Emergency steering with pull push
Discussion
Today, I was interested to read this:
www.dft.gov.uk/stellent/groups/dft_rdsafety/documents/page/dft_rdsafety_508279.pdf
On page 51, there is a laboratory study of the effectiveness of push-pull and cross-hands steering methods. The samples used are very small, only three groups of 8 male drivers, divided into full licence holders, learners, and non-drivers. However, it is interesting to see that learners and full licence holders exhibit slightly more accurate steering control when using crossed-hands, while non-drivers seem to be equally proficient at both.
No doubt that because the study is British, the learners and full licence holders would have been trained under the DSA-approved pull-push technique. Yet it seems to be the licence holders and learners -- those who are actually trained in pull-push, who perform better using crossed-hands.
The study itself isn't statistically significant, but it definitely hints at greater effectiveness of cross-hands -- especially for learners -- and provide food for thought for a more extensive and realistic study of steering techniques.
www.dft.gov.uk/stellent/groups/dft_rdsafety/documents/page/dft_rdsafety_508279.pdf
On page 51, there is a laboratory study of the effectiveness of push-pull and cross-hands steering methods. The samples used are very small, only three groups of 8 male drivers, divided into full licence holders, learners, and non-drivers. However, it is interesting to see that learners and full licence holders exhibit slightly more accurate steering control when using crossed-hands, while non-drivers seem to be equally proficient at both.
No doubt that because the study is British, the learners and full licence holders would have been trained under the DSA-approved pull-push technique. Yet it seems to be the licence holders and learners -- those who are actually trained in pull-push, who perform better using crossed-hands.
The study itself isn't statistically significant, but it definitely hints at greater effectiveness of cross-hands -- especially for learners -- and provide food for thought for a more extensive and realistic study of steering techniques.
Regarding the study:
* They don't mention how often during the task a wheel rotation of more than 180 degrees is required. I think we all agree that fixed input is better when driving down a twisty road at speed (what this test was designed to simulate), but where fixed input falls down is when manouvering tightly, such as a hairpin bend or a small roundabout. That is the argument for push-pull. Personally, I used fixed input for twisty road driving, and push pull for tight corners around town.
* They do hint at the fact that whilst neither method proves a substantial advantage over the other method, during moments where car control is key (i.e. sliding), fixed input could be better.
Interestingly, I am starting to understand that the feedback from the car is more important than the actual position of a control. This study hints at that (output is the same for push pull and fixed input, but input, or position of the control is different), but also in my own recent experience in driving a car with a delay on the throttle. I am finding a delay on the throttle really messes up my driving to the point where I just can't drive smoothly anymore. This is presumably because I am not measuring throttle control by the position of my right foot (which would be the same between drive-by-wire and cable actuation), but rather from the reaction of the car. A delay between input and output breaks the feedback loop and makes smooth car control impossible.
* They don't mention how often during the task a wheel rotation of more than 180 degrees is required. I think we all agree that fixed input is better when driving down a twisty road at speed (what this test was designed to simulate), but where fixed input falls down is when manouvering tightly, such as a hairpin bend or a small roundabout. That is the argument for push-pull. Personally, I used fixed input for twisty road driving, and push pull for tight corners around town.
* They do hint at the fact that whilst neither method proves a substantial advantage over the other method, during moments where car control is key (i.e. sliding), fixed input could be better.
Interestingly, I am starting to understand that the feedback from the car is more important than the actual position of a control. This study hints at that (output is the same for push pull and fixed input, but input, or position of the control is different), but also in my own recent experience in driving a car with a delay on the throttle. I am finding a delay on the throttle really messes up my driving to the point where I just can't drive smoothly anymore. This is presumably because I am not measuring throttle control by the position of my right foot (which would be the same between drive-by-wire and cable actuation), but rather from the reaction of the car. A delay between input and output breaks the feedback loop and makes smooth car control impossible.
It is rather worrying to read how much driving data is based on ancient anachronistic studies! They quote a study done in 1957, back in the days of horrendously heavy steering and strange camber angles. This study said that a wheel raked away from you (ala Renault Clio) is the best way of controlling a car. Well, this may have been the case back then, but not now! Look at the steering wheel in a Formula 1 car! why is it like that? I wish manufacturers would wake up! Univeral joints surely must be capable of getting the wheel at a more sensible angle for the modern motorist?! I much prefered the steering angle and driving position in my Caterham to that of a silly modern "sit up and beg" car.
