[hpsdr] Another Metis down

[K1rqg at aol.com]

My Metis failed with the same symptoms as K5SO had. End result was
only the 4 power LEDs lit and no other activity and of course, I could  not
recover. Boot loader did not work.
 
Previous to this failure, I was able to use HPSDRProgrammer to update
Metis code on XP 32 bit machine. It never worked on WIN 7-64 bit PC.
 
At the time of failure, I was NOT updating Metis code, but attempting  to
load latest FPGA code in a spare Penelope board. I used the procedure
in the "How it is done" document "JTAG Programmer. I followed the 5
step process, which as I recall was missing reference to JP1, but  this
was displayed in the Programmer process. All went well and the  process
reported no errors and programing complete. Cycled power, and Metis 
went dormant, nothing referenced in any of the documents would allow
recovery. No boot loader, no other activity. 
 
I am certain this was self inflicted, but I cannot see anywhere along the 
process that an error was either encountered or displayed.  I believe  it
is related to JP1 function and/or sequence of when to use it! 
 
The only difference that I can tell you is that I was using a 3 slot  EU
Atlas board vs the 6 slot US board. I always keep this connected to
my XP PC for testing and FPGA upgrades, and keep the 6 slot Atlas
on the WIN 7-64 PC for on air operation.
 
I am currently running Diversity 18 (K5SO) code with fantastic  results.
I have developed some hardware interface systems to integrate 3 RF
probe antennas (plus 2 horizontal antennas) with 2 Mercury boards,
and have phenomenal performance, especially at LF and VLF  frequencies.
 
I am in the process of documenting the hardware that I have  incorporated
including software control of the antennas via the I2C buss from  Mercury
to the HiZ amplifiers at the RF probes. 
 
I have also encountered some issues with the transmit audio chain in  the
latest W5WC software, but not quite sure how to explain it, until  further
use. 
 
Thanks to ALL of you that have developed this great HPSDR project and
all the ongoing improvements, etc.
 
73 de Joe
K1RQG
 
 
In a message dated 2/22/2011 12:03:30 A.M. Eastern Standard Time,  
scotty at tonks.com writes:

*****  High Performance Software Defined Radio Discussion List *****

Hi  Joe,

The reason that you could not recover from your "fault" is that  the 
bootloader code also resides in the PROM alongside the operating  
code.  If the PROM contents get scrambled (which they apparently  
did), the FPGA does not load either of the two images from the PROM  
and nothing works, not even the bootloader.

It is hard to imagine  what could have messed up the image in the 
PROM, but we'll all be looking  to see if this happens again or to anyone 
else.

Dan and I powered up  and down literally hundreds of Metis boards and 
never saw that  failure.

73,
Scotty  WA2DFI


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openhpsdr.org/pipermail/hpsdr-openhpsdr.org/attachments/20110228/c5ae62d3/attachment-0003.htm>