How much of a performance upgrade do these mods net you?

Cost?

Does the mfgr warranty get voided?

thanks!

No headers. Car already has great headers.

Catback has been independently dyno'd for a 20hp gain, but this will only happen when all the input parameters are perfect in the eyes of the ECU. Downpipe gains have never been independently tested to my knowledge.

If you throw an engine code, and you have DP's, chances are the warranty won't cover it if it can be linked to anything to do with the oxygen sensors, either directly or indirectly. No problem with cat-back.

Low end gains for non-res catbacks and downpipes is about 25 lb-ft of torque between 2500 and 4500 rpm. This makes a significant difference in driving, providing some good torque down low. At the top end the conditions did not allow me to accurately dyno. The ECU was dialing everything back (stock and with Millteks) and closing down the intake valve in the air box. If I were to guess, I'd guess a 25 to 30 hp gain at top end under best conditions with the catbacks and downpipes.

I personally like the sound that they make with the downpipes. Others don't. It's a matter of personal preference.<ul><li><a href="http://www.stratmosphere.com/milltek_rs4_exhaust.htm">Stratmosphere</a></li></ul>

The cost is high (check the Stratmosphere website per the link below). However,the sound is wonderful, especially at WOT under load, yet fairly civil around town. Others have documented gains on dynos; I know my car had a little more power and torque on the track against stock RS4s.

One added benefit - when I fired my car up after dinner at a local restaurant last night, it set off a rather obnoxious alarm system on the Honda Element parked next to me. People stared, but I snickered as I drove away.<ul><li><a href="http://www.stratmosphere.com/milltek_rs4_exhaust.htm">http://www.stratmosphere.com/milltek_rs4_exhaust.htm</a</li></ul>

Milltek (Stratmosphere)
Magnaflow
Tubi
GMG

Milltek are the only DP's though.

i've been meaning to ask ELEVENS about it...not much feedback from him, other than the initial report.

...regarding cost, basically the range is under $1000 for the Magnaflow (some of us got it for $750 shipped, but I don't know if that discount is still available) to (alledgedly) $3000 for the Tubi. Add another $2200 for the Milltek downpipes. To my knowledge, no one has demonstrated a performance advantage for any of the exhausts mentioned, so it boils down to sound quality and tip design for most people.

The only portion of the warranty that gets voided with a cat-back is the warranty on the exhaust itself. With downpipes, the O2 sensor warranty would be affected also. This is a gray area, so don't quote me.

I want to bypass the ECU and put in a switch to control the flap directly, so that I can do comparisons with VAG-COM and on the dyno. It looks like the best approach will be to purchase a second flap that I control, while leaving the solenoid on the 1st flap connected to the ECU, so there are no ECU fault codes.

I've been told by a tuner that normally the best way to proceed is to keep the stock solenoid installed, but disconnected, and place my own solenoid in the circuit, so that the ecu doesn't know the difference. I can neither confirm nor deny this.

It's the solenoid that the ecu knows about. For testing, I want to be able to control it to open and close the flap, so that I can test the open and closed condition in real time.

over time as a result of the ECU adjusting down for the exhaust? I have heard some rumors of initial gains on the dyno but once the ECU adjusts you might actually lose power? Has anyone done a follow up dyno to test power under the same conditions?

Personally, I believe that it's unlikely that the ECU adjusts down to remove power. What was posted was that the ecu will retard timing (if I was reading between the lines correctly.) As a control on power, that just does not make sense. It would be much simpler for the ecu to control the throttle body, or the injector pressure and timing. Normally timing is retarded in response to the ping detectors.

...and he had significant gains with just a cat-back. Perhaps he can clarify.

after several months of driving to see what the result is with the ECU being fully adjusted.

we are all just guessing. For example, is there a torque sensor? Is it maxed out at 317lb-ft or is it more. Maybe it is more at peak, but not so at other points in the power band. Or vice versa.

Repeated dyno tests over varied time intervals may provide a clue, but you need very similar air temp and humidity conditions. Even then, up to 15 hp variation between runs means nothing.

If anyone here knows anything about statistics, and about tuning cars &amp; dynos, you also know that to get any meaningful data on this, it would take multiple dynos by VIRTUALLY EVERY OWNER of a particular exhaust brand to give you a conclusive result. Take 200 RS4's all with the same configuration of Milltek exhaust. Dyno them all 5 times each on the same dyno in similar climate conditions, with each run 2 weeks apart. Now take the average hp gain of all the test results and you will have a conclusive result.

But my solenoid is still connected to the vacuum line, it just doesn't have any flap to control!

But I think it's hurting my gas mileage. I'm constantly giving it near-full throttle up to 4-6K in 1st few gears up to 40-50mph from traffic lights just to, well, as Enzo Ferrari once said, "hear the music of the noise"!

There is no torque sensor. "Actual torque" is a calculated function in the MED 9 ECU.

<img src="http://www.teraspeed.com/RS4/sensors.jpg">

<img src="http://www.teraspeed.com/RS4/actuators.jpg">

This might have been the case on older ECUs, but with broadband O2 sensors, almost all adaptation is real time. Air fuel ratio is adjusted on a per crank rotation, per cylinder basis. There are long term adaptations to compensate for air leaks in the intake system and degradation of mass air flow sensor.

There is also a long term adaptation for timing, which is cranked back every time there is a knock. That specific adaptation has a very short time constant for retard, and a very long time constant for advance. As soon as you get bad gas, or run at high ambient temperatures and knock occurs, timing is retarded instantly, and the adaptation constant is changed. It takes a very long time for advance to occur.

If you want full engine power, reset the ECU to reset the timing adaptations. The ECU will adjust air/fuel ratio in real time.

It's available on Bentley.

Audi of America, Inc.
Service Training
Printed in U.S.A.
Printed 07/2006
Course Number 921603