View Single Post
      05-25-2012, 11:45 AM   #1
Drives: Cobb Stage 2+ 135i
Join Date: Apr 2011
Location: Roanoke, VA

Posts: 2,197
iTrader: (0)

Datalogging With Cobb

There has been a surge in members new to Cobb asking about how to datalog and graph the results. Here's a quick DIY on doing this. This is just to get you started on the right foot. There's a lot to learn and it's pretty fun doing so.
1) Setup the datalogging list or just use the Default list.
*RPM, Accel Pedal Pos, TPS ACT, Load Requested, Actual Load, Req Boost, Boost Mean Abs, Cyl 1-6 Timing Correction, Cyl 1 timing, Bank 1 and 2 lambda(AFR) and Boost
2) Find a long, straight, level and empty stretch of road.
3) COMPLETELY DISABLE traction control, if you do not do this you will have throttle closures
4) This step depends on how comfortable you are/how safe it will be to redline 4th gear. If you are comfortable doing that, then you should a 3rd-4th gear pull. If you are not comfortable at those speed(~130mph) then do a 2nd-3rd gear pull.
4.1) On the AP go to the logging menu and select datalog. The screen then should say something like "Press OK to Begin Recording"
4.2) Get your RPMs to about 2,500. Allow your RPMs to settle and then press "OK" on the AP IMMEDIATELY before going WOT.
4.3) Be sure to hold the Accelerator to the floor all the way, do not allow it to return at all.
4.4) Shift into the next gear and go all the way to redline and then let off the throttle.
5) Once you have your datalog recorded you need to connect it to your computer and use the AP manager to transfer it onto your computer.
6) At this point you can use Excel, Zasquatch's excellent Excel grapher( or Virtual Dyno(
6.1) Using VD setup up your car profile and then load your run. Select the icon in the lower right corner of the Dyno summary box and that will bring up the data viewer.
6.2) select the channels you want to view and review your data.

Here's a list and explanation of the BASIC(i.e. major) channels that should be logged.

How to read the Data: I'll go over the main logging channels and what you SHOULD be seeing.

AFR(lambda): AFR is the ratio of air to fuel as measured by the O2 sensors in the DP. AFRs should be between ~11.7 and ~12.3. You will see a spike during the shift but the AFRs should immediately settle back down after the shift.
Here's a pic of how AFRs should look

Boost: Requested Boost and Boost Mean Abs are what the DME use and are useful when helping to determine if you have a boost leak. Boost is more universally accepted amd understood. Boost the pressure in the Chargepipe as measured at the TMAP sensor. This will vary depending on what map you're running. On stage 2+ Sport you should be seeing ~16.psi tapering down to ~12.5psi. On stage 2+ Aggressive you should be seeing ~18.5psi in the mid-range tapering down to ~12.5psi near redline
Here's a pic of how boost should look

Charge air temps(IAT): IAT is what the Charge air temperature is. This will vary wildy due to mods(namely FMIC and if meth injection is used) and ambient temps/humidity. The lower the IATs the better.

Tming and Timing Corrections: Timing and Timing corrections are interlinked as timing corrections determines how your timing looks. As soon as you go WOT Timing will spike upwards until the TMAP sensor sees boost and the DME enters "spool mode" at this point the timing will take a serious dive. It"s when the timing begins to recover from this dive that you need to start paying close attention. Timing should steadily climb upwards through the rev range until redline. Timing corrections occur when the DME sees something it doesn't like from the knock detection system or IATs get too high. Timing corrections occur in increments of 3. A -3 timing correction will result in a 3 degree pull in timing(see pic). If you are experiencing constant corrections across multiple cylinders you need try a less aggressive map. Timing on Stage 2+ sport should be ~3 in the midrange climbing to ~7 near redline. Stage 2+ Aggresive should be ~5 climbing to ~11 near redline.
Here's a simple pic of timing correction in action

Here's a pic of a good timing curve

Load: Requested load is what the car wants to see/what power it wants it to make. Actual load is what "power" output the car is making. Load is largely determined by boost and when graphed the plot for Actual load and boost wil mirror each other. On stage 2+ sport the requested load line should be flat at ~172, on stage 2+ Aggressive it will taper from ~188 at redline to ~160 near redline.
Here's a pic of stage 2+ sport load

Stage 2+ Aggressive

TPS: TPS req is what the DME is commanding and Actual is what the throttle blade position is. Requested should be ~99% and Actual should be ~80%
TPS Actuall

These are just examples. Your numbers may vary slightly depending on conditions. The main thing is that all the curves should be smooth without sudden drops or spikes.

Here's a graph of it all together:

OK, so here are some of the most common issues encountered in Data logs. I've already discussed what these channels are above^ so i'll just be showing what the actuall graphs look like and just the important information. I'll be circling the problems(when applicable) in red.

Timing Corrections
This car is running the stage 1 Sport map on 91 ACN fuel. There are timing corrections throughout the whole pull and at times on all 6 cylinders. These corrections are repeated every pull.

This is how the timing looks on the same car during the exact same pull with all those corrections.

So, what to do about it? Well, That depends on the cause of the corrections. Consistant timing corrections isolated in 1 or 2 cylinders is most likely a result of a mechanical issue. Routine maintenance(plugs, coils, injectors, port cleaning) can usually take care of or greatly reduce coorections caused by this. However, corrections like those above are almost certainly caused by running a tune more aggressive than the fuel will allow. This can either be fixed by running a less aggressive map OR if you happen to be lucky enough to have access to e85 then you can begin experimenting with e85 mixes to eliminate the corrections.

This is the same car, running the exact same map in similar conditions on a ~80/20 mix of ACN and e85 and the corrections are gone and the tiiming is exactly where the DME wants it.

Boost Leak
A boost leak is, fortunately, very easy to detect. The DME is hyper sensitive to boost leaks and will usually throw a P30FF code. This is not a 100% guarantee that a small leak(or a leak at the right place) will trigger a boost leak code thats why looking at requested and actual load will give you a heads up if you believe you have an issue. This leak will appear as an inability to hit requested loads. This car is running the stage 1+ aggressive map and at times is 50 points below load target and 4+psi below requested boost. The way virtual dyno scales data doesn't show how big the difference between requested and actual really is. That's when paying attention to the actual values becomes important(or use excel for a more visual heads up.)

Throttle Closures
Throttle closures are usually caused by 1 of 2 things. Traction control induced or in response to overboost. The DME will use the throttle plate to trim the boost to requested levels and are a huge concern. If a large overboost is detected it will slam the plate shut. These are the throttle closres that you can feel. Keep in mind that the TMAP sensor in PRE throttle plate so that during the throttle close the boost will spike due to the restriction created in the charge pipe. Whats important is immediately before and after the throttle closure.
This is an example of a throttle closure trimming the boost a little. TPS actual goes from 81% to 72%.

Here is an overboost induced throttle closure. Boost mean overshoots Requested boost by about 1PSi and then the DME reacts by dropping the TPS to ~50%. This spikes the boost during the closure but as soon as the throttle plate opens back up the boost is dropped to just under requested.

Last edited by rader1; 07-20-2012 at 11:55 AM.
rader1 is offline  
Reply With Quote