Smart Car of America Forum banner

Explanation for the huge understeer bias

13K views 55 replies 17 participants last post by  Miss Mercedes  
#1 ·
A couple thousand miles ago I started to notice my smart felt a little floaty on rough surfaces, most notably gravel. More so than before, that is... this just brought it to my attention. Tire pressures were the same 36R/44F I've been using for some time. My first thoughts for what had changed were that either a shock had died or I'd picked up a bit of outward toe somewhere. Checked the shocks by jumping on the bumpers... none were quite as good as they used to be, but all seemed plenty good enough. Checked for toe out as well as I could in my driveway, but due to not having an alignment machine I couldn't find anything but a gross error... which I didn't find.

A few days after those checks, I figured out it wasn't anything to do with the slip ratio vs. grip curve like I thought it was, it was just an oversteer bias. The car still feels plenty understeery on pavement, but when I hit fine-scale bumps it goes a bit oversteery where it had been merely neutral before. My best explanation for the difference is that the rear de-dion tube suspension has a lot more unsprung weight than the front McPherson strut suspension. The engineers presumably just tuned the spring / damping / sway bar rates to keep it from being oversteery in any situation, resulting in a big understeer bias in most situations.

As for what changed about my car, I found that my rear tires have worn far too much for their miles... and the pattern is as if they're underinflated, which is odd since they're stock sizes and have been run at the factory pressure recommendation since I got them installed. My front tires have dead even (and much less) wear, and I've been running them at 44 lbs (sidewall rating). Looks like I'll have to crank up the rear pressure a bit, which will lose me my directional stability I got by using those crazy pressures. :( Anyway, I expect the wear lost me a bit of grip in back, leading to a bit more of an oversteer bias overall that I just can't feel on pavement because it's so far from neutral to start with.
 
#2 · (Edited)
36R/44F is a bit on the high side. I believe the door jamb sticker says 29F/36R (assuming stock tire sizes) Higher tire pressures will cause the car to float a little more as there will be less tire on the road (so to speak) as over inflated tires will ride in the center of the tread. Tire pressures are recommended by the factory to give the best traction and ride for a specific car/suspension setup and the pressure listed on the tire will give you maximum weight capacity for the tire, but won't take the car's characteristics into account.

If you are going with non-stock tires, height or width, then you will have to do the trial and error method. I'd lean towards the same ratio of front to back pressure (but that's just me)

Confusing since you say you have been running factory pressures, but the front was set at 44 (??)
 
#3 ·
Also, the skinny front tires are part responsible for the understeer.

As for the tires, unless you have wedges installed, the camber angle combined with overinflated tires can cause that wear.

My smart has the wide tire factory option and it just looks like they'll have insane wear if I don't fix it.
 
#6 ·
The rear camber can cause the odd rear wear, but it didn't on my old Contis for whatever reason.

Presumably upgrading the front tires will make it even more oversteery on gravel.

Did you ever consider running a set of tires with a softer tread compound? It might make a big difference in the feel of the car if you're willing to live with the shorter tread life.
Absolutely, just haven't got around to affording the new wheels/tires yet (there's just about zero choice in stock sizes).
 
#5 ·
@jimvw57:

Rear is at factory spec of 36, front is 44 where the factory says 29.

I'm aware of what the factory says, but I'm also aware that the factory specs are a balance between various factors, and I'm not afraid to experiment to figure out what works best for me. I found that I really liked the extra directional stability I got from high front and low back pressures, so that's what I've been using. Also, you may note that my tires at a vastly different pressure than the spec are doing far better for wear than my tires at the factory spec. :wink:

A Pacejka curve is a graph of forces transmitted through a tire vs. the slip ratio at a tire (just look at the black line in this example):

Image


There are 3 main characteristics of interest here: how quick the graph ramps up (precision), how high the peak is (static grip), and what height the line stabilizes at past the peak (kinetic grip).

