R/C Tech Forums

Go Back   R/C Tech Forums > General Forums > Electric On-Road

Reply
 
Thread Tools Search this Thread
Old 03-14-2009, 11:11 AM   #16
Tech Lord
 
wingracer's Avatar
 
Join Date: Nov 2007
Posts: 12,861
Trader Rating: 24 (100%+)
Default

Quote:
Originally Posted by SystemTheory View Post
But if racing without thinking about setup/theory is what you enjoy, go for it!
Where did I say anything about not working on set-up? I probably spend more time and energy on set-up than anyone that isn't paid to do this for a living.

All I'm trying to say is that I suspect if you spent a lot of time optimizing your car to pull the maximum amount of g's on a skid pad you would probably end up with a poor handling car in actual race conditions. I could be wrong, I've never done skid pad testing.

Here are some of my reasons for thinking this way:

1. Skid pad will tell you nothing about transitional handling. It only deals with steady state cornering forces. As an example, very high horsepower cars are usually set-up too tight (understeer) to be able to get good skid pad numbers. They require a big reserve of rear traction to be able to accelerate off the corner. This is why many F1 drivers are always complaining about understeer. This could easily be solved by reducing the rear wing but the loss of forward bite off the corner results in slower lap times despite the improved balance. A lower horsepower car on a very tight track might require a much looser (oversteer) set-up in order to get around those hairpins.

2. Especially with foam tire cars, pulling more g's is easy. Just slap a softer set of tires on the car. This will not always result in lower lap times though. You run into problems with traction rolling or, especially with stock 1/12th scales, bind the car up too much in the fast corners and scrub speed. Sometimes a harder tire that reduces skid pad g's will also reduce lap times.

3. Since we are talking ozite here, how many sets of tires and gallons of Paragon is it going to take before the skid pad carpet matches that of most tracks?

4. Cars often handle differently in fast corners versus slow corners or whether you are off power or on power.


The gist of all this is that race car set-up is a compromise. In order to gain performance in one area often reduces performance in another. The trick is finding which areas need to be optimized and which can be allowed to suffer in order to gain the bast lap time with a particular combination of car, driver and track.

Again, I'm not saying skid pad testing is useless, it very well may provide usable information but trying to set-up a car strictly for pulling the best skid pad numbers will not work. It's just like engines and dynos. A motor set-up to turn the absolute highest hp number will usually not be the quickest on the racetrack. Yet, kept in perspective and used as just a tool and not the end all be all of performance you can still get good info from a dyno.
__________________
Sean. Certified speed crazed mowron.
Team Shepherd USA
www.ashfordhobby.com
wingracer is offline   Reply With Quote
Old 03-14-2009, 11:12 AM   #17
Tech Lord
 
wingracer's Avatar
 
Join Date: Nov 2007
Posts: 12,861
Trader Rating: 24 (100%+)
Default

Quote:
Originally Posted by SystemTheory View Post
Trail braking a dump truck will not win a formula race.
No, it won't win an F1 race but it just might win a dump truck race.
__________________
Sean. Certified speed crazed mowron.
Team Shepherd USA
www.ashfordhobby.com
wingracer is offline   Reply With Quote
Old 03-14-2009, 12:10 PM   #18
Tech Champion
 
robk's Avatar
 
Join Date: Aug 2001
Location: Macho Business Donkey Wrestler
Posts: 7,586
Trader Rating: 22 (100%+)
Default

Wingracer has posted the Correct.
__________________
A mutually re-enforcing cascade of failure

"Failior [sic] crowns enterprise." Robert Goddard

I-Lap Scoring Systems http://www.rclapcounter.com/
robk is offline   Reply With Quote
Old 03-14-2009, 01:06 PM   #19
Tech Adept
 
Join Date: Mar 2009
Location: Metro New York
Posts: 139
Default

As they say, "The Devil is in the details." I can't dispute wingracer's experience, so I'll just point out that no simulator engine can handle all of those details, and like everything else, skid pad testing is good for what it's good for.

