Create VCDS log files and prepare them for the forum. (Articles)

When diagnosing TDI problems, log data is often essential.
With VCDS (VAGCOM), VAG drivers can access a high-quality tool that, thanks to continuous log sequences of diagnostic data, provides a significantly better insight into engine operations than the original VAG workshop equipment, which can only document a snapshot (without the data before and after).

Given the large number of available data block (MWB) options, or extended data block options, it can be difficult for some drivers to identify the most appropriate MWB(s) to use when troubleshooting problems, and to prepare the logs in a way that makes analysis and assistance from external parties easy.
You can refer to the following tips for this.

Before, here are 2 hints:

The engine behavior was programmed for normal operation. Only there can high engine loads (full throttle) be recorded for a long enough period at any speed in order to be evaluated as log data. Logs when the system is idle are usually useless, except for idle-state monitoring and similar data.

The load pressure data is provided by the diagnosis as absolute values, including, for example, an ambient pressure of approximately 700 - 1000 mbar, depending on the weather and altitude. For the colloquial overpressure (as overpressure!), you must therefore subtract the ambient pressure from the overpressure values.


1. Limiting to the truly necessary measurement values

VCDS (VAGCOM) allows the simultaneous logging of 1 to 3 MWB, i.e., 4 to 12 individual measurement values. Newer control units offer so-called "extended measurement blocks". Which one is used depends on the specific control unit. If no MWB are available in the specific control unit, you need to select the required measurement values in the "extended measurement blocks". This selection can be saved and reloaded to avoid having to repeat the search process each time.
Since the data transfer rate from the MSG to the laptop (sample rate) can usually not be changed, the data transmitted within a certain time period can either be...
a) Many MWBs are distributed, but only a few individual values are obtained per measurement size.
b) distribute fewer MWB, while obtaining more individual values per measurement. Due to the higher temporal resolution of the data series, for example, Accurately trace regular processes.

As shown in the "MWB Overview" sheet in the appendix, several measurement values appear in multiple MWBs: For example, the measurement values of the classic "air mass flow meter" can be found in MWB 3 and 10. Otherwise, these two MWB are fundamentally different:
MWB 3 provides all the data for the AGR system, namely the engine speed, target air mass, actual air mass (which is the actual value from the air mass meter), and the control input via the electromagnetic valve's throttle position.
As long as only the air mass is to be controlled instead of the AGR function, 60% or 80% of the data is redundant (timestamps, target air mass value, AGR ratio, and optionally, RPM)!
For this, the volumetric flow rate must always be considered when examining the air mass.
MWB 10 provides an overview of the most important "air parameters," such as intake pressure, air mass, ambient pressure, and the throttle position, but without the engine speed.
If you log MWB 10 together with MWB 8 (which provides the most important volume limitations and RPM), you have all the important size data for an initial analysis in case of performance issues, without unnecessary data and with the highest possible sampling rate.
Furthermore, the "simultaneous" data for air mass and boost pressure in the MWB 10 are best suited to provide clues about a faulty turbocharger or leaky or partially blocked air/ Obtain exhaust gas pathways.

Standard objection from new VAGCOM/VCDS users:
"But MWB 3 actually provides the target air mass flow! If that is achieved, it automatically means that the engine is not having any air flow problems!" "If it's not achieved, either the LMM is faulty, or there's a lack of boost pressure!"
Incorrect! The air mass value in the MWB 3 is only relevant for the AGR function and is typically emitted in most 850 mg/stroke TDIs at full throttle.
However, this is not enough to overcome the haze in engines with high specific fuel consumption (ASZ, ARL, BPX . . .), meaning that they can still show a loss of performance despite reaching the target air mass!
Conversely, significantly less than 850 mg of air/stroke is sufficient for the 1.9l-90PS variants to achieve full power.


Therefore, it is advisable to log as few MWBs as possible, but to ensure that all the necessary measurement values are available for analyzing the specific problem. The information in the MWB overview (useful for . . .) can be used as a guide.
Are there "Turbo" buttons available in the measurement blocks? If so, please activate them! For extended measurement values, please use the grouping function!


Apparently, torque specifications are irresistible to many drivers. However, these do not provide the actual torque available for performance, but only the calculated internal engine torque, without deducting for friction losses, etc. Therefore, maximum values far exceeding the prospectus data do not necessarily mean that you have a particularly powerful engine, but only that you have logged diagnostically useful data, which are largely useless for error diagnosis.


2. File format .csv or .xls

VAG-COM/VCDS writes log files in the space-saving .csv format. These files can be analyzed directly with KDataScope without any further conversion. KDataScope is a VCDS Log Viewer and is included with every diagnostic system from our automotive diagnostic shop.
.csv files, please always upload only as they are provided by VCDS, meaning do not modify the file in any way. Will the file be converted to Excel/OpenOffice Calc, please save it as .xls? Everything else results in a meaningless jumble.

The recommendation is to use .csv files, so no conversion is needed!

If you want to collect and compare values from one or more log files, Microsoft Excel can also be used as an editing tool.
When working with Excel and saving, column separation formatting is often lost, and the next time you open the file, you'll see the data from all the columns crammed together on the left side as a jumble of numbers.
This can be painstakingly repaired using the "Data -> Text to Columns" menu, but the more time-saving and safer method is to save the .csv file as a .xls file after opening it in Excel, which can then be processed with comparatively less risk.

Alternative: convert the log file to .xls using a CSV converter (which can be found in newer VCDS versions after installation in the VCDS directory).

Will the fresh .csv file be opened automatically the first time you click on the number grid in the specified format? This can also help you open the .csv file from Excel (using "Open File"): Sometimes the data appears (for some reason) neatly organized in separate columns, as expected. Then immediately save as .xls, see above!


