On Tue, Nov 20, 2007 at 10:53:33PM -0800, Andrew Greenberg wrote:
> Hi Mark,
> Sorry, I wasn't totally paying attention. Let me review a few things:
> > 4-byte big-endian "ID"
> > 4-byte big-endian timestamp
> > 8 bytes of data
> This is only the case for *certain* CAN messages on the bus. Many of the
> messages had no bytes of data, and some had one or two. All of the CAN
> message IDs and data lengths are all specified in the firmware directory:
> http://svn.psas.pdx.edu/psas/trunk/c/firmware
> ... except, uh, svn web is down. Jamey, can you fix this? Sorry.
> And just to show I wasn't paying attention, I'll re-ask the question:
> did we pack them into same size structures? We must have. But someone
> correct me if we didn't.
> > Help me interprit the following:
> > 00001490   00 00 73 01  00 00 1E B4  34 00 04 A5  07 31 07 75
> Oh god, don't make me do this again. But, if you really want to, when we
> get access back to svn, I'll dig out the old firmware headers and
> interpret a few of them for you.
> Finally, sorry, but I still don't understand why you're not just using
> the text logs after we parsed them. I believe we went over them, and I

Sorry, but the text data looks corrupted in the time domain.  What else
can we do but this?

> believe they're correct, with the exception of the jiffy time stamp. But
> we never relied on the jiffy time stamps anyway: we always used the
> accelerometer messages as a 2.5 KHz "heartbeat" to the system.

If the accelerometer messages are coming in at 2.5Khz then why is it
that we one only one accel data item per jiffy before the launch?

BTW 2.5Khhz does fit with the observed 24-26 samples per jiffy after the
lauch started.

I'll redo my analysis with the 2.5KHZ assumption for the sample rate and
ignore the time stamp.  This may clean up some of my issues with the x/y
acceleration data.

> In any case, we can discuss this all in more detail on 12/5.

I look forward to that.



Attachment: signature.asc
Description: Digital signature

psas-software mailing list

Reply via email to