[rjahl]

I've been working with a program called Testo to monitor and create data logs from my car.

Testo is a Program created by Pheno and friends that uses Inpa Ediabas functions to retrieve data using our existing cables. If you have Ediabas running, Testo should install fairly easily. The following is a link to the download and instructions.

http://www.bimmerforums.com/forum/sh...ime-graph-view

My plan has been to use Testo to monitor and record engine performance prior to making planned incremental software changes. I've already learned a few items such as the Disa Valve cut out at high RPM, how air temp affects my full load ignition timing and how the Automatic transmission kills torque through shifts. I have some data that leads me to think that the DME might be doing some predictive torque control in low gears as well.

Yes, I have a spare DME and tools required to create my own tunes. I'm just very short on the knowledge and experience required.

If other members contributed logs from their cars along with a list of modifications we should be able to validate these changes through the logs. As an example, I've logged my car with an old air filter, then with a new air filter and found the new filter made a 5% difference in air flow. These were stock Mann filters but way too many miles on the old one. My butt dyno barely noticed the change.

I see this as being an opportunity for everyone to validate their new exhaust system or intake without going to a dyno. We are not getting rear wheel HP measurements but we can measure change. If enough people post logs we can develop some baselines.

Attached is a file used to define custom jobs. If you replace the CustomJobs.xml file in the Testo Config directory you will be able to run a Job called E90 for both the MSV70 and MSV80. I'm assuming that anyone that's been able to get Inpa running will have no problems managing the technical challenges.

I'v been running the E90 custom job from a rolling start to about 80 MPH allowing the car to shift by itself. Data rates are not really very good so I am going to try rolling starts in third gear and run until red line. This will keep the car in one gear longer and provide more data points.

[rjahl]

Following up on my first post,

The following are my latest data logs

[rjahl]

Not to confuse the issue but Testo can also provide us with nearly unlimited trouble shooting possibilities.

The following is a quick snap of a Vanos check where I compared the actual Vanos cam positions against the target set points. This log was kept small to keep the data rates high and I simply revved the engine a few times to see if the Cams could keep up. White lines is the Exhaust Cam Set point and the red line is the actual position. I have another chart for the Intake cam.

[Taskmaster]

When you have time, hit me up and we can lo my car. INPA works on my computer, but due to a driver issues, I can't use the USB to serial cable anymore.

[Mike.]

if anyone gets this working in the NY area, Im down to do some data logging

[hassmaschine]

Its raining so I'm driving the E90 - I'll try to get some logs this week.

[rjahl]

Originally Posted by Mike.

if anyone gets this working in the NY area, Im down to do some data logging

Testo is not the easiest software to get running. First you need to have INPA up and running and then Testo. Just like INPA nearly nothing works until you are actually connected to a car. You can put Tool 32 in simulation mode to help setup custom jobs but they will not run until you are connect to the car.

There is a version of Testo that's stand alone but I rarely use it.

[rjahl]

I finally started working on my own tune. Small adjustments to the full load valve lift, full load high rpm Lambda, Disa and Rev Limits. Nothing earth-shattering just a little at a time.

I can's say that I can feel any difference so far. Shifting seems to be quicker, there is a chance the soft limiter was kicking in before the transmission started to shift. My Automatic shifts somewhere around 7,000 RPM. I't hard to catch the exact number and my logging methods are not quick enough.

I'll try to log these changes later in the week.

I've come up with some theories on the Torque figures the DME are reporting. Is it possible these are Raw numbers before friction and pumping losses? If I use the Modeled losses stored in the Parameters section of the tune and subtract those figures form the reported Torque values the numbers start to make sense.

[hassmaschine]

the only thing I ever felt a big difference was by working on the ignition maps. altering vanos/vvl definitely made a small difference on the dyno (~8whp or something), but it was pretty hard to tell by the seat of your pants. DISA will mainly affect the transitions since the resonance frequencies are a physical property that you can't alter.

I wouldn't mess with the lambda much - it goes quite rich mainly to protect the cats in the last cell of the curve. I leaned that out a little bit, but adding fuel anywhere else just lost power immediately.

I think the biggest potential is in the overbearing DSC/ASC and for the automatic guys, possibly shift points and torque reductions during shifts (if that can be altered on the DME side). Also, the soft limiter is pretty weak as well.

