28 replies to this topic

[gruntguru]

As the owner of a Focus with Drive By Wire and manual transmission, I have always been annoyed by the characteristics of the throttle. At a guess I would say the first 20% of throttle travel produces 80% of the available torque. From past experiences fitting quadrant linkages to carburetted vehicles, I have always found it is easier to drive smoothly and economically when the throttle-torque relationship is more linear or even biased the other way.

I can only guess that manufacturers map DBW in this way to make the car seem more lively in a road test situation. Can anyone confirm this?

Does anyone have experience with altering the characteristic? I have thought of modifying the pedal-potentiometer mechanism or perhaps biasing the potentiometer with an external resistor. Clearly the ultimate fix would be re-mapping but I doubt that would be cost effective.

[desmo]

I rented an AT Focus this year and it was similar. The mapping seemed annoyingly biased towards tip in, rendering the remainder of the pedal travel dead and uninteresting. Maybe it helps with the mileage cycles or it was focus grouped.

[Greg Locock]

It's not just Focus, I find it very easy to accidentally generate wheelspin in a diesel Mondeo auto, which is both embarrassing and potentially expensive. Yes, throttle progression will have been exhaustively cliniced. I could look it up and find out more but then I'd have to kill you.

https://dspace.lboro.ac.uk/2134/4613

Also bear in mind the progression /may/ be adaptive on the Focus, (as shift schedules are typically) so if you are other users hoon around in the car you may find it tending towards sharper shifts and higher revs.

So far as modifying the signal goes, most throttle pedals have an analogue pot on the shaft and it is only digitised at the EEC, so a bit of op-amp trickery will give you the ability to roll your own.

[gruntguru]

So far as modifying the signal goes, most throttle pedals have an analogue pot on the shaft and it is only digitised at the EEC, so a bit of op-amp trickery will give you the ability to roll your own.

I will probably have a look but I won't be surprised if the thing has two pots (for redundancy) and probably with mirrored signals. Also the ECU probably does continuity checks so it won't like any apparent change in impedance. >> Limp home mode.

[gruntguru]

https://dspace.lboro.ac.uk/2134/4613

Also bear in mind the progression /may/ be adaptive on the Focus, (as shift schedules are typically) so if you are other users hoon around in the car you may find it tending towards sharper shifts and higher revs.

Car is a manual so shifts and revs don't come in to it.

Quote from linked paper ("conclusions")
" . . . . drivers considered subjectively to be more aggressive in nature preferred a lower IIP (Throttle gain). This indicates that they prefer or require more control over the available power"
That statement would certainly explain the preference stated by myself and other posts here but certainly wouldn't indicate adaptive mapping of the throttle on Focus or Modeo.

IMHO the Focus throttle is mapped to achieve the most favourable response from first-time drivers. Unfortunately that does not coincide with mapping that is most satisfying to a longer term user eg owner.

[MatsNorway]

Some chip tuning can probably fix that.

[Magoo]

I will probably have a look but I won't be surprised if the thing has two pots (for redundancy) and probably with mirrored signals. Also the ECU probably does continuity checks so it won't like any apparent change in impedance. >> Limp home mode.

Generally three on a Ford, oh fun. Also three matching TPS inputs. There are multiple continuity and input check routines, all of which will throw a malfunction lamp instantaneously.

See if you can lengthen the effective lever arm of the pedal. You can also relocate the pedal sensor assembly (APPS) on the firewall to change the pedal's action point relative to heel point on the floor. Adding return spring force might also help you obtain what you want. Beyond that you are spooning into a can of worms.

[Tony Matthews]

Adding return spring force might also help you obtain what you want.

Which reminds me of a tuning trick supposedly perpetrated by some Italian mechanics when a customer complained of lack of grunt. Replacing the throttle return spring with a weaker one gave instant "Mama Mia!"

[Greg Locock]

While it might seem attractive to give first drive customers the perception of grunt, this has to be weighed against the Things Gone Wrong at 3 Months in Service (TGW@3MIS) where drivers will complain about silly drivability issues. TGW is tracked at a system level, and you are expected to reduce TGW every year. So there is a counterbalance in the system.

[giskard]

That exlains why the more plebeian the car (e.g. Camry), the more sensitive the throttle. Non-sporting drivers like sensitive controls.

I suppose the same reasoning applies to the brakes? Yet another peeve of mine.

Edited by giskard, 22 July 2013 - 15:53.

[blkirk]

My Acura TSX does the exact same thing. It's very annoying. I hooked up an OBD-II reader and logged throttle position for my drive home one day. It only needed ~16% throttle to cruise at highway speeds (65 mph). By the time you get to 19% throttle, you're travelling at 80 mph. It's really, really hard to not speed in that car. You basically have 0.5" of useful pedal travel between idle and "Oops! I didn't mean to rear-end that car in front of me".