RobM77 said:I'm with you on this Rob. I drove my mothers Renault thingy recently and it was a complete pain due to having imprecise steering dead ahead (it wandered and need constant minor correction due to a steering lag) and brakes which had a long travel followed by massively over-servoed action making smooth brakeing almost impossible. Add in CVT gearbox for extra control prcision (not) I can't comment on the throttle control as it only moved with foot firmly planted on the floor, so it only had two effective throttle positions.
Regarding the study:
* They don't mention how often during the task a wheel rotation of more than 180 degrees is required. I think we all agree that fixed input is better when driving down a twisty road at speed (what this test was designed to simulate), but where fixed input falls down is when manouvering tightly, such as a hairpin bend or a small roundabout. That is the argument for push-pull. Personally, I used fixed input for twisty road driving, and push pull for tight corners around town.
* They do hint at the fact that whilst neither method proves a substantial advantage over the other method, during moments where car control is key (i.e. sliding), fixed input could be better.
Interestingly, I am starting to understand that the feedback from the car is more important than the actual position of a control. This study hints at that (output is the same for push pull and fixed input, but input, or position of the control is different), but also in my own recent experience in driving a car with a delay on the throttle. I am finding a delay on the throttle really messes up my driving to the point where I just can't drive smoothly anymore. This is presumably because I am not measuring throttle control by the position of my right foot (which would be the same between drive-by-wire and cable actuation), but rather from the reaction of the car. A delay between input and output breaks the feedback loop and makes smooth car control impossible.
On the steering specifically, I tend to drive instinctively, and have never conciously used any "technique". If I thought about it I probably use a fixed hands position, but modified. I won't cross hands, but will steer one handed while the other hand gets a grip further up the wheel, and will then have two hands on one side of the wheel. Having a Chim with no PAS may have some bearing on this technique; in tight turns requiring a lot of lock having two hand on one side of the wheel helps with leverage! I find push pull way too much of a compromise, and too slow.
I may just be a bit weird though. I could never get the hand of tying shoelaces the conventional way, so made up my own technique. I don't touch type, but a few years ago when my wife was a secretary and trained touch typist she was quite put out that I could type significantly faster than her, but my hands were "all over the place". Home keys, what the flippin heck is that all about?!
Thumbs on spokes? There are parallels with bike riding I guess, when its accepted wisdom to rest thumbs on the top of the bars rather than wrapping around, thereby avoiding the most frequent biking injury - broken thumbs. Frankly, I never fancied the alternative injury (I like my face even if no-one else does), and I found that a degree of control was compromised. In the end I simply favoured not running into anyone as a much better technique of avoiding injury.
Yes, interestingly, I have taught myself to type and can go quicker than most typists (I have been tested a few times, and come out in the top end of typists speeds). I used a weird technique! All fingers are going, but it isn't strictly touch typing.
I am the same with raquet sports - I have played since I was a kid with the wrong grip, but find it easier. I play with the head of the raquert in the same plane as the palm of my hand - I find this more natural. Every coach I've ever had has tried to get me to rotate it upwards, but it just isn't the same!
I am the same with raquet sports - I have played since I was a kid with the wrong grip, but find it easier. I play with the head of the raquert in the same plane as the palm of my hand - I find this more natural. Every coach I've ever had has tried to get me to rotate it upwards, but it just isn't the same!
victormeldrew said:
I drove my mothers Renault thingy recently and it was a complete pain due to having imprecise steering dead ahead (it wandered and need constant minor correction due to a steering lag)
Funny you should mention that as the last Renault Megane I drove had the worst steering I can recall in a "modern" car.
As you say imprecise and it did indeed seem to have a lag effect. After two days and a fair few miles I concluded that if I drove it much longer I would be in serious risk of dumping it in the boondocks on a fast sweeper.
RobM77 said:
Not to mention the Megane's hilariously daft throttle delay! A friend of mine has one for a few months and he was shocked!
Did you see Top Gear on Sunday? Thought about the fly by wire discussion when they announced the £507 for the Sport Chrono package on the Cayman gave you a chronometer and better throttle response!
Seems to answer the question whether manufacturers deliberately degrade throttle delay in drive by wire applications.
Flat in Fifth said:They certainly didn't in the case of the Maser 4200, it was notoriously difficult to drive bacause of the sensitivity of the throttle. I drove one an a Supercar experience, and got on fine wih it, but by heck it was lively! Most people were having real trouble with it.
Not to mention the Megane's hilariously daft throttle delay! A friend of mine has one for a few months and he was shocked!
