For about a decade, automobiles and heavy duty vehicles have had controllers that are networked together, but initially the network was too slow for most data acquisition applications. With the adoption of the CAN bus within the last few years, the data from an in-vehicle network offer substantial benefits to the test engineer.

This article describes the advantages and disadvantages of both direct sensor data and network data, and also the issues that exist if both approaches are used simultaneously.

Acquiring Data Directly from Your Sensors

There are benefits of a dedicated system that acquires data directly from sensors:

• Parametric data is always available
• Generally there are no time delays
• Acquisition rate is controlled by the user

Unfortunately, installing sensors is time consuming, expensive and may irreversibly modify a vehicle. In addition, signal conditioning and data acquisition hardware are required along with compatible software.

In-Vehicle Network Data

Here are some of the benefits of acquiring in-vehicle network data from the CAN bus:

• Sensors, signal conditioning, data acquisition hardware and some embedded software are supplied as part of the standard vehicle
• The bus information is used to control the vehicle, some of it transmitting as fast as 250 samples/sec
• Time and money can be saved relative to using added sensors.

The in-vehicle network exists both to control and service the vehicle. As a result, a vehicle interface data acquisition tool has the lowest priority on the network. Here are some disadvantages of in-vehicle network data:

• Parameters are buried inside messages, so we need to know the message ID, parameter location, and scaling information, as discussed in our previous eNewsletter.
• Diagnostic messages must be requested, so we need to know how to do this.
• Many parameters are difficult to acquire without a special database, which may be hard to obtain.
• Controllers may not respond immediately to requests if they are busy, causing potentially significant & unpredictable delays.
• Sample rates may be too low to meet the our needs.

In summary, in-vehicle network data are generally a preferred choice if the parameters we need are available to us, and if they can be acquired at a sufficient sample rate. Analog data from our sensors are required if the above conditions are not met. An additional benefit of analog data is verification of the network data, such as during accident reconstruction work.

If both sources of data are utilized, technicians must also be careful to time align the data. One technique is to acquire a rapidly moving signal with both techniques (engine RPM or brake light on/off are good options), and then to align them by shifting the data with a Z transform.

Some reference links

Click here to learn how to simultaneously acquire sensors, GPS, and in-vehicle data.

Click here for information on HEM Data’s DAWN product to acquire this information.

Click here to learn about DAWN’s Diagnostic Vehicle Database Editor to create and import databases.