Causes for Understeer???

[MrAWD]

The reason a softer suspension is good for teaching this is because it is not very accepting of the mid corner corrections that new drivers often make when the look up halfway through a turn and notice they offline. I can not count how many times I have been driving through a corner with an inappropriate entry, tried to correct, and had the car just flop around but not change direction. Had I been driving a stiffer, more responsive car I would have been able to make that mid corner correction, then thought I did a good job because I got through the corner ok, but probably not realized that I could have been much faster if I set up for the corner properly in the first place.

This is another of the common misconceptions about the better suspension or better car per say. If you have two cars, one stock (let say DS) and other one nicely modified (fits in the ESP or SM). Better suspension on the car two will usually give you better times on the typical autoX course. You should be able to do more with a better prepared car. But, one thing either car can not do and that is to recover your time after you make a mistake. No matter how better the car is prepared then the stock one, once you find your self in the middle of the turn two or three feet off your line, the time is lost and there is nothing that car or you can do to make up for the time lost.

Another thing is that if either of two cars allows you to make such a correction in the middle of the turn, which means that you didn’t get in there fast enough to start with. At the edge, you can do only small amount of corrections and change your car from where it is to where you want it to be. If you are off by more then two or three feet, and you still can make that correction, you are still loosing time overall and that can not be your fastest time that you could do that day.

Basically, don’t expect your suspension to cover for your mistakes and which is even worse blame it for the bad run!


Fedja

 

[jtmcinder]

Instructors often have to tell their students to slow down their hands and feet. They rarely -- if ever -- have to tell them to speed up their hands and feet. This applies to all of their students, no matter what car they drive or how it has been prepared.

Heavy, huh?

- Jtoby

 

[Revolution]

No, it is never good to understeer. Period! If your car understeers, that will mean just longer time for it to get into the position where it needs to be. Also, if you need your car to understeer to get into the correct position, that implies that you already made a mistake and turned your steering wheel too much too early.

The worst one of all is that it will delay the point in time where you can go full throttle.



Fedja

I can understand where you are coming from but in my defence I would rather understeer in a chicane (some section) of the course rather then deel with over exsesive over steer snap.

After driving for a little bit you sart to look at driver preferences when definning a drivers syle. I would never understeer on purpose but after braking according to the turn I would rather understeer slightly then over steer abruptly.

Fedja you have probably been auto xing alot more then I have so you at times will have a better understanding of a situation,as would jtmcinder and dg-fnr ,but i still stand by my understeering statement . How can you expect some one to properly use an over steering technique when they cant minimize their corner entry understeer.

To fix understeer is a tap of the brakes away ,this could be applied in many forms and ways but in the end whether the car be softer or harder the driver will still have to overcome the braking delema conserning understeer.

 

[MrAWD]

I can understand where you are coming from but in my defence I would rather understeer in a chicane (some section) of the course rather then deel with over exsesive over steer snap.

I don’t know why you using excessive oversteer as a defense for the understeer. They are both tuning problems and need to be corrected. And, it is much easier to correct understeer then oversteer.

After driving for a little bit you sart to look at driver preferences when definning a drivers syle. I would never understeer on purpose but after braking according to the turn I would rather understeer slightly then over steer abruptly.

Hmm, it is natural tendency for any car to oversteer when braking and turning. There is hardly anything one can do there to defeat laws of physics. The good thing here is that this usually helps and people do use it to help rotation of the car. And if you don’t like it, finish your braking before the turn, so you don’t have that problem any more.

Fedja you have probably been auto xing alot more then I have so you at times will have a better understanding of a situation,as would jtmcinder and dg-fnr ,but i still stand by my understeering statement . How can you expect some one to properly use an over steering technique when they cant minimize their corner entry understeer.

Two quite different behaviors and I don’t know why you linked them together. As I said earlier, brake before the turn and then drive flat out through the turn!!

