forum.milcan.org

Specs - General :: RE: MILCAN_software

Author: juergen.klueser
Subject: Re: MilCAN software
Posted: Fri Mar 28, 2008 4:15 pm (GMT 0)
Topic Replies: 3

Hi Alberto,

for Libraries I cannot answer, but for tools. The Vector CANalyzer and CANoe tools have an option J1939 with a similar message layout as MilCAN. For this option there exists a free adaptation for MilCAN.

There's a white paper showing this. You can get it at http://www.vector-worldwide.com/vi_downloadcenter_en.html
For product select "CANoe"
For type select "Application Note"
For standard select J1939

Press search and you will get "AN-ION-1-3501 How to use CANoe.J1939 with customized protocols"

Regards
Juergen

Specs - General :: RE: general_questions

Author: peri
Subject: Re: general_questions
Posted: Sat Nov 11, 2006 4:04 pm (GMT 0)
Topic Replies: 1

Alberto wrote:

Hello everybody,

I know that MilCAN is intended to be compatible with CANopen and SAE J1939 but what is the fundamental difference between those protocols, I know that is for militar purpose but is it more secure or faster ?

Another doubt I have is, what is the duration of one slot ? Is is the same duration as in CAN protocol ?
How is implemented synchronization in MilCAN, does it use any special hardware to implement it ? a clock ? or is it implemented by software ?

Thank you very much

Regards

Alberto

CANopen, J1939, and also MilCAN are higher layer protocols that operate on-top of CAN (not replacing it). They add functionalities to a CAN network in order to satisfy the requirements of their target operating environment (CANopen for industrial process control, J1939 for vehicle systems communication and diagnostics, MilCAN for military vehicle subsystem communication).

The purpose of MilCAN is to provide a standardised communication protocol for military vehicle sub-systems, offering deterministic communication between the applications layers of the attached devices. CAN itself provides deterministic communication at the Network level (through its arbitration mechanism) but can not guarantee delivery latencies for an application message (determined from the time the application of the source device triggered a message, due to a timed or external event, and the message was delivered and processed by the application of the destination device).

The MilCAN specifications define an operating environment that can be implemented in software level and is (roughly) placed between the User Application and the CAN controller. By following the guidelines and requirements of the Specs, a level of deterministic communication with "guaranteed" delivery times is achieved by scheduling the operation of the whole network. Basically, it allows you to schedule all message transmissions within the network based on the priority of each message, allowing higher priority tasks not to be affected by lower priority ones. This also indirectly increases the usability of the CAN network as the user can increase the load on the bus as message transmissions can be spread within the operational timeline (something not easily implemented in an unscheduled network).

The MilCAN "Slot" is a time period defined by the consecutive transmission of a specific CAN frame (the Sync Frame). This is the PTU (Primary Time Unit) based on which all message transmissions of Synchronous frames are pre-scheduled. The duration of the PTU is user-definable and does not depend directly on the CAN protocol. The smaller the PTU length the higher scheduling granularity achieved.

The synchronisation functions are software based and require one or more timing sources (depending on the specific implementation) such as a hardware Timer found in most microcontrollers. The Timers are used to trigger the generation of the Sync Frame and also to detect the timeout of a Sync Frame (for fail-over operations). The actions following the triggering are software based in order to handle transmission of the Sync Frame CAN message.

To summarise, MilCAN is a software based protocol with minimal hardware requirements. It is an open standard and can be implemented by anyone. Although designed for use in military subsystem networks, it can be used in any CAN based network that requires deterministic and prioritised communication at the application layer.

/peri

Specs - General :: general_questions

Author: Alberto
Subject: general_questions
Posted: Thu Nov 09, 2006 5:00 pm (GMT 0)
Topic Replies: 1

Hello everybody,

I know that MilCAN is intended to be compatible with CANopen and SAE J1939 but what is the fundamental difference between those protocols, I know that is for militar purpose but is it more secure or faster ?

Another doubt I have is, what is the duration of one slot ? Is is the same duration as in CAN protocol ?
How is implemented synchronization in MilCAN, does it use any special hardware to implement it ? a clock ? or is it implemented by software ?

Thank you very much

Regards

Alberto

Specs - General :: RE: MILCAN_software

Author: Alberto
Posted: Mon Nov 06, 2006 7:58 am (GMT 0)
Topic Replies: 3

Hi peri,

Actually I'm looking for both of them, analysers and Libraries/stacks. So if there are some commercial companies that sell those Libraries, could you tell me please some of them, or any interesting link, as I was looking for them and I couldn't find it.
Most of people told me that the only way is I have to implement that libraries.

Thank you very much again.

Specs - General :: RE: MILCAN_software

Author: peri
Subject: Re: MilCAN software
Posted: Fri Nov 03, 2006 10:41 am (GMT 0)
Topic Replies: 3

Alberto,

Are you looking for simulation/emulation modules of MilCAN for CANbus analysers (such as CANoe/X-Analyser/CANanalyser), or MilCAN libraries/stacks to use in a system?

As far as I know there are no add-on modules for such analysers. Libraries/stacks are available from a few commercial companies and contractors.

/peri

Specs - General :: MILCAN_software

Author: Alberto
Subject: MILCAN_software
Posted: Tue Oct 31, 2006 3:43 pm (GMT 0)
Topic Replies: 3

Hi,

I wonder, if somebody knows any software module for MILCAN ptrotocol, like DeviceNet for Canalyzer with CAN.

Thank you very much.

Specs - General :: RE: questions

Author: peri
Subject: Re: questions
Posted: Thu Jan 05, 2006 12:41 pm (GMT 0)
Topic Replies: 1

dspcan1 wrote:

dear sir,I have some question as follows:

1.what's your aim to improve the CAN BUS? for better using it in combat vehicle? AND I knew that the combat vehicle called weasel2 had a successful application of CAN BUS.

