PXA-G49/USB-ICP-80C51ISP

[jdr]

Hello,
Programming a PXA-G49 with an 80C51ISP. I have the Flash Magic utility setting at: Advanced Options/Hardware Config: "Assert DTR and RTS while COM port open"
The attached scope images are of the TX/RX lines and they indicate that the chip gets the "f"... looks like it responds incorrectly the first time then correctly the second time and then a transmission begins... then fails. It throws a "unable to communicate" error.
I am using this setting because I can not get past the autobaud "f" without holding the PSEN down.
What could be the problem?
Thanks for your time.
Jeff

[Andy Ayre]

Please post a debug file. Instructions are here: http://www.flashmagictool.com/debugfile.html

[jdr]

I've attached the debug file. It consists of a series of "Read Device Signature" commands with different baud rates and one command to program.
Thanks for taking a look.
Jeff

[Andy Ayre]

All I see is that the Future Designs interface is echoing back what is sent to it. The command that Flash Magic sends is not acted upon, only echoed back. So there is no confirmation coming back saying that the command was processed.

[jdr]

Isn't the echo coming from the PXA-G49? The ISP firmware is echoing back what the Flash Magic is sending. Maybe I am confused about that but I thought that the fact that the echo existed meant that the chip is in the ISP mode. I am using the "terminal" option in Flash Magic and getting the same echo but it is very inconsistent ... input matches output then it will start sending garbage. Are my assumptions incorrect?

[Andy Ayre]

The interface you are using converts ISP to ICP. ICP is a hardware-level protocol. The ASCII responses are coming from the interface.

[jdr]

The binary ASCII echo is from the TX pin 13 on the microcontroller, after a lower case "f" has been sent to RX pin 11 to establish baud rate. The microcontroller's ISP firmware is auto echoing the character (Application Notes An 716 page 3). no?

[Andy Ayre]

I suggest you contact Future Designs. From your debug file it seems Flash Magic is sending a valid command but is not getting a valid response back. We can tell you what Flash Magic is doing, and we can fix any bugs there if necessary, but we can't tell you what the Future Designs interface is doing. I'm sure they will quickly be able to identify what step is missing.