When you inflate your tires more, you lose some grip, but you gain a lot of precision. A floaty feel is caused by bad precision right around the x = 0 mark, not bad grip as such.
 
#16 ·
A good technical explanation [wikipedia.org]

I adjusted tire pressures that way since after I disabled my power steering I was having some issues with tracking straight in ideal conditions (no wind). The issue seemed to just be a lack of directional stability as defined in the linked article. To fix that, you need to put more rubber on the ground in back and less in front... so I did, and it worked.

Imagine a rear-heavy car with the same size and pressure tires all around. If it ends up sliding exactly sideways, it won't self correct, you'll end up pointing the wrong way. This is because the tires with less weight on them grip better per weight. A front-heavy car with the same tires will end up facing the correct direction. Suspension can correct for this to some extent, but not completely. What you really need to adjust is the tires. Get wider tires in back or adjust rear pressures closer to max grip in back than front.

front end isn't directly connected to steering wheel - there's a computerized link in there to make the thing stable at speed. I figured that out after driving it on the freeway where my Cooper S rides up and down the troughs created by truck traffic and you can feel the response through the steering wheel. In my wife's smart, same grooves yet zero feedback throught the steering wheel, just rock steady even though you can sense the front end wandering ever so slightly side to side while the stabilization software keeps you going straight down the road.
There is a mechanical link, but the power steering is electric and always tries to return the wheel to center. It causes problems when you've got a crosswind and have to fight both the wind and the PS motor. I disabled my power steering for that reason (my area gets a whole lot of wind). I've got better feedback for it, but it isn't quite as good at tracking straight in ideal conditions.
 
#9 ·
front end isn't directly connected to steering wheel - there's a computerized link in there to make the thing stable at speed. I figured that out after driving it on the freeway where my Cooper S rides up and down the troughs created by truck traffic and you can feel the response through the steering wheel. In my wife's smart, same grooves yet zero feedback throught the steering wheel, just rock steady even though you can sense the front end wandering ever so slightly side to side while the stabilization software keeps you going straight down the road.
 
#11 ·
I hate to tell you this, but the steering wheel is directly linked to the steering rack. While the smart does have electric power steering, it is an assist type. While it offers power assist, it also helps maintain center as well. I will agree that the smart doesn't offer the highest feedback in the world, however, I do not find it objectionable. Certainly isn't as bad as the old cars from the 70's & 80's with the "arms shot up with novocain" feeling.

I don't know of a car on the road that doesn't have a mechanical link from the steering wheel to the steering rack. That would be a lawyers dream come true if there was ever a failure.:rolleyes:
 
#10 · (Edited)
Interesting experiences shared here... I have the original 155 & 175 Contis. After getting tired of the constant mushy understeer set up for undoubtedly good reason, I set the tires to 35psi all-around. Running it that way for a couple years, no unusual wear. Feels more balanced to me, and more responsive.
 
#15 ·
WhitenBlack06 mentioned about cars ON THE ROAD Yet CT says "Such systems already exist in certain work equipment/vehicles and it's inevitable that they'll also make their way to passenger cars & trucks eventually" so I guess he agrees that they are not on the road yet. Sounds like WhitenBlack06 did his research...
 
#17 · (Edited)
LOL

You try to make an argument out of everything I post. Maybe you and Kermit should join forces.

The fact of the matter is that I'm simply saying that steer-by-wire will be something we see on cars in the near future. Like you said, I obviously agree we're not there yet, so what is the argument about again?
 
#18 ·
I see... you can dish it out but you can't take it.
Just stating the obvious. No argument, just what I see.

Drive by wire is not a common item on cars used ON THE ROAD. Someday, maybe but then someday pigs will fly!!

Done with this.
 
#21 ·
That's what they said about drive-by-wire too. And of course we all know what happened there. Steer-by-wire is coming eventually and nothing will stop it. The safety features and fail-safes that will be built into the system will be more than enough to satisfy all but the most rabid opposers of the technology.