My simple question is this. In a turn of radius R and sweep angle A, you must cover a given distance S. Can you go faster along curve S while pulling fewer g's ? The skid pad helps you learn this lesson in the simplest study, that's all I meant to convey.
SystemTheory is offline   Reply With Quote
Old 03-14-2009, 03:16 PM   #20
Tech Lord
 
wingracer's Avatar
 
Join Date: Nov 2007
Posts: 12,861
Trader Rating: 24 (100%+)
Default

Quote:
Originally Posted by SystemTheory View Post
My simple question is this. In a turn of radius R and sweep angle A, you must cover a given distance S. Can you go faster along curve S while pulling fewer g's ? The skid pad helps you learn this lesson in the simplest study, that's all I meant to convey.
In some corners, yes you can. If you look at telemetry data from real race cars, they actually spend very little time at the maximum lateral acceleration. The tires can only have so much traction and some or most of it is either used for braking entering the corner or for acceleration off the corner. Typically you will see only a split second of max lateral acceleration right at the apex. Let's say that max acceleration is 1g. A good driver (let's say it's Schumacher because I have seen his data in a magazine before and he really is a master) will stand on the brakes near the end of the straight using all his traction for braking. As he begins to turn into the corner he will ease off the brake to have the traction needed for cornering but does NOT jump all the way off the brakes. He gradually releases the brakes while gradually increasing steering input so that linear acceleration reduces while lateral acceleration increases. He wont hit that full 1g lateral until right at the apex as he finally comes all the way off of the brake. This last for a tiny fraction of a second as he then starts to reduce steering input and adding throttle. Now as the lateral acceleration drops, the linear acceleration increases.

So what does this mean for achieving max g's? Well, let's say this car Herr Schumacher has been driving is really well balanced and he is very happy with the car. Now some engineer comes along and says "this spring rate will improve grip." So they make a change and go out again. This time the max lateral g seen for that split second in the middle of a corner is 1.1g. This is better right? Except the laptimes are slower and Schumacher is griping that the car is junk. How can this be?

Well, we did increase grip but in the process we made the car very unstable under braking so he can't go into the corner as hard. The extra lateral grip allowed him to let off the brake sooner, thus increasing the max g number at apex but having to brake so much sooner and so much more carefully has hurt the straightaway speed and laptimes.

Or, maybe this gave the car a bad throttle push exiting the corner. He wasn't able to stand on the gas as hard on exit and therefore straightaway speed suffered.

Of course, now that we know the car is capable of more grip, perhaps we can play with some other things to improve the transitions while maintaining that grip.

My point is, skid pad data MAY provide some useful data but you can NOT set up a car on skid pad data alone and expect good results. For instance, take a car with a 500lbs, 500 hp engine and run it on a skid pad. Now replace the motor with a 400 lbs, 200 hp engine. The reduced weight is going to improve the skid pad numbers but laptimes will be much slower on a real racetrack.

One RC class where I could see skid pad testing being useful is spec rubber tire stock TC's. Their low power means they spend more time at a more or less steady state, grip can often be hard to find and you don't have the option of slapping a softer set of tires on it. If you could find some grip on the skid pad without having to resort to any unpractical set-ups you just might find some laptime improvement.
__________________
Sean. Certified speed crazed mowron.
Team Shepherd USA
www.ashfordhobby.com
wingracer is offline   Reply With Quote
Old 03-14-2009, 04:03 PM   #21
Tech Adept
 
Join Date: Mar 2009
Location: Metro New York
Posts: 139
Default

Einstein was my hero growing up (I was/am a strange kid). He once said, "I wish to understand the mind of God, the rest is just details." Another Physicist shot back (wish I could remember who it was), "God is in the details." I bet Einstein smiled at that one!

I would argue that the average g's through the corner of displacement S must go up when the elapsed time goes down, or else you violate some law of Physics. On a dirt car drifting through a high speed corner, kicking up a roost or spinning the tires, it could be a steering vector, rather than pure grip, but one must get centripital force to pull the mass m around that turn, and thus get around the corner in a lower elapsed time, I am fairly certain.

I agree with you 100% that track tuning involves many trade-offs, and even running a Monte-Carlo analysis, it is almost impossible to predict the optimum line on a race track using hours of computer time.