To fix understeer is a tap of the brakes away ,this could be applied in many forms and ways but in the end whether the car be softer or harder the driver will still have to overcome the braking delema conserning understeer.

Any tap of brakes means not accelerating as fast as you could through out the turn! And again, you have me confused with bringing that corner entry into this discussion


Fedja

 

[DG-FNR]

Well... the car will be a good deal faster with some oversteer in it.

To clarify, given that power-on oversteer is an effective impossibility, you want some trailing-throttle oversteer to help rotate the car. You wind up with a setup where lift-throttle steps the tail out, and you catch the car with the throttle. You can then throttle-steer the car.

The master of this was GH Sharp back when he was driving a Boxster. He'd agressively turn in, get the tail moving (not sliding, just moving) and then when the car was pointed where he wanted it to go, roll on the throttle to plant the rear. This was a technique I incorporated into my bag of tricks, and when the car and the course let you use it, it is VERY fast.

Where this becomes less useful is at high speed. As a rule of thumb, the faster the peak speed you'll see, the less oversteer the setup can stand. Autocross cars are setup with buckets of oversteer, but an IRL car on a super speedways can stand NO oversteer, and are always set up with varying degrees of understeer.

It takes time, however, to learn to drive a car with a lot of oversteer, and different talent levels are capable of working with different setups. I like a lot of oversteer, but there's a definate line over which the car becomes just too twitchy to deal with and too difficult to push hard. A lot of my setup work over the past couple of years has been playing with the amount of oversteer I can make use of before I run out of talent. Any more or any less than that magic amount, and I will be slow.

What I absolutely cannot stand, however, is understeer. I have dialed in less oversteer and gone faster, but not once, on any car, have I ever moved a car from oversteer to understeer and gone faster.

Understeer makes you work harder. I was at a test session, and this kid was there in an STX WRX. There was a light rain going on, and this poor kid was really struggling. He asked me to try out his car and tell me what I thought.

Well this car had the worst understeer I had ever experienced. We blew the first corner completely, when I braked about a car too late. The second corner I braked at the right point, but plowed through it when I tried to carry a little more speed through the turn than the car would let me get away with. All subsequent turns were OK, and I did another lap to confirm my impressions - and it was some huge amount faster than what the kid had done.

We get out of the car, and his jaw is down around his ankles. "You were turning like 2 car lengths before the cones!" Well, yeah, because it takes that long for the car to react, so you need to think ahead, account for the push, and drive accordingly.

However, the course Gods aren't always going to be so nice to you. You may encounter elements that are closer together than can be anticipated, and then your only choice is to slow down. That works - but it is SLOW.

DG

 

[jtmcinder]

To clarify, given that power-on oversteer is an effective impossibility...

Hey, Charles, remind me again what your car was like with the Cusco center before you put the VC back in.

- Jtoby

 

[ACM]

Hey, Charles, remind me again what your car was like with the Cusco center before you put the VC back in.

- Jtoby

Let's just say the oversteer reminded my of a pi$$ed-off pit bull

For different reasons though...

Actually my car is set up for power-on oversteer - again, finally. I had the car like this 2 or 3 seasons back, lost it as the car "developed", and now I'm getting it back again. Somewhat unpredictably, due to the misbehaviour of the VC and rear (un)LSD last year, but I think I'm back on track again. The Cusco has improved the front grip significantly.

 

[Revolution]

I don’t know why you using excessive oversteer as a defense for the understeer. They are both tuning problems and need to be corrected. And, it is much easier to correct understeer then oversteer.

Hmm, it is natural tendency for any car to oversteer when braking and turning. There is hardly anything one can do there to defeat laws of physics. The good thing here is that this usually helps and people do use it to help rotation of the car. And if you don’t like it, finish your braking before the turn, so you don’t have that problem any more.

Two quite different behaviors and I don’t know why you linked them together. As I said earlier, brake before the turn and then drive flat out through the turn!!