[rjahl]

I wonder if we can use the exhaust sensors to monitor the cat temperatures and find the actual limits. In any case I made small changes in the last two cells in the Full load Lambda Tables. Do you think these were too much?

RPM Original Change
5504 0.901 0.910
6496 0.871 0.898


RPM Limits were raised by a modest 50 RPM

Disa higher engine speed threshold raised by 130 RPM

Getting control of the traction, stability and torque reduction control systems might put a lot more FUN back into our cars. Did you every try messing with the "Torque Reduction at high speed map"?

[hassmaschine]

Yes. I zero'd it out.. I can't tell if it does anything though.

[rjahl]

Well,

I logged the car with the modified Intake valve Set Point(Full Load). Set Point tables were increased from 9.7 to 9.9 mm.

I recorded no noticeable change in air flow on the engine. None. A new air filter was more noticeable.

I was not expecting much but I was hoping to at least see a numerical change.

[Taskmaster]

Originally Posted by rjahl

Well,

I logged the car with the modified Intake valve Set Point(Full Load). Set Point tables were increased from 9.7 to 9.9 mm.

I recorded no noticeable change in air flow on the engine. None. A new air filter was more noticeable.

I was not expecting much but I was hoping to at least see a numerical change.

Nowhere, not even at 2000rpm?

[rjahl]

Originally Posted by TheAxiom

Nowhere, not even at 2000rpm?

I'll take another look at that speed, It's tough for me to get good data that low. Although, I just started using a strategy that gets me better data at lower RPM's. It takes a little more runway, LOL but doing the full run in third gear gives me a nice chart to work with. The problem is the car is going a little too fast by the end of third, certainly north of any speed limit around here.

[rjahl]

Is anyone able to provide a better German translation to the following fields?

I'm trying to understand what the actual values the DME is reporting for torque. Nothing seems to work back into the Torque tables in the DME or even the torque values expected for the N52.

Better yet, Does anyone know if the DME will is capable of reporting improvements in the tune that does not directly affect airflow or fuel such as ignition timing? I know the power modeling is pretty good but just how good is it?

[Taskmaster]

Originally Posted by rjahl

I'll take another look at that speed, It's tough for me to get good data that low. Although, I just started using a strategy that gets me better data at lower RPM's. It takes a little more runway, LOL but doing the full run in third gear gives me a nice chart to work with. The problem is the car is going a little too fast by the end of third, certainly north of any speed limit around here.

Highway

[rjahl]

Top of third takes me over 80 and the Hiway patrol is blitzing my area lately. I like the southbound stretch of 275 just after the 75 split. It's flat straight and the traffic normally runs fast...

[Taskmaster]

Originally Posted by rjahl

Top of third takes me over 80 and the Hiway patrol is blitzing my area lately. I like the southbound stretch of 275 just after the 75 split. It's flat straight and the traffic normally runs fast...

Hmm, fair enough. I use 75, but that is completely opposite end of town for you. Back roads might have to be an option.

[mmmnot]

Originally Posted by rjahl

Top of third takes me over 80 and the Hiway patrol is blitzing my area lately. I like the southbound stretch of 275 just after the 75 split. It's flat straight and the traffic normally runs fast...

Have you considered investing 50 bucks or so and doing these runs on a dyno for a proper 4th gear pull?

I love what you're doing though - just figured that would be the most controlled testing environment.

[Taskmaster]

Originally Posted by mmmnot

Have you considered investing 50 bucks or so and doing these runs on a dyno for a proper 4th gear pull?

I love what you're doing though - just figured that would be the most controlled testing environment.

Probably not the best for what he's doing. He needs a proper baseline yes, but for those small changes, it doesn't make sense to dyno each one until he's done.

[mmmnot]

Originally Posted by TheAxiom

Probably not the best for what he's doing. He needs a proper baseline yes, but for those small changes, it doesn't make sense to dyno each one until he's done.

Fair enough

[rjahl]

I still have far too much to learn before thinking about a dyno run. I'll take this in small baby steps. Change one parameter then check the the affects.

I'm just an amateur so don't expect perfection or fast progress from me alone.

As mentioned by Hass, Two of the biggest gains will be item that can't be measured on a dyno. Traction control and torque reduction at gear shift. These are huge hits.

I am also beginning to think that AITs are something that can not be simply factored on a dyno run. I don't have solid evidence yet but it seems like the DME pulls away more timing when it's hot. I live in Florida, so "hot" is not uncommon