Did you see the autonomous Prius that drove a blind man to Taco Bell? If that's possible, then steer-by-wire is child's play in comparison.

Google's Autonomous Prius Drives Blind Man to Taco Bell | Autopia | Wired.com
 
#22 ·
The confidence of your assertion does not answer the basic question of how you steer with no electrons available. The autonomous Pious is an apples to Aardvarks comparison.

Aside from the tech side of things, you also need to make things the public will buy. A vehicle with no direct connection between steering wheel and road will be a tough sell. When the Japanese start buying steer by wire in their home market, I'll see the light.
 
#23 ·
The confidence of your assertion does not answer the basic question of how you steer with no electrons available.
Like I said, there will be safety features and fail-safes. Something as simple as a capacitor can be used to supply "electons" in the extremely rare event of a total electrical failure. That's just one example of a way to provide steering long enough to get safely off the road.

The autonomous Pious is an apples to Aardvarks comparison.
It wasn't meant to be a direct comparison....it was meant to illustrate that there's no stopping technology. If they can make a car that can drive itself safely with no human input, then why does it seem so outrageous that we'll see cars that can be turned without a direct mechanical connection between the steering wheel and the wheels? You suggested safety as the main concern; I suggested that safety will eventually be a non-issue.
 
#25 ·
Experimenting with tire pressures

It seems strange to me that lay people with extremely limited resources would try to "out guess" a multi-billion" auto company where suspension and tire pressures are involved. Remember Benz is the company that failed the Moose test when they developed the first gen. "A" series. That event cost them some serious dollars to rectify. I'll bet they did a serious amount of testing when they developed the new rear engine, rear wheel drive smart.

Remember you don't just develop a system for dry pavement it must function when driving on various road surfaces and in different weather conditions. You can experiment and maybe achieve slightly better handling in one situation only to create a safety hazard with another condition.

It may be analogous to switching to a different brake pad compound in order to reduce brake dust accumulation on your wheel rims while unknowingly decreasing braking ability leading to an accident. You just don't assume anything.
 
#27 ·
It seems strange to me that lay people with extremely limited resources would try to "out guess" a multi-billion" auto company where suspension and tire pressures are involved. Remember Benz is the company that failed the Moose test when they developed the first gen. "A" series. That event cost them some serious dollars to rectify. I'll bet they did a serious amount of testing when they developed the new rear engine, rear wheel drive smart.

Remember you don't just develop a system for dry pavement it must function when driving on various road surfaces and in different weather conditions. You can experiment and maybe achieve slightly better handling in one situation only to create a safety hazard with another condition.

It may be analogous to switching to a different brake pad compound in order to reduce brake dust accumulation on your wheel rims while unknowingly decreasing braking ability leading to an accident. You just don't assume anything.
It's true that we don't know everything about the car, but there are some principles of car handling that apply anyway. If you can show that I'm probably doing bad things for safety, that's one thing... but really, anything I do *might* compromise safety somehow, but most won't, and considering my goals here I'm probably improving it...

Are you guys driving the same little fork lift engine powered golf cart I am or is there a new Ferrari/Lamborghini microcar that's come out. Mine's the little plastic skinned city cart, the one that parks in half a space...It's not a race car by any stretch of delusional psychiatry....
Just because it doesn't have much power or grip is no reason for us to not care about handling at all...
 
#26 ·
Are you guys driving the same little fork lift engine powered golf cart I am or is there a new Ferrari/Lamborghini microcar that's come out. Mine's the little plastic skinned city cart, the one that parks in half a space...It's not a race car by any stretch of delusional psychiatry....
 