Any tap of brakes means not accelerating as fast as you could through out the turn! And again, you have me confused with bringing that corner entry into this discussion


Fedja

Fedja I totally agree with you in more ways then one , to much over steer and too much understeer is bad . In my previous statements I was trying to explain that in my own personal experiences when trail braking into a corner ,yes the car would begin to rotate depending on my steering inputs, but never has my car ever snapped into excessive over steer unless I was going way to fast or just broke way, way to early . Hence my understeer coments at corner entry.

When my car would begin to rotate, never was the rear leading the way but more like it was subsequently sliding sideways into the next barrier . I would then point the steering wheel in the direction I wanted to go and just progressivly added gas.

To me I was still understeering the car ,even though it did rotate, I still had to my nose diving into the corner. I will take your advice and brake sooner ,maybe about a half a car sooner to and see what the outcome would be.

But to be honest with everyone following this thread , I would and do jerk back and forth the steering wheel becasue my car feels like it has some sort of retard on the actions I introduce.

It's frustrating for me because if I would want a crispier feel im going to have to look into modifieing the car again which will cost an arm and a leg. It's not as easy learning to drive a car that's sloppy at times because I could be doing 4 things right and 6 things wrong ,and still not know the difference. If I were to try and better myself as a driver then racing school would cost me just as much.

So I think driving slow and progressivly adding speed would help out alot of stokers or mildly modded cars in auto x . Sometimes we just think about the win instead focusing on nurchering and development.

 

[bjones18]

Fedja I totally agree with you in .....But to be honest with everyone following this thread , I would and do jerk back and forth the steering wheel becasue my car feels like it has some sort of retard on the actions I introduce.

It's frustrating for me because if I would want a crispier feel im going to have to look into modifieing the car........sometimes we just think about the win instead focusing on nurchering and development.

My 2g does not have "steering torque feedback" proportional to the vehicle response. I am assuming this is "standard" for Eclipse/Talons....very common. Very few vehicles have this "well tuned" steering torque response (BMW 3, 5, Lincoln LS are a few which do have good steering torque feedback). Steering torque feedback can be a drivers first clue to what is happening at the contact patch. It is NOT easy to accomplish.

An easy first step to improve this is tire selection. Tires with a continuous solid-center circumferential-band (no siping on center band; and NOT Aqua-tread or Kuhmo style) is a great first step toward improving steering feel. Other tire properties to improve feel are not as easily recognized; leading to back to back testing to find what you are looking for.

 

[n2o4fun]

Somewhat unpredictably, due to the misbehaviour of the VC and rear (un)LSD last year, but I think I'm back on track again. The Cusco has improved the front grip significantly.

Cusco Tarmac center correct?

 

[ACM]

Yes, the Cusco Tarmac.

Charles

 

[jtmcinder]

Is anyone ever going to make a Torsen Type-3 for our center? Or does our layout (with a ring-in and concentric-out) make this impossible? And what's in the center of the new Evo?

- Jtoby

 

[MrAWD]

And what's in the center of the new Evo?

EVO has a clutch type controlled with hydraulics…pressure variation changes the locking amount from fully locked at 50:50 on one end and all the way to fully opened on the other...

Fedja

 

[jtmcinder]

Cool. Thanks. But when it's open, what's the split? Did they manage to get a planetary in there or is it still 50/50?

- Jtoby

 

[jtmcinder]

I was googling around, looking for more information, and I found a site that is a perfect example of internet idiocy: http://www.superstreetonline.com/featuredvehicles/53460/

Yes, maybe it goes without saying that it's a magazine for Honda owners, but you got to love it anyway.

The amusment peaks when you look at the captions for the pictures down the right side.

- Jtoby

 

[MrAWD]

Cool. Thanks. But when it's open, what's the split? Did they manage to get a planetary in there or is it still 50/50?

It is 50:50 at full lock!

Fedja

 

[MrAWD]

The amusment peaks when you look at the captions for the pictures down the right side.

Good find!!


