Haudi S4

Tried it tonight, my car does the same thing.

AvramD

I'm serious, the car continues to accelerate. MPH increases. Momentum <b>NEVER</b> causes acceleration. It <b>only</b> maintains velocity, never increases it.

Momentum is "a body in motion tends to stay in motion, and a body at rest tends to stay at rest." There is no "a body in motion tends to speed up."

AvramD

I'm not sure what to make of that. As I said, it came pretty close to going away at around 1400 miles before they reset the throttle. I wonder if it really goes away, or if it just gets more tolerable.

I'm horrified at the suggestion of having to wait <i>another</i> 1400 miles, much less 2500, before this goes away. And even more horrified to have to wait that long to verify that it *doesn't* go away. That would be almost 10% of my lease.

egghead_S6

..I have some experience with closed loop systems on wind tunnel models. We use them to provide positional control and dampening to the suspension components and wheels as they are moving up and down on the rolling road plane during aero prove out. We also use them to control the rolling road plane section of the wind tunnel itself.

I have noticed this throttle behavior in my 2000 S4 and my current S6, and all Audi's for that matter. Not a terribly unnerving event; you can learn to anticipate which speed/gear/throttle positions are best for maximum power delivery. Just listen to what the car is telling you. Your throttle is fine. You need to adjust your program, not the car's.

AvramD

And if you can't, then again as below, I'd suggest you consider the possibility that your car doesn't have the problem we are describing.<ul><li><a href="https://forums.audiworld.com/s4b6/msgs/40053.phtml">https://forums.audiworld.com/s4b6/msgs/40053.phtml</a</li></ul>

CH S4

Got my S4 (2004 model) in early August and it has been running very well, for the most part. Have been able to run it pretty fast on the Autobahn (about 250 kmh)for a long time where it performed flawlessly. It is truly a huge improvement over the last S4 (stock and with the Abt chip). About a month after I got it I started noticing that shifting was getting rougher and not smoother. It has a problem with the revs hanging, especially between 1700-2300 RPM and only from 1st to 2nd and 2nd to 3rd. I have it at the shop now and they have not been able to find the problem. It is espcially annoying because at times the car accelerates when it should be slowing down (particularly when I downshift and it bangs into gear at 2300 rpm and keeps gaining speed..) So far they have replaced the electronic module for the engine but that did not do it. I hope that Ingolstadt can shed some light on the subject as I am not looking forward to driving the car in the winter if I can't control the engine speed and can't downshift without it slowing down. Also, I feel that it is putting a lot of wear and tear on the drivetrain... Anyone have any clearer idea as to what it might be?

mtu_audi

As far as torque management goes it can apply to either manual or auto. Since you know your engine torque output (because the ECU calculates it) as well as your expected losses through the drivetrain and your gear ratios, output shaft torque could be calculated by the ECU therefore you could limit engine torque if your expected output shaft torque is higher than some given limit. There can also be a sensor for feedback as to exactly what the output shaft torque which can then be used to compare what is expected and what is actual.

As far as octane is concerned, it can't me measured. It could be inferred though based on the knock sensors because if you are getting pre-ignition, it might be due to the octane in your tank. Therefore if knock is detected, engine timing might be retarted.

One thing that we may be unrealizing is that closed-loop ETC systems take advantage of things like air temperature. Therefore when we see decreased performance, we perceive it as something that is going wrong, in reality, it is just a reaction to environmental conditions. Closed-loop systems can take advantage of environmental changes. But remember that OEM need to be concerned with emissions and reliability of the engine. Closed-loop systems allow us to better manage these two factors as well as performance, so it is up to the OEM producing the ECU software strategy to try and balance these variables.

One question I had was are you concerned about how sometimes you seem to have more power than other times or is it that there is a delay in power output?

SelmerMk6

You might want to consider angular momentum and the principles of conservation of angular momentum. Perfect example is an ice skater spinning with her arms up in the air and extended. When she pulls her arms in and close she spins faster. Her angular velocity must increase to keep the angular momentum constant.

mtu_audi

Don't mind my other question in my follow-up to your other response. You are obviously concerned about the delay time there is between when you press on the pedal and when the throttle plate changes position. This I may be able to offer suggestions.

I can guarantee that VW/Audi has a requirement somewhere for their validation that states the maximum delay time to acheive a throttle position angle is 'x'. That total time though is a summation of the maximum velocity that the electic motor in the throttle can move, and the total time the processor can receive a pedal command and issue a throttle command. Now, you won't be able to measure how long the computer delay is unless you knew on the ECU which input pin is connected to the pedal signal and which output pin is connected to the electric motor on the throttle that provides the amount of voltage to move the plate. Then you would need to have a setup that could precisely start a timer when you issue a pedal command to this input pin and stop the timer when it detects voltage to the throttle. As far as how fast the throttle plate moves, you would need to know which input pins on the ECU receive the throttle position sensors so you could calculate the velocity. Anyways I'm starting to type what I am thinking off the top of my head. If you need any more information/clarification on anything I said, don't hesitate to ask. Actually, If I had a pinout of the VW/Audi ECU and knew the pedal spec and throttle position spec, I could measure it at work. Prolly not to practical though.

mtu_audi

Most ECU code is done in C. The compiler is chosen based on which kind of processor you are using. Motorola Power PC, ST Microelectronics, Infineon all make microprocessors that are used in ECU's.

And by the way, its not that the ECU guesses wrong, it is either doing something deliberately or is a malfunction or "bug" in the software.