Re: CANopen and the need for time-stamping
Heinz-Jürgen Oertel <[email protected]>
| Newsgroups | gmane.comp.hardware.bus.can |
|---|---|
| Organization | privat |
| Message-ID | <1975370.KhLKBFvDlb@heinz> |
Am Donnerstag, 23. Oktober 2014, 13:30:04 schrieb Janakiraman, Hareesh:
> Hello,
> Some CAN implementations have a built-in timer for time-stamping of messages. These timers also provide a time-out interrupt should transmission or reception of a mailbox does not occur in a pre-determined time. The eCAN module of in the 28xx MCUs from TI is an example. My question is this: Is the presence of a built-in timer absolutely essential for CANopen implementation? (Note that I am not referring to timers that are external to the CAN module, since all MCUs have one or more of “general-purpose” timers). I believe the answer is “no”, since I see in the websites of many 3rd parties who develop CANopen stacks a list of processors and I know not all of them have a built-in timer/counter as part of the CAN IP.
>
> Hareesh Jana
Your guess is correct. Such a timer is not absolutely necessary. Most applications live without it. If needed the CANopen application can use a normal timer (and of course a piece of software) to monitor receive or transmit time-out conditions. But having a hardware support in the CAN controller is great.
--
mit freundlichen Grüßen
Heinz-Jürgen Oertel
emtas GmbH
signature.asc
(application/pgp-signature, 490 B)
-----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.22 (GNU/Linux) iQEcBAABAgAGBQJUSW2KAAoJENViPQy9fcm71/0H/AxM/jgRLBHe+KEzOiAbwhbh R9cx90SYFiXj3MQlVoaWMnd9Hr1PqEWLLjbTk9pFI18zcGIXzgo/BwccHmuomyAt P9W+ZSdrfC1ncobIF9XNBfrmiBFK5X+AEy7OS9sKKDRm0zAUauwq0qJOvEPYrjV/ 7kz41zCTWCKFHvZ06FULJl0OGxCDVrkSjBXkZ8skICmDI5S+RZvy8FFkM/vYuxB8 1UR9VhDzuEWNxO75/OwZKB+Gjd/Ytqkm5OyU8aOQ5rfSw3YuFJMI+ZnRHz0j962J 4SGf2RbkTDtXXPY69ttoTQa4nJ6JCDzeuO0PO0YWGXDUfs8Kr7peS+GHqLeOcBQ= =IdVH -----END PGP SIGNATURE-----