Fedja

 

[MrAWD]

I have only a link of type "how it works?" for the AYC

Sorry...



Fedja

 

[jtmcinder]

It is 50:50 at full lock!

I don't think that you're understanding my question. All diffs are 50/50 when locked, even the STi 35/65 ACD. My question is whether the Evo's center is 50/50 when open or whether it has a rear bias like an STi when open.

- Jtoby

 

[MrAWD]

I don't think that you're understanding my question. All diffs are 50/50 when locked, even the STi 35/65 ACD.

Are you sure about this? As far as I know, fully locked STi has some noises when turning and I believe that comes from the 35/65 lock. Unless it never goes to the full lock mode, I think it is rather that and not 50:50.

My question is whether the Evo's center is 50/50 when open or whether it has a rear bias like an STi when open.

From what I know so far about the ACD, it acts just like an open diff (which is 50:50 with no slip) and let say adjustable viscous coupler next to it. When fully opened, it is the same as any other open diff. With 50% locking it would allow 25-75% slip if pushed that way! And that would go either way! The thing is that 60:40 weight distribution will always brake the rear end before the front one and that is why it ends up rear end happy. Also, diff can be done that way to always sends more to the rear and let the front to catch up a bit later, kind of what we have in our stock center diff.


Fedja
Ps. Of course the numbers that I used are just for the concept proof purpose, and not the actual data.

 

[jtmcinder]

I am absolutely sure that an STi center is 50/50 when locked because it is a coaxial planetary-type differential. The way that these work is by having the differentiating gears unequal, such that it is easier to differentiate in one way than the other. (Remember: torque is inversely proportional to shaft speed.) But it is a coaxial planetary, so when it is not differentiating the outputs must turn the same speed so it's a 50/50 when locked.

Charles' Cusco is exactly the same.

The noises when turning in locked mode are caused by the rears trying to track inside the front. Darned good tires on those cars stock. Enough to make the locking system strain on good, dry pavement. Do not turn it on for a dry autocross run. Leave it in automatic.

- Jtoby

 

[DG-FNR]

Are you sure about this? As far as I know, fully locked STi has some noises when turning and I believe that comes from the 35/65 lock. Unless it never goes to the full lock mode, I think it is rather that and not 50:50.

The STI is capable of going "past" fully locked and overdriving the rear wheels. The EVO is not capable of doing this.

From what I know so far about the ACD, it acts just like an open diff (which is 50:50 with no slip) and let say adjustable viscous coupler next to it. When fully opened, it is the same as any other open diff.

Yes. In essence, it's an intelligent locker.

With 50% locking it would allow 25-75% slip if pushed that way!

Yes, but you're looking at it backwards.

At fully locked (100% locking) the torque split is 50% front.

At fully unlocked (0% locking) the torque split is 100% front ASSUMING FRONT WHEELSPIN (which is the more likely case)

So at 50% locked, and again, assuming front wheelspin, the bias will be 75% front, 25% rear.

You _can_ get it to go the other way, but in able to do that, you must have _rear_ wheelspin and no front wheelspin. Given that we MUST be talking about throttle-on here, and given that rearward weight transfer is moving grip from the fronts to the rears, rear wheelspin is going to be rare. Possible, but rare.

The EVO diff, if programmed properly, acts like a perfect Quaife (or a Quaife acting perfectly acts like a properly programmed ACD, one or the other)

Where the STi is different is that 100% locked equates to a 35/60 torque split, not a 50/50 torque split (al least conceptually, the mechanism may function slightly differently). It "overdrives" the rear axle, and that's how it generates power-on oversteer.

The thing is that 60:40 weight distribution will always brake the rear end before the front one and that is why it ends up rear end happy.

Ah, but grasshopper, weight distribution under power is NOT 60/40. 60/40 is the upper bound; the actual distribution is going to be somewhat more rear biased. The power and grip (and the higher the CG) the more rear bias you have.