#32 ·
Let's revisit the original post on this thread - the complaint was about "huge understeer bias" with the smart's handling. Wading through all the hypertechnical stuff that ensued (not that it isn't pertinent) if the car was actually understeering severely on a consistent basis you would expect accelerated front tire wear. Because understeer is by definition loss of grip of the front tires, driver input is to turn, the tires can't hold, the car tries to plow staright ahead. The post actually stated that his front tires were wearing evenly, and less so than the rears. I think what is actually happening is that the stability system actually commendeers driver input to match what the tires/ suspension is designed to handle. Driver perception is that the car is understeering because front end response doesn't match driver input. That, to me, connotates no direct link between driver and front wheels.

researching the smart car you'll find that the original fortwo had to be redesigned just prior to release in the late 90's due to "significant problems with stability at speed"

I agree that true drive by wire is on the horizon.

A Cessna 172 still uses cables and mechanical links. A Boeing 757 is fly by wire.
Guess which one has a better safety record.
 
#34 ·
Let's revisit the original post on this thread - the complaint was about "huge understeer bias" with the smart's handling. Wading through all the hypertechnical stuff that ensued (not that it isn't pertinent) if the car was actually understeering severely on a consistent basis you would expect accelerated front tire wear. Because understeer is by definition loss of grip of the front tires, driver input is to turn, the tires can't hold, the car tries to plow staright ahead. The post actually stated that his front tires were wearing evenly, and less so than the rears. I think what is actually happening is that the stability system actually commendeers driver input to match what the tires/ suspension is designed to handle. Driver perception is that the car is understeering because front end response doesn't match driver input. That, to me, connotates no direct link between driver and front wheels.
Definition of understeer: the front tires run higher slip angles in corners than the rear tires.

Smarts do this, no question about it. When cornering on fairly smooth surfaces, the rear end stays firmly planted (low slip angles) while you can get the front to do just about anything you want. There's a mechanical link, because I disabled the power steering and I'm still doing fine. It's got tons of feedback with the PS disabled, and I have no doubts about what it's doing.

If I crank the wheel way over with ESC (stability control) off, it'll push crazily and I'll probably end up off the road. With ESC on, it'll yank the car around by engaging the inside rear brake, generally upsetting everything about the handling. I've never gotten to either of the above points. If I did regularly, then yes I would get huge front tire wear... and I'd have probably wrecked the car by now.

ESC isn't it either. It behaves the exact same way with the ESC disabled, and I don't make the ESC do anything in my normal driving. The only times on dry roads that ESC has come into play on my car, I had no need for it and wasn't even cornering that hard to start with. On snow, it's nice... but that's different.

I take corners fast by slowing to my minimum speed at the entrance of the corner, dialing in lock, and flooring it. I adjust steering through the corner to maintain my target line, but there comes a point when I'm at max front grip. This point is obvious from the steering feedback. If I get there, I don't turn the wheel any further, I let off the gas. This tranfers load to the front tires, making the handling much more neutral and giving me plenty of front grip to make it around the corner.

Within this process, the electronics don't interfere at all, and I never go past the front tires' max grip point. They're never sliding as such. Most people don't even think of tires as slipping until they get to 10% slip ratio or so.

This will wear the tires faster than sedate cornering, but there's no comparison as far as tire wear goes between this and careless fast cornering that ends up sliding like crazy (more than ~10% slip ratios).
 
#33 ·
The little bit I saw about a Cessna 172 said it was built in 1956, Boeing 757 were built in 1981, jet vs single engine plane. big difference in technology there. I did hear the 451 smart was lengthened to make it more stable.
 
#37 ·
Most fun with oversteer was an old fiberglass dunebuggy with an 1835 and close 3rd and 4th. Big bias ply tires in the rear and skinnies in the front. You could slide it any where. It was super responsive to steering and throttle input. It was so much fun. Then I went and made it a drag car and ruined it.
 
#42 ·
Multiple levels of redundancy are built in to a modern jet transport to support the entire airplane and all of its systems. Without the overall layered redundancy, fly-by-wire could never happen. It takes a bit more than a capacitor. :rolleyes:
What automaker is going to expose themselves to the extreme liablity of drive-by-wire guidance. Conceptually, of course, but practically, I doubt it. Until the automakers can design in an acceptable level of redundancy at a cost point that will be acceptable to them and to the consumer, it ain't gonna happen.
What about ongoing maintenance of these systems required to maintain the level of performance required? Who pays for that?

