no way to compare when less than two revisions
Differences
This shows you the differences between two versions of the page.
— | can_faq:can_bus_off [2018/08/21 23:08] (current) – created - external edit 127.0.0.1 | ||
---|---|---|---|
Line 1: | Line 1: | ||
+ | ==== CAN Bus-Off condition/ state ==== | ||
+ | |||
+ | BusOff is an error state of the CAN Controller. Only the transmitter can switch in the state BusOff, if the //Transmit Error Counter// exceeds 255. | ||
+ | |||
+ | (more details needed on this place) | ||
+ | |||
+ | In real life a CAN controller can switch in the mode ErrorPassive sporadic. But the mode BusOff is a critical error and should never arrive without a hardware error or an other critical error. | ||
+ | |||
+ | ---- | ||
+ | '' | ||
+ | |||
+ | there are 2 different ways to recover from bus off : | ||
+ | * automatically after 128 occurrence of 11 consecutive ' | ||
+ | * A node can start the recovery from »bus off« state only upon a user request. | ||
+ | |||
+ | That's right that we can find on the market these 2 ways according the used CAN core. This difference must be taken in account to have the same behavior in drivers and network management according the choosen CAN controller | ||
+ | |||
+ | ---- | ||
+ | |||
+ | '' | ||
+ | |||
+ | There is actually no difference. | ||
+ | |||
+ | If that were not the case then a node with a defective receiver would try to send, go bus off, then go bus on again and try to send thereby | ||
+ | using 100% of the bus. The 128 * 11 makes it possible for 128 other messages to sneak through while this defective node goes through its bus | ||
+ | on/off cycle. | ||
+ | |||
+ | The interesting thing about handling bus off situations is that on a wired CAN bus, unless there has been a major bus disturbance, | ||
+ | off situation almost never happens. | ||
+ | |||
+ | ---- | ||
+ | |||
+ | '' | ||
+ | |||
+ | The ISO states that auto-bus on is not required, i.e. devices with auto-bus on feature are not 11898 compliant. We had this discussion with the Motorola TouCAN (which is still not conformance tested) and other CAN-contoller-cores. I heard also from several sources from CAN-industry that auto-bus-on is not wished very often. It makes more sense to let the application decide what should happen. Especially in the automotive-sector the auto-bus-on-feature is very negative. | ||
+ | |||
+ | '' | ||
+ | |||
+ | I would support Anastasios. We always suggest our customers, in the case our driver reports the BusOff situation to the application, | ||