Could not detect target chip

[joshkeys]

Hi guys,

I have not been able to locate a solution to my problem in the forums yet, so hopefully someone can help.

I am using the ICD-U40. I am able to connect to the ICD and even upgrade the firmware to the latest (CCS 1.39), but I am unable to connect to any target device (currently PIC18F4320). I have the connections proper as I have tested it with a working ICD-S40 by just swapping the RJ-12 cable. What I beleive happened is the first time I used the U40 I had a jimmy rigged RJ12 cable where 2 wires were shorted I believe, therefore I most likely blew an output driver on the U40. As I stated the device is recognized in windows and the LED comes on and I can do everything but program or basically recognize the target PIC. So I am wondering if anyone would know what chip may have blown on the U40, I have opened it up and see no visual indication (swollen cap, black resistors or board), but if someone knew which chip it was I may be able to swap it out. My warranty ran up a year ago so I would rather repair it myself. Anyways, thanks

Josh

[Foppie]

Sorry for bringing up this older post but I have searched this forum and this was the only post that described my problem.

I have here 2 ICD-U40s with exactly the same problem. is there someone with a solution?

Jos

[joshkeys]

I never did find a solution. i had to buy another ICD programmer.. sorry.. Someday if I have time I will probably troubleshoot the old programmer some more. If you ever find out feel free to tell me.

Thanks,
Josh

[Goodwinchang]

Actually My ICD-U40 has the same problem also. Why CCS doesn't have any countermeasure for it?

[joshkeys]

The reason CCS does not have a "fix" for this is because it is most likely a hardware problem. Hook up an oscilloscope to your ICSP data and clock lines and also your Master clear and ensure you are seeing your Vpp on your MCLR and that your data and clock lines are functioning as they should (see CCS ICD troubleshooting guide for what it should look like).

Josh

[opusofok]

I had this same problem with an 18F87j11. The problem was the chip was not accepting the external clock. It was running off the internal osc and that was running very slow. I used the troubleshooting routine and stepped thru till I reached the point where it read the ID. The first read would come back as FF but the second read came back ok. After that I could close the troubleshooting box and program the part.

The way I figured out what was happening is I checked the osc out with a scope. The clock I was using is 48Mhz but the osc out was running at 17khz (approx).

Hope this helps.

Opusofok[/quote]

[mike]

I had the same issue and it turned out that the ICD had some pins that had not been soldered. It was difficult to detect because the pins were covered in flux and are quite small.


I now have an issue where my computer gets really slow if I am running the ICD program while connected to the ICD programmer. If i pull the usb cable out, the computer returns to normal.

Shoddy soldering and shoddy programming in my opinion.