Speaking of the Google's Prius test. Very impressive, but what about the child who runs out between parked cars, or the texting person who blows through a stop sign. If every trip was a controlled experiment, it would be great. That's just not the real world.
 
#43 ·
Multiple levels of redundancy are built in to a modern jet transport to support the entire airplane and all of its systems. Without the overall layered redundancy, fly-by-wire could never happen. It takes a bit more than a capacitor. :rolleyes:
Do yourself a favor and research drive-by-wire accelerator pedals. You'd probably be surprised by what you read. I never knew much about it until I began to do my own research as a result of Toyota's recent troubles with "runaway" vehicles. I suspected that it was driver error and it turns out I was correct. Very similar to the whole Audi 5000 stories in the 80s.

Anyway, the point is that "drive-by" technology is already in our cars. If you don't think we'll see more of it, I think you're not paying attention to the way the technology is headed.

What automaker is going to expose themselves to the extreme liablity of drive-by-wire guidance. Conceptually, of course, but practically, I doubt it. Until the automakers can design in an acceptable level of redundancy at a cost point that will be acceptable to them and to the consumer, it ain't gonna happen.
Of course not...do you think they'd sell cars that didn't have an acceptable level of safety? They already did it with drive-by-wire (throttle) control. And it's a huge success across every manufacturer's vehicle line-up.

What about ongoing maintenance of these systems required to maintain the level of performance required? Who pays for that?
The systems we have now are designed to last the life of the vehicle. Actually, beyond the life of the vehicle. And it seems that they are doing just that. No maintenance is normally required.

Speaking of the Google's Prius test. Very impressive, but what about the child who runs out between parked cars, or the texting person who blows through a stop sign. If every trip was a controlled experiment, it would be great. That's just not the real world.
Do you realize that the Prius isn't the only autonomous car driving around on public roads? And that these cars use radar and lasers to detect objects/people that may be in its path? Heck, we already have cruise control that sensor the distance between your car and the car in front of you and automatically adjusts the speed in response.
 
#44 · (Edited)
I'm so impressed. You really do have an answer for everything.

But, do yourself a favor . . . don't assume that you're the consummate expert in all things. I'm starting to believe that you're not so much a Critical Thinker and much a Critical Critic.
I'm quite familiar with drive-by-wire accelerators. I've been driving them since '08. Linear throttle response is a bit different than subjective steering input.

I don't see any good reason to become another target of your one-upsmanship.
Have a good one.
 
#47 ·
I'm so impressed. You really do have an answer for everything.
Thanks, but I think you're giving me far more credit than I deserve. I don't have an answer for everything, but I do have an answer for things that fall within my scope of knowledge and/or expertise.

But, do yourself a favor . . . don't assume that you're the consummate expert in all things.
Expert in all things? I've never thought that at all. That said, if you want to have a discussion, please stick with the topic instead of resorting to ad hominem fallacies, which is what you've done so far. Sadly, this type of behavior seems to happen quite a bit on this forum. Instead of getting personal with me, you should really just post your opinion (or facts even!) on the subject.

I'm quite familiar with drive-by-wire accelerators. I've been driving them since '08.
Almost everyone who owns a new car (since '08) has been driving them, so I guess we're all familiar with them. Then again, I'm quite sure that most people have no idea if their throttle is controlled by a cable or not.

Linear throttle response is a bit different than subjective steering input.
Subjective steering input? LOL. You think that steering input is subjective? Unless your steering input takes place in your mind, then I don't think it makes any sense to say that steering input is subjective.

But either way, throttle control isn't all that different from steering control, as far as the actual driver's input. Think about it. Or don't.

I don't see any good reason to become another target of your one-upsmanship.
Have a good one.
You too! Thanks!