Re: SDRAM problems

[Casainho]

On Fri, Jun 18, 2010 at 4:12 PM, Opendous Support
<opendous...@gmail.com> wrote:
> Casainho,
>
>  Please test this firmware.  I am about to switch to the Micron DDR
> so this might be useless but please check to see what output you get.
> It appears I have at least one stuck pin.  I set the firmware to run
> at a UART speed of 57600 so that it will work with an 8MHz USB AVR
> running USBtoSerial from the latest LUFA release.  With or without
> UART I get the same errors.  It is not enough to just test fixed
> patterns of data as all that tells you is that one or more addresses
> hold the correct data but not whether addressing works.

Right now I am very busy to get my 3D printer working, for printing
enclosures for this board/Lyre project.

Bob should get his board tonight and he will work on code, since he is
more a firmware guy.

For now I trust on my SDRAM/board and if there are any problems, Bob
will get them ;-) -- Later I may start working also on the firmware.

By the way, I bought FAKE poliamide/Kapton tape from DealExtreme, so,
if anyone is thinking to buy it from there (for gluing the LCD) please
avoid it:
http://casainho-emcrepstrap.blogspot.com/2010/06/fake-katpon-tape-from-dealextreme-do.html

>  You can connect the debug UART as pictured and communicate with:
> gtkterm --port /dev/ttyACM0 --speed 57600
>
> Later,
> Matt,
> Opendous Inc.
>
> =!=!=!=!=!=!=!=!=!=!=!=!=!=!=!=!=!=!=!=!=!=!=!=!=!=!=!=!=!=!
>
> Relevant Output for sdram_test-2010-06-19-1100.zip:
>
> HW_DRAM_CTL33 = 0x204100c8
> 0x40000000 = 0x00400100
> 0x40000004 = 0x00400101
> 0x40000008 = 0x00400102
> 0x4000000C = 0x00400103
> 0x40000010 = 0x00400104
> HW_EMI_CTRL = 0x1c024040
> 0x40000000 = 0x00400100
> 0x40000004 = 0x00400101
> 0x40000008 = 0x00400102
> 0x4000000C = 0x00400103
> 0x40000010 = 0x00400104
>
> SDRAM failed Test(s)!
>
> Success Count=0x00200000, Error Count=0x00600000
> Last Correct Address=0x41fffffc
> DONE
>
> =!=!=!=!=!=!=!=!=!=!=!=!=!=!=!=!=!=!=!=!=!=!=!=!=!=!=!=!=!=!
>
> Relevant Output for memory_test.tar.bz2:
> Note 0x00000100 = 0b100000000 = 256
>
> 0x40000000 = 0x00000100
> 0x40000004 = 0x00000101
> 0x40000008 = 0x00000102
> 0x4000000C = 0x00000103
> 0x40000010 = 0x00000104
> MemFail at 0x40000000 = 0x00000100 should be 0x00000000
> MemFail at 0x40000004 = 0x00000101 should be 0x00000001
> MemFail at 0x40000008 = 0x00000102 should be 0x00000002
> MemFail at 0x4000000c = 0x00000103 should be 0x00000003
> MemFail at 0x40000010 = 0x00000104 should be 0x00000004
> MemFail at 0x40000014 = 0x00000105 should be 0x00000005
> MemFail at 0x40000018 = 0x00000106 should be 0x00000006
> MemFail at 0x4000001c = 0x00000107 should be 0x00000007
> ...
> MemFail at 0x400003f8 = 0x000001fe should be 0x000000fe
> MemFail at 0x400003fc = 0x000001ff should be 0x000000ff
> MemFail at 0x40000800 = 0x00000300 should be 0x00000200
> MemFail at 0x40000804 = 0x00000301 should be 0x00000201
> MemFail at 0x40000808 = 0x00000302 should be 0x00000202
> ...
> MemFail at 0x40000bf8 = 0x000003fe should be 0x000002fe
> MemFail at 0x40000bfc = 0x000003ff should be 0x000002ff
> MemFail at 0x40001000 = 0x00000500 should be 0x00000400
> MemFail at 0x40001004 = 0x00000501 should be 0x00000401
> MemFail at 0x40001008 = 0x00000502 should be 0x00000402
> ...
> MemFail at 0x400017f8 = 0x000005fd should be 0x000005fe
> MemFail at 0x400017fc = 0x000005fe should be 0x000005ff
> MemFail at 0x40001800 = 0x000005ff should be 0x00000600
> MemFail at 0x40001804 = 0x00000700 should be 0x00000601
> MemFail at 0x40001808 = 0x00000701 should be 0x00000602
> ...
> MemFail at 0x400047f8 = 0x000011fd should be 0x000011fe
> MemFail at 0x400047fc = 0x000011fe should be 0x000011ff
> MemFail at 0x40004800 = 0x000011ff should be 0x00001200
> MemFail at 0x40004804 = 0x00001300 should be 0x00001201
> MemFail at 0x40004808 = 0x00001301 should be 0x00001202
> ...
> MemFail at 0x40004ff8 = 0x000013fc should be 0x000013fe
> MemFail at 0x40005000 = 0x000013fe should be 0x00001400
> MemFail at 0x40005004 = 0x00001501 should be 0x00001401
> MemFail at 0x40005008 = 0x00001500 should be 0x00001402
> MemFail at 0x4000500c = 0x00001503 should be 0x00001403

[Casainho]

On Fri, Jun 18, 2010 at 5:29 PM, Opendous Support
<opendous...@gmail.com> wrote:
> Casainho,
>
>  I bought a new NHD-2.4-240320SF-CTXI%23-T so I can test it properly as well.

Good!

>  Did you also buy the Micron DDR?

No, we (me and Bob) didn't bought nothing yet. We would love your help
on commenting the schematic Bob made for LCD and buttons expansion
board: http://lyre.svn.sourceforge.net/viewvc/lyre/hardware_design/lyre_v1_io/

When we get board approved, then we can buy in one time all components needed.

Our boards have already the other SDRAM so I don't see why we should
spend more money and efforts to solder another SDRAM. Could you
explain?

[Opendous Support]

>Our boards have already the other SDRAM so I don't
>see why we should spend more money and efforts to
>solder another SDRAM. Could you explain?

Winbond has obsoleted the W9425G6EH-5 so you will not be able to
make the same board in the future. Your debugging efforts for the
W9425G6EH-5 may not be useful for other SDRAM ICs.

When you do order the LCD, get a rework kit and solder wick as well.
http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=SMD1NL-ND
http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=473-1061-ND

You will be able to safely and easily remove the IC.
http://www.youtube.com/watch?v=FTQqjggeklo
http://www.youtube.com/watch?v=7kyaz4Zrd78