3. Clear Definition of Measurement Column

Depending on the used label file (if any), the measurement column in the first lines of the log file may be ambiguous, missing, or in English, with "automatic guesses" from VAG-COM, VCDS, or also with correct headings.
If the headings don't match, you should label the columns according to the MWB overview in the appendix: this significantly simplifies the analysis of the log, as not everyone knows, for example, what the temperature was in MWB 7 according to the MWB overview.


4. Focus on important passages in the logs

Generally, it is important that the logs should accurately represent the specific problem , for example,
-> in case of performance issues, run the affected speed ranges at full throttle in one go
-> In cases of errors in the loading pressure rule, it should be possible to trigger as many as possible in sequence.
...and so on.
Furthermore, it is advisable to remove longer sections of the log that do not contribute to clarifying the problem (e.g. Phases of the "Performance Deficiency" Problem:
Please do not record any nonsense instead of the operating conditions that should be investigated!

The "Log Example" document in the appendix shows two logging trips with the Ladedruck MWB 11, with the vehicle 1 additionally using the MWB 8.

With Wagon 1, the significant full load range was almost exclusively utilized during the log.
The log of Wagon 2, on the other hand, includes a leisurely ride with short periods of full load, but only the latter are relevant here.
In this vast amount of data, they can only be found through painstaking searching for LD-target values above approximately 1500 mbar: a few clicks on the "Stop / Continue" log button would have made the results much more manageable. Of course, you could also delete the partial load and thrust phases retrospectively, but here they remain as an example in the log to deter further errors.

Content of Example Logs:
Vehicle 1 exhibits significant over- and under-boosting fluctuations in the boost control system, which, in combination with a worn-out MAF sensor, also causes the turbulator to fall below the torque limit.
Vehicle 2 exhibits significantly more precise boost pressure control.

By analyzing the additional calculation columns K and L, it is possible to easily identify loading errors and their consequences for the limits. To make them easier to find, passages of particular importance in Vehicle 1 are highlighted in color.


5. Represent data in charts (optional, use KDataScope instead)

The logged trends of individual measurement values are best understood when they are presented in graphs. Therefore, the following form should be aimed for: either only the diagram is made available online, or the log file with the integrated diagram.

With VAGScope, you can create charts, but the labeling options are limited compared to Excel charts, for example.
Those who have access to Excel or similar software and are familiar with charting functions (such as color selection, grid lines, axis formatting, legends, etc.) can therefore create the "most appealing" charts.
An example is shown in the "with diagram" sheet. There, the log data from the log example is presented graphically.

Due to the un-trimmed walk of Wagen 2, the diagram initially appears confusing and confusing.
If you activate the "Vehicle 2" filter in column M using the dropdown menu, and the "important" filter in column N, then the chart will only display the passages with full load, and it will become significantly more informative immediately. Here's an approximate representation of how a log diagram might look when inviting helpful people instead of driving them away (see above, under No. 4: Deleting unimportant passages).

If the "important" data for both vehicles is displayed (select column M "all"), the diagram shows the charging pressure settling behavior of both vehicles in a compact form, including the limits for vehicle 1. Whoever has the data from a comparable vehicle should approximately enter it into the log in this format.
(Note: The filters are used here only to illustrate the difference between a clear and a cluttered diagram.) If logs are posted on the forum, longer passages containing "non-essential" data should be deleted from the beginning.

As an Excel chart type, "2-Axis Line Chart" is generally recommended, as the value ranges of different measurement variables are often different. Having 2 scales is very helpful in this case.

If the desired and actual values of a measured quantity (air mass, spray start, pressure from MWB 3, 4 or 11) are shown, they should definitely be assigned to the same axis. From the trajectory of the two curves, you can immediately identify the rule errors, instead of having to calculate and search for them in the table.

To fully utilize the diagram area, the Y-axes should be manually scaled to approximately the range between the maximum and minimum values of each log .
For example, in load pressure logs, values typically do not appear below 900 (rpm or mbar). Therefore, the left Y-scale in the appendix begins at 900. The right Y-axis ends at 76, as this is the maximum value for all data on the right axis.
By choosing appropriate intervals, the axes can be aligned so that the grid lines also have corresponding values on the right scale.


At MWB 11-Logs , the numbers representing the key ratio are often "upside down" for many MKB, meaning that interventions by the MSG to increase the loading pressure result in a lower key ratio, and vice versa, which makes it difficult to understand the control sequences.
Then the charts become more informative when you create a calculated column "100 - Tastature Ratio" and use these values instead of the logged data in the chart (see: Column K of the "with diagram" sheet.

If the effect of the intercooler (IC) needs to be investigated, then at least the MWB 7 and 11 must be logged, because without engine speed and boost pressure, no assessment is possible.
However, driving speed plays a relatively unimportant role in everyday situations - unless, for example, you are driving a heavy trailer up a steep mountain road at full speed.
Since the LLK has a certain heat storage capacity, the LLT changes more slowly than initially expected. In addition, the LLT is generally at its highest level in the Pmax range. Therefore, only logs in which the engine operates at the slowest possible speed, i.e. at high RPM (at least 4. (Accelerate) and rev the engine to at least 4000 rpm.
As a reference, the ambient temperature is always required, as it forms the basis for the heat dissipation of the LLK. In addition to (or instead of) the log data from the real LLT, a calculated column "LLT Log value minus Outdoor Temperature" should be included in the log file or diagram (Column H in the LLT sheet of Annex, abbreviated LLT net)!
The actual LLT (suction hose temperature) is actually only needed to discuss potential thermal problems or performance degradation due to excessively high LLT.


If irrelevant/unnecessary data is logged due to an MWB selection (e.g., ambient pressure and gas inlet position for MWB combinations 8 and 10), then these should not be displayed in the diagram (-> delete the data series).