating the throttle like a hot potato. If you want to argue that Ayrton Senna would be faster by being smoother on the throttle, you're a braver man than I am.
Sjmarcy, I think it's unfair of you to cite a couple of top level drivers who go fast by getting brutal with the throttle, and forget about, just for example, drivers like Jenson Button, who go fast by being far smoother with the throttle. Granted, he gets hyperactive with the throttle every now and again (which kind of ruins my argument to be honest!) but most of the time, he is smoother on the throttle than Michael or Rubens, and yet just as fast, or faster.
For people who aren't familiar with the concept of throttle tapping/throttle schizophrenics/whatever, here's a Senna video that shows him treating the throttle like a hot potato. If you want to argue that Ayrton Senna would be faster by being smoother on the throttle, you're a braver man than I am.
There were times back in the 80s turbo era where they had amazing horsepower. And some of the low budget teams had normally aspirated engines which could be larger to try to even things out a bit.
In the rain there were a number of times Senna (in a boosted car) could not deal with the normally aspirated cars with their sharper throttle control. So "nobodies" were running with him at times. He had to deal with the lag and soft/inconsistent response and even he could not make up for it.
If you are ever in a tightish turn with a car that enjoys/is cursed with trailing throttle oversteer then try out the senna throttle chops. It works wonders on many German cars. Save 1/4 second per turn all day long by doing that. Technically it is "wrong" but the clock says otherwise. I did it a lot on my old E30 M3.
I think that Senna developed the blipping from his karting days. The tracks down there are extremely bumpy so they also do it while braking. So that they have brakes on when there is traction and then off for an instant while they have none and then back on. You can see this in some karting viddies. When I watch karting I make a point out of observing the foot actions.
#116
I'd make the point that Sav was the troll here. He started a thread with a provocative title. And proceeded from there. I did not start this thread. I have not called anyone names. Such as "whiny" or "idiot" or "troll" or accusing people of unethical things or not being able to drive, etc.
I just respond with data/information and my experience which I believe is also valid. So far I have not been semi-banned or e-blasted due to my responses. I have not noticed anyone editing and changing my posts besides me. I try to stay civil and factual. So I have that going for me. Maybe this forum is improving? And discussions are actually possible?
If this site is to grow and produce increased and more stable income, a mature perspective might help with that possible goal. Then content quality will rise. And folks can improve their Miatas.
In the mean time, those developing turbo cars might want to bear in mind some of my points. They will speed you up! Seen it a million times over the years.
I just respond with data/information and my experience which I believe is also valid. So far I have not been semi-banned or e-blasted due to my responses. I have not noticed anyone editing and changing my posts besides me. I try to stay civil and factual. So I have that going for me. Maybe this forum is improving? And discussions are actually possible?
If this site is to grow and produce increased and more stable income, a mature perspective might help with that possible goal. Then content quality will rise. And folks can improve their Miatas.
In the mean time, those developing turbo cars might want to bear in mind some of my points. They will speed you up! Seen it a million times over the years.
#117
In F1 you can see the teams that are not quite up to speed making more mistakes. This is because they have to be right at 10.000 tenths all the time to try to make up the gap...which is hard to do all day long.
I'll see if I can find a nice example from a 911 qually lap at Spa. Excellent driving! You can see the corrections to steering and see/hear the throttle action. This driver says that if the car is set up for more understeer, then it becomes boring and slow. You might agree with me that he seems to be at the limit in many of the turns.
Here is a Spec Miata running pretty fast. These are grippy no-power racecars in case you are not familiar with them. And the spec setup is pretty understeery. Despite this, you can see lots of steering and throttle and brake inputs used to control the car to good effect.
Last edited by sjmarcy; 08-26-2011 at 11:54 AM.
#118
Moderator
iTrader: (12)
Join Date: Nov 2008
Location: Tampa, Florida
Posts: 21,026
Total Cats: 3,123
To my understanding, a good race driver is never at 0% throttle mid-corner but applying enough throttle to maintain his speed and not continue to decelerate. A good driver will also be in his powerband at that point to allow for ready acceleration out of the corner. In a turbo car that would mean that it was spooled up, running at higher rpms with a partially opened throttle.
The problem with marcy's argument is that to stab the throttle on corner exit without spinning the car out means one of two things. Either the car is being driven poorly and is not actually cornering at its limits of adhesion, or you are driving such a gutless wonder that there is almost no difference between WOT and closed throttle. The argument leads me to believe that marcy has never driven a car with a decent amount of power at or near its limits in a controlled environment.
The fact that stabbing the throttle is a bad idea for cornering at the edge of adhesion in a car that makes any kind of power makes the argument moot.
Cliffs: Driving like a spaz is bad. Therefore we don't need to know why driving like a spaz is better without a turbo.
The problem with marcy's argument is that to stab the throttle on corner exit without spinning the car out means one of two things. Either the car is being driven poorly and is not actually cornering at its limits of adhesion, or you are driving such a gutless wonder that there is almost no difference between WOT and closed throttle. The argument leads me to believe that marcy has never driven a car with a decent amount of power at or near its limits in a controlled environment.
The fact that stabbing the throttle is a bad idea for cornering at the edge of adhesion in a car that makes any kind of power makes the argument moot.
Cliffs: Driving like a spaz is bad. Therefore we don't need to know why driving like a spaz is better without a turbo.
#119
Moderator
iTrader: (12)
Join Date: Nov 2008
Location: Tampa, Florida
Posts: 21,026
Total Cats: 3,123
Just look up data traces on ANY racecar driven fast. And you will find many rapid driver inputs at key corners. Then look at the slower drivers, and you'll find fewer corrections. The closer you get to the true limit, the harder it gets and the more little corrections you will see.
I'll see if I can find a nice example from a 911 qually lap at Spa. Excellent driving!
I'll see if I can find a nice example from a 911 qually lap at Spa. Excellent driving!
Here's an example.
#120
To my understanding, a good race driver is never at 0% throttle mid-corner but applying enough throttle to maintain his speed and not continue to decelerate. A good driver will also be in his powerband at that point to allow for ready acceleration out of the corner. In a turbo car that would mean that it was spooled up, running at higher rpms with a partially opened throttle.
The problem with marcy's argument is that to stab the throttle on corner exit without spinning the car out means one of two things. Either the car is being driven poorly and is not actually cornering at its limits of adhesion, or you are driving such a gutless wonder that there is almost no difference between WOT and closed throttle. The argument leads me to believe that marcy has never driven a car with a decent amount of power at or near its limits in a controlled environment.The fact that stabbing the throttle is a bad idea for cornering at the edge of adhesion in a car that makes any kind of power makes the argument moot.
Cliffs: Driving like a spaz is bad. Therefore we don't need to know why driving like a spaz is better without a turbo.
The problem with marcy's argument is that to stab the throttle on corner exit without spinning the car out means one of two things. Either the car is being driven poorly and is not actually cornering at its limits of adhesion, or you are driving such a gutless wonder that there is almost no difference between WOT and closed throttle. The argument leads me to believe that marcy has never driven a car with a decent amount of power at or near its limits in a controlled environment.The fact that stabbing the throttle is a bad idea for cornering at the edge of adhesion in a car that makes any kind of power makes the argument moot.
Cliffs: Driving like a spaz is bad. Therefore we don't need to know why driving like a spaz is better without a turbo.