So God is your details, as far as I'm concerned. I hope others here recognize this and benefit from your knowledge.
SystemTheory is offline   Reply With Quote
Old 03-14-2009, 04:45 PM   #22
Tech Adept
 
Join Date: Jun 2008
Location: Minneapolis, MN
Posts: 238
Trader Rating: 10 (100%+)
Default

It appears we have a physicist arguing with an engineer. Either way, great discussion.
theghostmouse is offline   Reply With Quote
Old 03-14-2009, 04:49 PM   #23
Tech Regular
 
Parnelli's Avatar
 
Join Date: Nov 2008
Location: Minnesota
Posts: 341
Trader Rating: 13 (100%+)
Default Yada yada yada

Well so much for thinking out loud. I agree the best way to optimize my VTA car's set-up and learn how to drive it fast is frequent visits to a permanent carpeted race track. I do not have that option more than once per month unless I move to Madison Wisconsin. As much as I like the Badger state (Road America, Lake Michigan, Peter Egan and Friday night fish fry's) this is probably not happening in the next 10 years.

Based on the discussion so far It appears that many areas of 1/1 scale sports car racing do not translate to 1/10 scale RC racing. I can accept that, my goal is continuous learning and incremental improvement. For example tomorrow I will spend the time and money to drive 2.5 hours each way to race 1 class on carpet in Superior so I am / can be committed. As Michael Delaney said almost 40 years ago "everthing else is just waiting"
__________________
Parnelli

Drives like a Washburn monkey
Parnelli is offline   Reply With Quote
Old 03-14-2009, 08:37 PM   #24
Tech Lord
 
wingracer's Avatar
 
Join Date: Nov 2007
Posts: 12,861
Trader Rating: 24 (100%+)
Default

Einstein is a hero of mine as well and the devil is in the details, so why try to gauge the performance potential of a car on just one detail?

Race engineers often come up with ideas that on paper should be faster. Sometimes those ideas work, other times they result in a car that may be capable of faster lap times but no driver can drive the thing.

Different types of cars require different types of set-ups. Grossly overpowered cars (Can-Am of the 60's and 70's comes to mind) are more about linear accelerations. They typically ran less camber to improve braking and acceleration. They were all about slowing down and speeding up as fast as possible. This would be like 1/8th or maybe 1/12th mod on a tight, slick asphalt track. Lower powered cars like formula fords are all about corner speed (lateral accelerations). They don't have the power to accelerate quickly so maintaining corner speed is a must. This would resemble 1/12th stock and stock TC's and may benefit more from your skid pad test's.

Take a look at Leman's style racing. The LMP-2 cars are much lighter than the big LMP-1 cars and pull more g's in the corner. They can sometimes compete for overall wins on tight, twisty tracks but most of the time the brute horsepower of an LMP-1 car wins out.
__________________
Sean. Certified speed crazed mowron.
Team Shepherd USA
www.ashfordhobby.com
wingracer is offline   Reply With Quote
Old 03-14-2009, 09:36 PM   #25
Tech Elite
 
Join Date: Aug 2002
Location: Houston, Texas
Posts: 3,777
Trader Rating: 27 (100%+)
Send a message via AIM to John Stranahan
Default

A lot of tuning that you do on the track is to tune the transition of the car. By this I mean you enter a corner unrolled or flat. At the apex your car has reached maximum g's and is fully rolled over. How your car behaves in this interim is very important and the subject of most of your tuning. On the skid pad we have a steady state. The car is always fully rolled. It is at maximum g's. It could very well perform the best here with setting that are poor in the transitions. Lap times on the road course could go down.

Instead Make a figure 8 with cones or better yet arcs of some smooth barrier material. Practice going close to the barrier, accelerating in the middle. This is 95% of your performance. Being able to slow properly, approach the barrier close without touching it, and then accelerating out without tire spin. The other 5% is setup. This assumes the car is not ill behaved. Usually the ill traits are gotten out easily and the next 5% with some difficulty. Being able to approach the apex at speed slow down and not hit but pass close to the barrier is the key.

A stop watch is used with anticipation. Accuracy is much much better than .2 s. You watch the car approaching a mark you anticipate the exact time. Your finger is down on the button at the instant the car touches the mark. There is error, but not even .1 s with practice.

On any video game that I have tried the physics is so poor that they are virtually useless for gaining any RC car skill. My Nephew says in response that that is the Physics that they would like to have.

16 x 20 feet on the driveway is almost big enough for some 1/10 scale practice. Oval is OK too. You have straights and then turns to deal with.