2.Is the property of MILCAN on real-time and fault-tolerance better than that of CAN BUS?if so ,what's your proof?

3.Can I use MILCAN to finish my own design? what's the form of the milcan product,card or chips?

thanks!

MilCAN is an open higher layer protocol operating on top of CAN BUS, providing deterministic communication to the network. MilCAN nodes operate synchronously through a redundant bus-wide synchronisation protocol, based on which data transfers are scheduled. Using an 8 level prioritisation scheme combined with a set of guidelines and restrictions on message triggering and transmission, the operation of the network becomes deterministic with guaranteed delivery times on each of the priority levels.

MilCAN is an open protocol and can be freely implemented and used. Compliancy to the protocol and other MilCAN capable devices can only be achieved by adhering to the Compulsory and Optional Enhancement statements of the protocol.

Specs - General :: questions

Author: dspcan1
Subject: questions
Posted: Thu Jan 05, 2006 11:12 am (GMT 0)
Topic Replies: 1

dear sir,I have some question as follows:

1.what's your aim to improve the CAN BUS? for better using it in combat vehicle? AND I knew that the combat vehicle called weasel2 had a successful application of CAN BUS.

2.Is the property of MILCAN on real-time and fault-tolerance better than that of CAN BUS?if so ,what's your proof?

3.Can I use MILCAN to finish my own design? what's the form of the milcan product,card or chips?

thanks!

Events :: 3rd MilCAN newsletter

Author: peri
Subject: 3rd MilCAN newsletter
Posted: Fri Dec 02, 2005 12:10 pm (GMT 0)
Topic Replies: 0

The 3rd MilCAN Matters newsletter was release on 2nd December 2005.

For more information see the MilCAN website.

CAN Controllers :: RE: MilCAN Bus for PC

Author: peri
Posted: Thu Nov 24, 2005 5:50 pm (GMT 0)
Topic Replies: 3

Technically yes, you can implement MilCAN within your application. MilCAN operates on top of the CAN protocol / CANbus.

The "snag" (but also the advantage of MilCAN) is the real-time characteristics. The Sync Frame is the primary function that requires deterministic execution. The operation of the MilCAN network is based on the generation of the Sync Frame. Transmission of frames is coordinated by the Sync Frame sent on the bus by the active Sync Master. All nodes need to "react" on this message with a very small latency (10s of microseconds). Also, the time slices (PTUs) defined by the periodic generation of the Sync Frame include an "exclusion zone" where transmission of operational messages is prohibited (in 1Mbit CAN the PTU is 2msec).

As you can see these real-time characteristics require the MilCAN stack to operate with strict timming, something not possible using standard operating systems. Classic Windows will not be able to support it. Not sure about the Embedded versions. If you can achieve guaranteed reponses to CAN activity in the range of 10-100usec you should be able to implement the stack in them.

CAN Controllers :: RE: MilCAN Bus for PC

Author: dave_m_moore
Subject: MilCAN For PCs
Posted: Thu Nov 24, 2005 5:05 pm (GMT 0)
Topic Replies: 3

Thanks for your response peri.
So what I think your saying is that if I have a standard CANbus card, I can implement the MilCAN bus behaviour in my application? (essentially 'on top' of the standard CANBus behaviour). Is this right?.

I've read that the main attribute of MilCAN over CAN is a 'sync frame' message containing a slot number. Each node is allocated certain frames to respond within.

I read with interest you comments on the need to use an RTOS. Does Windows qualify for this?(!). I was thinking of something along the line of:

http://www.phytec.com/can/hardware/pccaninterface/peakpci.htm

Would this be adequate?.

Thanks for your comments in advance,
Dave

CAN Controllers :: RE: MilCAN Bus for PC

Author: peri
Posted: Thu Nov 24, 2005 2:24 pm (GMT 0)
Topic Replies: 3

Dave,

As MilCAN is a higher layer protocol, any CAN based interface can be physically attached on a MilCAN network. The "MilCAN functions" are handled by the processor the CAN controller is connected to. This can be for example (a) a PCI card with a CAN controller connected to a PC, or (b) a PCI card with a CAN based microcontroller connected to a PC, or (c) a CAN based microcontroller with a USB port connected to a PC, etc.

In example (a) the MilCAN protocol will have to be implemented within the PC operating system, while in (b) and (c) in the microcontroller. Due to the realtime requirement of the MilCAN protocol it is necessary for the MilCAN stack to run on a deterministic environment. Example (a) will require an RTOS, while (b) and (c) can use the microcontroller processor to run the stack.

CAN Controllers :: MilCAN Bus for PC

Author: dave_m_moore
Subject: MilCAN Bus for PC
Posted: Wed Nov 23, 2005 5:32 pm (GMT 0)
Topic Replies: 3

Hi All,
Can I get a PCI card for a PC that will handle a MilCAN bus?. Can I just get a normal CANBus card or won't this support MilCAN?.

Ta in advance,
Dave

Events :: 1st MilCAN Newsletter

Author: peri
Subject: 1st MilCAN Newsletter
Posted: Mon Oct 31, 2005 5:53 pm (GMT 0)
Topic Replies: 0

On 1st November 2004 the first MilCAN newsletter was published.

CAN Controllers :: RE: CAN Controllers that disable re-transmission

Author: gweatherhead
Subject: Caution: Aborting on Error
Posted: Fri Sep 02, 2005 9:15 am (GMT 0)
Topic Replies: 3

Just a word of caution. When using controllers which have the abort facility, once a transmission error has been detected you MUST be certain that you can service the interrupt and abort before the controller retries.

As such I do not recommend this facility, unless you have set it up in a controlled manner (e.g. in your own FPGA).