PCI card internals corrupt

[Cyber Point]

Hi,

We have installed the card. Windows says the card is OK. But if we start the file: AESTest.exe we get the following error.

PCI card internals corrupt


Please advice

[Cyber Point]

AES Intelligent Money Handling Equipment Interface

OS Version 5.1

Checking Driver File:

Found at: C:\WINDOWS\system32\Drivers\AesIMHEI.sys

Timestamp : Sun Feb 08 15:59:00 2004



Checking DLL:

Found at: C:\WINDOWS\system32\AesIMHEI.dll

Timestamp : Sun Mar 28 16:14:00 2004

Version 1,1,0,1



Checking Device Access:

AES Intelligent Money Handling Equpiment:

Interface 2, code version 04010401

*** PCI card internals corrupt ***

[davebush]

Your problem is that you a running a very old (prototype) version of the Firmware on the Milan PCI card, and the modern PC side software does not interface to it.

You have essentially two choices:

• 1/ Return the card to your supplier and ask them to provide you with one running the current (1.8.14) release of the firmware.
  2/ Use the hardware “bootstrap” function of the card to load in the recent build.

If you are interested in this latter route you will need:

• 1/ A Re-programming lead (which you will have to make yourself)
  2/ The Hitachi Flash programming utility
  3/ A copy of the firmware in S-Record format.

Covering each of these in turn:
Programming Lead

This is converted from a standard network cable.

• A “standard” network cable is one that has the colours of the cores going into the plug as follows: (Holding the lead with plug at the top and the clip at the back, reading from left to right)
  Orange & White
  Orange
  Green & White
  Blue
  Blue & White
  Green
  Brown & White
  Brown

To convert such a network cable into a reprogramming lead you should:

Cut the cable in half and connect three of the cores to a 9 way female D Type:

• Blue - Pin 3
  Green & White - Pin 2
  Orange - Pin 5

Connect the Brown core to the Brown & White core.

You can ignore the other leads.


The Flash reprogramming utility:
You will need the lead plugged into the back of the PCI card and COM1/2. There should be a switch will put the card into run mode or into reprogram/reset mode. That, obviously, has to be in reprogram mode.

The Hitachi software that does the reprogramming can be installed from: http://www.aardvark.eu.com/PCI/Utilitie ... 20v1_1.exe

Once you've installed it you need to set the FLASH / Settings menu up as:

• DeviceName=H8S/2623F
  InterfaceName=Direct Connection
  PortName=as appropriate
  BaudRate=38400
  Mode = Boot

You then use the browse button to locate the MainV4-1-8-14.a37 image and then press the big FLASH program button. (It may appear to find errors on the read after write check - but that's a bug in FlashSimple!)

The S-Record (A37) Image.
This can be downloaded from http://www.aardvark.eu.com/PCI/Binaries ... 1-8-14.a37 (You will probably need to right click on this link and select "Save as")

Note: that after re-programming the card you will have to remove the lead and wait for 15 seconds before trying to use it.

Good Luck!

[Cyber Point]

?? what's wrong ??

----

Starting operation
Opening port 'COM1' ...
Connecting to device 'H8S/2623F' on 'COM1'
Configuration:
'BOOT Mode' connection - using emulated interface
Opening port 'COM1' ...
Booting the device
Initiating BOOT SCI sequence
Attempting 9600
Attempting 4800
Attempting 2400
Attempting 1200
ERROR: BootSCI() failed

1 error(s), 0 warning(s)

[Cyber Point]

We have found this on your forum:


--

To re-program the PCI card. You will need the lead plugged into the PCI card and COM1/2. There should be a switch will put the card into run mode or into reprogram/reset mode. That, obviously, has to be in reprogram mode.

--

Where is the switch ??? There are some jumpers on board !! but no switch.

[aardvark]

The error messages show that the Hitachi software is failing to communicate with the embedded micro. I can only suggest that you re-check the connections.

In the early days of the project, we supplied pre-built cables with a switch connecting the brown/brown&white cores.

If you know something about hardware, and want to try troubleshooting the cable - replace the brown/brown&white connection with a switch. With the switch open, and something like hyperterm running (9600,n,1), you should see diagnostics output by the firmware and it should react to a <return> with a '$' prompt.