I, too, had been thinking about trying to add some sort of biasing resistor into the system to make it less sensitive. I hadn't thought about the redundancies and sanity checks that are certain to be lurking about in the system. I should have known better after all the trouble I had helping someone diagnose their 911->Boxster engine transplant. Maybe I'll enlist the help of the EEs at my office. They build real-time control systems for a living, so if there is a way to lie to the ECU, they should have the skills to find it. Good to know that there may be a market out there for retrofit modules to make throttles less touchy.

Edited by blkirk, 22 July 2013 - 16:52.

[rory57]

I have always been annoyed by the characteristics of the throttle. At a guess I would say the first 20% of throttle travel produces 80% of the available torque.

I hav'nt driven a Ford in ages except those with a Diesel engine but my recollection of carburreted Fords is as you describe. It did make downshifts easy, having that eager throttle response.

I think they did the same with steering response; all 'turn-in' but no grip!

My daily drive (Toyota MR2) is just the opposite, a very gentle throttle response, and it has a cable to the throttle. It could do with pair of Webers!

[MatsNorway]

Surely a chip tuner should be able to do something.

[pugfan]

That exlains why the more plebeian the car (e.g. Camry), the more sensitive the throttle. Non-sporting drivers like sensitive controls.

I suppose the same reasoning applies to the brakes? Yet another peeve of mine.

Another addition to the sample size: My Mother-in-Law's 2005 Hyundai Sonata has a ridiculously sensitive throttle to the extent that it is quite difficult to take off smoothly with if you're used to driving anything else. Pushing past 30% of throttle results in no extra urge at all.

[munks]

gruntguru - My FR-S has the same problem. However, it also comes with a 'Snow' button that changes the mapping to something more controllable (I'm not sure ATM if it reduces maximum torque, however). Does the Focus have anything similar - a mode for low-traction situations?

[kikiturbo2]

the first renault clio sport that came with DBW throttle (due to the need for ESP) had an optional Ragnotti pack available from REnault sport... this comprised of a new ECU and old cable operated throttle.. LOL

[gruntguru]

gruntguru - My FR-S has the same problem. However, it also comes with a 'Snow' button that changes the mapping to something more controllable (I'm not sure ATM if it reduces maximum torque, however). Does the Focus have anything similar - a mode for low-traction situations?

No such luck.

It would be great if Ford were to offer some software upgrades/options - wouldn't fit with corporate goals of course.

[Greg Locock]

Of course one huge issue you are all carefully forgetting about is that using the angle of a butterfly valve is a lousy way of controlling the torque produced by a variable speed engine with more or less torque proportional to manifold depression.

At low rpm a butterfly valve is simply insensitive. Always is always will be. Say your engine has a red line at 7200 rpm. At typical cruising revs of 1800 that means that the 'full throttle' MAP is reached at (18/72)^2 of fully open butterfly area. That's 1/16 of fully open area, which is 20 degrees.

So if you had linear pedal travel to butterfly angle, at cruising revs you'd effectively need only 22% of your pedal travel to go from zero to full torque.

Also, I think that paper has a fair bit of idealistic wishful thinking in it. BMWs are probably driven by people who think they are more 'sporty' than MBs, yet MB has the taxi driver throttle pedal you lot think you want, BMW is fairly typical of any other mass produced car.

[gruntguru]

Certainly wasnt forgetting rpm dependence. I would assume the Focus has a 2D throttle map anyway.
As to driver preferences, I had one experience of fitting a linkage with adjustable progression to a race car. I had it set up for near linear torque-throttle. One driver loved it while another wanted it super sensitive from tip in. From Motec data the first driver spent some time at all throttle angles searching for the grip limit while the second used the throttle like a light switch - mostly at zero or 100%.
Surprisingly there was not much difference in pace between them.

[munks]

What does "taxi driver throttle pedal" mean?

[Greg Locock]

Reputedly MB developed their throttle pedal to be extremely lethargic at the request of taxi drivers who were keen to save fuel. Whether that is the true reason I don't know, but of the cars that were evaluated MB always felt as though the pedal was dead.

[gruntguru]

On the subject of 2D throttle mapping. May not sound like a realistic situation but try this in a manual Focus. Allow the car to slow to idle in 1st gear then try to squeeze the throttle gently enough to accelerate without a jerk. This is a situation typically seen in a traffic jam and you are trying to operate the vehivcle at walking pace or slower. Clearly at such low revs the throttle should have a "taxi driver" map since 95% of full torque occurs at - probably about 5% throttle.

Edited by gruntguru, 25 July 2013 - 23:01.

[Magoo]

Of course one huge issue you are all carefully forgetting about is that using the angle of a butterfly valve is a lousy way of controlling the torque produced by a variable speed engine with more or less torque proportional to manifold depression.

At low rpm a butterfly valve is simply insensitive. Always is always will be. Say your engine has a red line at 7200 rpm. At typical cruising revs of 1800 that means that the 'full throttle' MAP is reached at (18/72)^2 of fully open butterfly area. That's 1/16 of fully open area, which is 20 degrees.

