Chief Delphi - CAN control weirdness

Chief Delphi (http://www.chiefdelphi.com/forums/index.php)

- Electrical (http://www.chiefdelphi.com/forums/forumdisplay.php?f=53)

- - CAN control weirdness (http://www.chiefdelphi.com/forums/showthread.php?t=79589)

CAN control weirdness

I was looking at the Jaguar CAN documentation, and I noticed a couple of things about the CAN setup:

The documentation talks about a "voltage control mode", where the voltage is varied according to the CAN command. Does this mean that the Jaguar in this mode is just sending an analog voltage instead of a PWM wave?
As I understand it, CAN will be run by daisy-chaining Jaguars together, with an additional safety verification provided in the form of a "heartbeat" PWM from the digital sidecar. However, if a point on the daisy chain fails, will the "downstream" Jaguars continue executing the last recieved command?(since there are no further commands coming through the CANbus and the heartbeat is telling it that everything is fine). Or is there some sort of additional safety system that will detect the loss of connection and shut down?

Re: CAN control weirdness

Quote:

Originally Posted by nathanww (Post 892909)

The documentation talks about a "voltage control mode", where the voltage is varied according to the CAN command. Does this mean that the Jaguar in this mode is just sending an analog voltage instead of a PWM wave?

No, it is still a duty cycle modulated wave form. The difference is that it will compensate to ensure a fixed voltage across the motor. Without this feature, drive strength is dependant on battery voltage and load. With voltage mode, some of this variability is compensated for by dynamically changing the duty cycle.

It is not capable of creating an analog voltage.

Quote:

Originally Posted by nathanww (Post 892909)

As I understand it, CAN will be run by daisy-chaining Jaguars together, with an additional safety verification provided in the form of a "heartbeat" PWM from the digital sidecar. However, if a point on the daisy chain fails, will the "downstream" Jaguars continue executing the last recieved command?(since there are no further commands coming through the CANbus and the heartbeat is telling it that everything is fine). Or is there some sort of additional safety system that will detect the loss of connection and shut down?

Without going into too much detail, this is not a problem. A Jaguar that is not receiving fresh CAN messages will shut down, even if the PWM heartbeat tells it to continue.

Re: CAN control weirdness

Quote:

Originally Posted by EricVanWyk (Post 892915)

It is not capable of creating an analog voltage.

[Nitpick]
I know what you mean, and you are correct in that respect BUT: the world really is analog, and all voltages are also analog, particularly voltages going to a motor.

[/Nitpick]

(Spoken by a true Analog EE) ;)

Re: CAN control weirdness

Quote:

Originally Posted by Don Rotolo (Post 892937)

Sorry Don. I actually had the "aww, thats not right but they'll know what I mean" thought train go through my head, but I was already off eating dinner. Please forgive me.

Re: CAN control weirdness

Even I admit it was nit-picking.

Off topic: Back when I was in college, Digital Electronics was "The Thing" to study. How hard could it be, one and zero? I chose Analog design, using the logic* that the world was built in Analog and so I'd never be out of work. It turned out that way, too. So, I defend Analog whenever I get the chance. Just a hobby of mine ...

Eric, just in case you didn't know, it's really a pleasure having someone so knowledgeable here in this forum, thanks for sharing your expertise with us.

Don

*No, not digital logic!