Measuring on CAN bus

Subjects:

  • General
  • Multimeter measuring
  • Oscilloscope measuring

General:
If there is a suspicion that there is a malfunction in the CAN bus, a diagnosis can be made by, among other things, measuring the voltage levels on the wires. The content of the CAN bus message is not yet important in the first instance. We can take measurements on the CAN bus wires with both the multimeter and the oscilloscope. The measurements with the multimeter do have a limitation; only an average value is indicated when measuring the voltages. The multimeter is sufficient when measuring an interruption or short circuit. The oscilloscope is required to measure the voltage levels.

How a CAN bus system works and how the structure of the messages is established is explained on the page CAN bus.
This page focuses on measuring the CAN bus with the digital multimeter and the oscilloscope.

Measuring with the multimeter:
The digital multimeter can be used to easily measure voltages or ohmic resistance. This section describes both measurements with possible measurement results.

1. Measuring Voltage Levels:
The multimeter must be correctly set to DC (direct voltage) and to a measuring range of 20V. Various situations can arise on the CAN network that must be taken into account when measuring.

  • Recessive when no messages are sent:
  • Alternately dominant and recessive when messages are sent continuously
  • The problem with this measurement is that the CAN bus never has a constant dominant signal. It is always alternating, so the 2nd column is closer to reality. Because a digital multimeter is a lot slower than the speed of the data traffic on the bus, the multimeter measures an average voltage. The average voltages for constant data traffic will therefore be:

In practice, there is never constant data traffic. One moment slightly more data will be sent than a moment later. For example, when more or less busy, the voltage: CAN-High in relation to ground will be 2,8 one time and 3,3 the other time. This also depends on the quality of the multimeter. A low-quality multimeter is often slow and then you are at the average voltages of the last table. With a very high quality multimeter you will come closer to the 2nd column; the refresh rate of the displayed voltages will be much higher.
This does indicate a guideline. If the CAN-High differs by a number of volts compared to ground during the sending of messages, then something is wrong. Therefore, we situate here below a number of error scenarios to learn to recognize the causes;

CAN-high shorted to ground:

  • CAN-high to ground: 0 volts;
  • CAN-low to ground: 1,5 – 2,5 volts;
  • CAN-high vs. CAN-low: between -1,5 and -2,5 volts. 

CAN-low shorted to ground:

  • CAN-high to ground: 2,5 – 3,5 volts;
  • CAN-low to ground: 0;
  • CAN-high versus CAN-low: between 2,5 and 3,5 volts. 

CAN-high shorted to plus:

  • CAN-high to ground: 12 volts;
  • CAN-low to ground: 1,5 – 2,5 volts;
  • CAN-high versus CAN-low: between 9,5 and 10,5 volts. 

CAN-low shorted to plus:

  • CAN-high to ground: 2,5 – 3,5 volts;
  • CAN-low to ground: 12 volts;
  • CAN-high vs. CAN-low: between -8,5 and -9,5 volts. 

CAN-high shorted with CAN-low:

  • CAN-high relative to ground: dependent on CAN transceiver;
  • CAN-low to ground: dependent on CAN transceiver;
  • CAN-high vs. CAN-low: 0 volts. 

2. Measuring Ohmic Resistance:
The measurements below show the ohmic resistance in three different situations: with a correctly functioning system, an open wire and a short circuit between CAN-high and CAN-low.

Interference-free:
On the page CAN-bus it is described that there are two terminating resistors in the network. The terminating resistors both have a resistance of 120 ohms. In a fail-safe system, we will measure a 60 ohm replacement resistance between CAN-high and CAN-low.

Note: we can only measure this if the power supply of all control units is switched off!

Interruption:
In the event of an interruption in a CAN-high or CAN-low wire, we no longer measure the replacement resistance of 60 ohms. In the picture we are only measuring the value of resistor R2 (120 ohms).

Short circuit:
In the situation where the CAN bus wires connect to each other (ie are shorted to each other), we measure a resistance value of approximately 0 ohms.

At the next fault, both CAN wires are interrupted. There will now be a lot of interference (noise) on the bus. Nodes 1, 3 and 4 can communicate with each other provided the interference and reflection is too great, causing the messages to distort. Similarly, node 2 and 5 can communicate with each other subject to the same problem.

Some CAN networks also function when one wire is interrupted. Error codes will be stored and the driver will be informed with warning lamps by messages from various systems. These are the networks equipped with a Fault Tolerant CAN transceiver. Depending on the transceiver used, different types of errors can occur without loss of communication between the nodes. These CAN transceivers can also function normally with the aforementioned faults with the short circuits to plus and ground (of course with various error messages).

Measuring with the oscilloscope:
In principle, searching for faults on the CAN bus with the scope is the same as with the multimeter; the voltages tell where the cause lies (short circuit with plus, ground or with each other). The outcome of these measurements can therefore be compared with the scenarios mentioned earlier when measuring with the multimeter. The CAN-High and CAN-Low signals can be viewed very nicely graphically on the oscilloscope.

This image shows a fully functioning CAN network. The CAN-High signals go from 2 to 3,5 Volts and the CAN-Low signals from 2 to 1,5 Volts. The receiving node looks at the dominant and recessive states; if the voltages of CAN-High are 3,5 and CAN-Low 1,5 volts, the bus is dominant. That means a 0. When both buses are 2 volts there is no bus activity and the node sees that as a 1. A 1 means recessive.

error: Alert: Content is protected !!