[hpsdr] HPSDR - FPGA updating

[w3sz]

Hi All,

I want to make sure I understand the physical 'rules' for programming 
the HPSDR Boards 'in situ' on the Atlas backplane, so I don't cause any 
problems when I do said programming.

 From my reading of the postings to this list, I believe what is below 
is a correct, complete, and not-overdetermined set of rules.  Is this 
correct?

1.  Ozy must be the most distant board from the Atlas Power connector

2.  The board closest to the Atlas Power connector must have the 'last 
JTAG' jumper in place.  This jumper should be left 'open' on other boards.

3.  If the boards are not sequentially positioned [i.e., if there are 
any gaps between the boards where an Atlas connector is left empty] or 
if the slot closest to the Atlas power connector is empty, then these 
empty slots must be filled with a DIN 41612 male plug [e.g. Mouser 
617-09-03-196-6921] that has pins A27 and A29 shorted together.

Are these three rules correct, necessary, and sufficient?

And two final requests for confirmation so all 'necessary' information 
can be found in one place for future reference:  it appears to me that 
the "A" terminals [e.g. A27 and A29] on the DIN 41612 connectors are 
those that are furthest from the Atlas power connector.  It appears that 
pin 1 is on the right and pin 32 is on the left, as one looks from the 
Atlas power connector towards the far end of the board.  Are these 
statements correct?

Thanks in advance, and

73,

W3SZ
Roger Rehr
http://www.nitehawk.com/w3sz



> On 8 Nov 2009, at 00:05, Phil Harman wrote:
>
>>> Roland/Phil
>>>
>>> Is the JTAG bus section of the Atlas bus bi-directional? Reason for
>>> asking is that, using the excellent back-plate for Pandora, Ozy is
>>> last on the bus with Penny/Mercury preceding. If I put the "Last
>>> JTAG"
>>> jumper on the board nearest the power connector (sorry, I'm working
>>> from memory without the rig in front of me) will Roland's method
>>> still
>>> work?
>>>
>>> Cheers
>>>
>>> Chris
>>> G4NUX
>>>
>>
>> Hi Chris,
>>
>> Not its not - you have to have the boards in consecutive slots and Ozy
>> farthest from the power connector.
>>
>> BTW, Excalibur has JTAG pass thought so you can add it anywhere in the
>> JTAG chain and program the other boards as before.
>>
>> 73's Phil...VK6APH
>>
>>
>>> On 7 Nov 2009, at 12:23, roland etienne wrote:
>>>
>>>> ***** High Performance Software Defined Radio Discussion List *****
>>>>
>>>> Some precisons:
>>>>
>>>> Instead of running the Altera Quartus IDE to access to the
>>>> programmer, you
>>>> can run a stand-alone version (the last is 9.1, a 122 MB file...).
>>>>
>>>> And of course, you can update only one of the device (Mercury or
>>>> Penelope),
>>>> so the best way to do it is:
>>>> Auto-detect, and then select the device that you want to update, and
>>>> change
>>>> the file to load the correct .jic file, don't forgot to check the
>>>> programm/configure checkbox for the device concerned, and then
>>>> start, and
>>>> wait..
>>>>
>>>> Hope that helps,
>>>>
>>>> 73, Roland F8CHK
>>>>
>>>>
>>>> -----Message d'origine-----
>>>> De : hpsdr-bounces at lists.openhpsdr.org
>>>> [mailto:hpsdr-bounces at lists.openhpsdr.org] De la part de roland
>>>> etienne
>>>> Envoyé : samedi 7 novembre 2009 12:19
>>>> À : phil at pharman.org; hpsdr at openhpsdr.org
>>>> Objet : Re: [hpsdr] HPSDR - FPGA updating
>>>>
>>>> ***** High Performance Software Defined Radio Discussion List *****
>>>>
>>>> Hello,
>>>>
>>>> The process to update the Flash EEPROM on the Mercury and Penelope
>>>> boards
>>>> without having to move the boards, as we did before, is really
>>>> simple using
>>>> the programmer interface within Quartus.
>>>>
>>>> After loading the soft witch make Ozy appears as a usb-blaster, you
>>>> run
>>>> Quartus, in Tools you run Programmer, you choose USB-Blaster as
>>>> hardware
>>>> programming tool.
>>>>
>>>> To test the JTAG chain, you can run the auto-detect function, witch
>>>> should
>>>> show you the EP2C8 and the EP3C25 if everything is correct (last
>>>> JTAG strap
>>>> in good position, in my case the order of the boards is Ozy,
>>>> Penelope, and
>>>> Mercury in last position).
>>>>
>>>> After the auto-detect, delete the two devices, and add file, first
>>>> the one
>>>> for Penelope, and then the one for Mercury. These files are the
>>>> *.jic files.
>>>> Check the program checkbox for the EPCS4 and the EPCS16.
>>>>
>>>> Then start to program, and wait...
>>>>
>>>> Now that the boards are in Pandora, this process is painless!!
>>>>
>>>> Best 73,
>>>> Roland F8CHK
>>>>

 1258899217.0