John
John Stranahan is offline   Reply With Quote
Old 03-14-2009, 10:07 PM   #26
Tech Lord
 
wingracer's Avatar
 
Join Date: Nov 2007
Posts: 12,861
Trader Rating: 24 (100%+)
Default

Quote:
Originally Posted by John Stranahan View Post
Instead Make a figure 8 with cones or better yet arcs of some smooth barrier material. Practice going close to the barrier, accelerating in the middle.
That's a damn good idea. Wish I had thought of it.
__________________
Sean. Certified speed crazed mowron.
Team Shepherd USA
www.ashfordhobby.com
wingracer is offline   Reply With Quote
Old 03-14-2009, 10:28 PM   #27
Tech Adept
 
Join Date: Mar 2009
Location: Metro New York
Posts: 139
Default

Thanks, Yoda (John)! Now I'll put down my lightsaber.

May be of note. In my stopwatch experiment, I designed a laser trigger for a $4.99 Walmart device. Two or three of these units would all log no less than 0.06 seconds when connected to a faster trigger signal, which I built into a 555 timer circuit, although the watch reads to 0.01 precision.

I tested people making two motions, start and stop the watch as fast as possible. Most of the times were 0.25, my best was 0.16, and my little brother, the race car driver, could average under 0.2 easily with practice.

I figure with a standardized ramp of proper length, a laser trap coupled to a 0.001s precision timer top and bottom ($30-50 bucks at Digikey), and a proper set of differential equations, I could build a "Chassis Dyno," thus getting data on the equivalent driveline inertia Jeq and driveline damping Deq from a roll down test with the pinion removed from an electric car. I figure some cars are just faster in the straight due to a better driveline.

I'm bad at building stuff, much prefer theory and troubleshooting, and the Chassis Dyno project bit the dust.

Last edited by SystemTheory; 03-14-2009 at 10:43 PM.
SystemTheory is offline   Reply With Quote
Old 03-15-2009, 05:28 AM   #28
Tech Regular
 
Parnelli's Avatar
 
Join Date: Nov 2008
Location: Minnesota
Posts: 341
Trader Rating: 13 (100%+)
Thumbs up Yes!

Thanks John for the enhanced clarity, now I am glad I posted this question. A small figure 8 test track makes a lot of sense in that I can run it both directions and get lots of transition feedback and raw speed data. I have a lap timing feature on my Spektrum transmitter, will that be adequate?
__________________
Parnelli

Drives like a Washburn monkey
Parnelli is offline   Reply With Quote
Old 03-15-2009, 09:39 AM   #29
Tech Elite
 
Join Date: Aug 2002
Location: Houston, Texas
Posts: 3,777
Trader Rating: 27 (100%+)
Send a message via AIM to John Stranahan
Default

Certainly a Spektrum lap timing feature is adequate. Another common radio feature is a repeated timed beep. Set it right at your lap times. Don't get too organized here as the course really is small depending on what power plant your run.

I'll touch on the stop watch again. Certainly reaction time is on the order of .2 -.25 seconds. If we are using the watch to get a lap time, there is no reaction time to deal with. You anticipate when the car is going to cross a mark. You can maybe see your self, arm swinging down in anticipation finger on the button, button gets pushed right at the time the car crosses the line. Probably only .005 s error. You would set up a completely different experiment to test this. The experiment would not be like the Pro tree at the dragstrip where you get a single light to start. More like the standard sequential light tree used for the lower classes.
john

Last edited by John Stranahan; 03-15-2009 at 10:34 AM.
John Stranahan is offline   Reply With Quote
Old 03-15-2009, 09:42 AM   #30
Tech Lord
 
wingracer's Avatar
 
Join Date: Nov 2007
Posts: 12,861
Trader Rating: 24 (100%+)
Default

When I was a kid my dad and I would often go to races and I would bring my stopwatch. I loved to time qualifying to see if my stopwatch skills were as good as the tracks timing system. I was almost always with .01 of the announced time.
__________________
Sean. Certified speed crazed mowron.
Team Shepherd USA
www.ashfordhobby.com
wingracer is offline   Reply With Quote
Reply



Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)
 
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are On
Pingbacks are On
Refbacks are Off



All times are GMT -7. It is currently 12:05 AM.


Powered By: vBulletin v3.9.2.1
Privacy Policy | Terms of Use | Advertise Content © 2001-2011 RCTech.net