So if you had linear pedal travel to butterfly angle, at cruising revs you'd effectively need only 22% of your pedal travel to go from zero to full torque.

Also, I think that paper has a fair bit of idealistic wishful thinking in it. BMWs are probably driven by people who think they are more 'sporty' than MBs, yet MB has the taxi driver throttle pedal you lot think you want, BMW is fairly typical of any other mass produced car.

[Joe Bosworth]

Greg and McG

The situation is actually much worse than that.

I once back in the dark ages set up a flow experiment with a butterfly valve restricting a pipe with a constant pressure at one end. Measured total flow (mass) and plotted against percentage of butterfly opening.

My data is long lost but in summary the 90% point of flow/mass was reached with the butterfly about 65% open, (of its travel of 90 degrees).

The butterfy opening to flow/mass was far from linear with most of the flow/mass change occuring between about 7 to 45 degrees of movement.

It is an easy experiment to set up and encourage anybody elae to do so and publish the data.

Regards

[gruntguru]

Worse still Joe - the IC engine (at constant speed) which provides the pressure drop across the throttle is more a constant volume-flow pump than a constant pressure pump.

This is the reason that (at low rpm where the butterfly valve is massively oversized compared to the WOT flow potential of the valve) a very small opening is sufficient to raise the manifold pressure to near atmospheric, and the torque to near 100%.

[munks]

Reputedly MB developed their throttle pedal to be extremely lethargic at the request of taxi drivers who were keen to save fuel. Whether that is the true reason I don't know, but of the cars that were evaluated MB always felt as though the pedal was dead.

I don't think you intended this, but the word lethargic sounds like a time delay to me, rather than a straightforward low-sensitivity mapping. Whether intended or not, it is my impression that drive-by-wire cars *do* have a very noticeable time delay (whether that's for fuel-consumption purposes or not, I don't know, but a simple map takes almost exactly zero time to compute). In any case, this makes gruntguru's 1st gear idle test about 10x more difficult to pass - you keep pressing harder because nothing's happening, and by the time it eventually happens your brain can't correlate that to what you did to cause it.

That's one of the biggest differences between my FR-S and my '95 M3: the M3 instantly responds to what my right foot does.

[Greg Locock]

Bad choice of words then. I'm talking about tip-in performance feel. The MB gives the perception that you have to bury the pedal in the carpet to get it to move briskly. That's fine by me, I'm always happy to turn the volume knob up to 11.

[jm70]

Since the advent of drive-by-wire people have been crying out for improved throttle response. Due to the fact that many of these modern drive by wire car have brutal driveability issues for so many folks, the Sprint Booster product was conceived, developed, and born as a simple solution that anyone could install.
Sprint Booster team developed and patented their product, investing a great deal of time and resources in R&D. As more and more people tried the product, they have reported back valuable feedback to make the product the fully realized solution it is today.

A lot of people have misconceptions on how Sprint Booster modules operate. We admit, we believed this product to be somewhat silly when we heard what it proposed to offer. Like many owners who modify their car, you are probably sceptical as well when it comes to a product that makes the claim "to fix or improve" throttle response on a car like your BMW.
In fact, Sprint Boosters have been more popular in the Euro car market than any other (BMW, Porsche, Audi and Mercedes Benz).

Sprint Booster is a simple plug and play unit that allows the end user to adjust the throttle pedals response and curve, to better match how the driver feels the engine should respond.
Before the advent of Drive-by-wire, a metal cable offered a direct connection to the engines throttle-butterfly (a valve that adjusts how much air is ingested). As modern technology has demanded a more flexible electrical system, and more compact engine bay drive by wire and its non-linear behaviour was implemented. Believe it or not, many manufactures believe that the delay and behaviour of there Drive by wire systems offer an "easier to drive" experience. However- if you own one of these, you know that these delays are very bothersome.

Sprint booster is the quick and easy fix and our customers have truly confirmed for us how much they enjoy the improvements! What a great way to improve your 5 6 or 7 series, a super quick DIY install can have you feeling a sporty difference in under 5 min

Quoted from an on-line ad.

Edited by jm70, 28 July 2013 - 21:18.

[Bloggsworth]

It's not just Focus, I find it very easy to accidentally generate wheelspin in a diesel Mondeo auto, which is both embarrassing and potentially expensive. Yes, throttle progression will have been exhaustively cliniced. I could look it up and find out more but then I'd have to kill you.

https://dspace.lboro.ac.uk/2134/4613

Also bear in mind the progression /may/ be adaptive on the Focus, (as shift schedules are typically) so if you are other users hoon around in the car you may find it tending towards sharper shifts and higher revs.

So far as modifying the signal goes, most throttle pedals have an analogue pot on the shaft and it is only digitised at the EEC, so a bit of op-amp trickery will give you the ability to roll your own.

Or change a linear to log or vice versa... Or leave it alone lest you void both your warranty and insurance!