Did you see Top Gear on Sunday? Thought about the fly by wire discussion when they announced the £507 for the Sport Chrono package on the Cayman gave you a chronometer and better throttle response!
Seems to answer the question whether manufacturers deliberately degrade throttle delay in drive by wire applications.
Are you sure that wasn't because of a fly-by-wire throttle? I've *never* driven a cable throttle car that was tricky. right up to 400bhp+ mid engined things like Noble M400s, or powerful Caterhams. However, the Nissan 350Z that I drove recently jumped if you so much as breathed on the throttle. The problem was a tiny delay. It was tiny, but enough to interupt the brain-foot link so you didn't know what you'd done till it gave you the power you'd requested, and that was usually a little more than you needed, all at once, thus the jerk.
cptsideways said:
Mr Ueo in action
Oh My Fooking God! Does he have a wheelbarrow to transport his testicles in?
crikey!!RobM77 said:
I've *never* driven a cable throttle car that was tricky.
Would reckon that the only cable throttle cars where I have had problems have been due to a sticky throttle.
Most manufacturers in their PR blurb about drive-by-wire throttles say something like "the position of the throttle pedal is measured twenty thousand times a second."
OK so they measure it, doesn't mean that they actually do something with it at the same rate.
Flat in Fifth said:
I've *never* driven a cable throttle car that was tricky.
Would reckon that the only cable throttle cars where I have had problems have been due to a sticky throttle.
Most manufacturers in their PR blurb about drive-by-wire throttles say something like "the position of the throttle pedal is measured twenty thousand times a second."
OK so they measure it, doesn't mean that they actually do something with it at the same rate.
True, some throttles I've used have needed a bit of oil! Drive by wire though is just a complete waste of time, and can be very dangerous as well. Imagine that car you mentioned that was tricky to drive in the snow and ice? My 330 is a nightmare..
Dead right about the sampling rate of the throttle position. This would just run at CPU speed I should imagine!
Doesn't mean you can't count to three before the throttle responds. The fact that they're mentioning such facts must mean that they are aware of problems. If you really probe a lot of salesmen and service staff at dealerships it becomes apparent that lots of complaints have been raised since this technology came in.RobM77 said:
I've *never* driven a cable throttle car that was tricky.
Would reckon that the only cable throttle cars where I have had problems have been due to a sticky throttle.
Most manufacturers in their PR blurb about drive-by-wire throttles say something like "the position of the throttle pedal is measured twenty thousand times a second."
OK so they measure it, doesn't mean that they actually do something with it at the same rate.
True, some throttles I've used have needed a bit of oil! Drive by wire though is just a complete waste of time, and can be very dangerous as well. Imagine that car you mentioned that was tricky to drive in the snow and ice? My 330 is a nightmare..
Dead right about the sampling rate of the throttle position. This would just run at CPU speed I should imagine!
I had a big issue with a 330ci that I bought a couple of years ago. I took it back within 50 yds of buying it from new because of the poo throttle response. They told me they were all like that. Had BMW tech man up who said it was a 'feature'.
In the end I flogged ot after 3 months and got a new STI8. Lots more fun.
andygo said:
I had a big issue with a 330ci that I bought a couple of years ago. I took it back within 50 yds of buying it from new because of the poo throttle response. They told me they were all like that. Had BMW tech man up who said it was a 'feature'.
Ah, yes. How to fix a bug without any code modifications: document it, then it's a feature. Am hinten durch Technik.
andygo said:
I had a big issue with a 330ci that I bought a couple of years ago. I took it back within 50 yds of buying it from new because of the poo throttle response. They told me they were all like that. Had BMW tech man up who said it was a 'feature'.
In the end I flogged ot after 3 months and got a new STI8. Lots more fun.
Very interesting to hear... I did the same thing. I rung them when I got home and told them that my car had a serious problem. They wouldn't give me my money back and said 'they all do it'. Mine's on Autotrader now.
RobM77 said:
I think maybe this letting go business is a technique for road cars with slow steering racks?
You don't need to let go, but with a relaxed grip, you can feel the feedback and let the car lead. I only let go when I've overdone the slide and know I'm too slow with the wheel to keep up. If you don't touch the wheel as you come out of the drift, it tends to tank slap anyway, which you could have removed by damping the wheel by holding it.
That Japanese guy is something else. I don't think it would be possible to keep up with the wheel rotationally at that speed however good you are!
Gassing Station | Advanced Driving | Top of Page | What's New | My Stuff