Does it ever reach 50/50, or beyond? *shrug* I dunno. I think Shep's car does. But it's a red herring anyway.

If the car is perfectly balanced mid-corner, and the torque split is 50/50, then adding power MUST cause understeer, because rearward weight transfer is moving grip rearward.

OK, so we give grip back to the fronts by reducing steering lock - but our centre diff, which is always (for lack of a better term) "seeking" to restablish a 50/50 torque split, will steal that grip to put power down with it. That means more understeer. The only way to get away from the understeer is to reduce power.

The STi centre diff isn't "seeking" 50/50, it seeks 35/65. That means it's possible to have enough front grip to sustain the cornering load plus 35% power, and have the rears on or over the grip limit, and you get oversteer.

What would be really cool would be a diff that not only has variable locking, but also variable torque split at full lock. Such creatures exist, XTRAC will build you one, and the price starts at $100k.

The WRC transmissions go even further. All 3 differentials are variable-lock and variable-split, so that power can be routed to individual wheels. Those start at $1000k.

DG

 

[jtmcinder]

The STI is capable of going "past" fully locked and overdriving the rear wheels.

Oh, do, please, say more about this. I am having great difficulty trying to visual what '"past" fully locked' might be and I have a lot of experience trying to understand things that you write. Is it fractal? Do the output shafts not only match in terms of rotation speed but also match in non-integer dimensions as well???

- Jtoby

 

[jtmcinder]

Where the STi is different is that 100% locked equates to a 35/60 torque split, not a 50/50 torque split

The only thing that is 100% is the fact that the above is wrong. Pure nonsense.

The torque split is created using unequal ratios WITHIN THE DIFFERENTIAL. (How many times does this have to be said, dude?) So when the differential is locked, the gears creating the unequal split have no effect.

- Jtoby

 

[MrAWD]

The STI is capable of going "past" fully locked and overdriving the rear wheels. The EVO is not capable of doing this.

Not quite getting this either…more info needed to understand this “past” thing!

With 50% locking it would allow 25-75% slip if pushed that way! And that would go either way!

Yes, but you're looking at it backwards.

At fully locked (100% locking) the torque split is 50% front.

At fully unlocked (0% locking) the torque split is 100% front ASSUMING FRONT WHEELSPIN (which is the more likely case)

So at 50% locked, and again, assuming front wheelspin, the bias will be 75% front, 25% rear.

Actually, you are the one that is assuming things that I didn’t say. And the things that I did say, you somehow omitted in your reply.

I didn’t say anything about where it will go. But I did say that it can go either way, which covers your option too!

You _can_ get it to go the other way, but in able to do that, you must have _rear_ wheelspin and no front wheelspin. Given that we MUST be talking about throttle-on here, and given that rearward weight transfer is moving grip from the fronts to the rears, rear wheelspin is going to be rare. Possible, but rare.

It is more then obvious that EVO is doing all of those magical things only if driven hard. If you try to baby that thing, it will be no different that a regular Civic or Accord with nice seats, Momo steering wheel and riceboy wing and hood.

About the what will be spinning, you again seeing a small picture and hanging in there as it is your life. First of all there is a huge difference between the way center diff is whether you are doing straight line acceleration or heavy cornering. Any straight line acceleration or counter steering is linked to the very heavy locking in the center diff. Only when cornering this opening of the diff comes into the play. So, tell us where you would see more weight transferred (basically outside front or outside rear) while doing heavy cornering on the car that has most of the weight in the front? Does it happen in your car that rear ever gets more weight then the front end?

If there is less weight transfer in the rear, you will have tire spinning on that end sooner then on the other end!

The EVO diff, if programmed properly, acts like a perfect Quaife (or a Quaife acting perfectly acts like a properly programmed ACD, one or the other)

There is no such a thing as perfect mechanical diff! Even if it is called Quaife! On the other hand, ACD can be programmed to do the near perfect thing!


Fedja
ps. had to split this in two, since it is already too long