Community
cancel
Showing results for 
Search instead for 
Did you mean: 
Altera_Forum
Honored Contributor I
1,000 Views

error CONF_DONE pin failed to go high : cyclone 4 evalboard

hi i am using the Cyclone IV GX FPGA Development Board. 

it has worked fine for the last few months but now when i try to download the sof file on the FPGA the prosses go to near 90% then i have the following 

Error: CONF_DONE pin failed to go high in device 1 cyclone 4 

the board is as i got it form Altera with no rework . 

i have tried different usb blaster cable , different pc, uninstall and reinstall the usb drivers and still got the same error. 

the CONF_DONE led lights up for an instant and then turn off . 

 

any help is welcome  

plz ask for any additional information  

Thx in advance for your help 

fadel
0 Kudos
10 Replies
Altera_Forum
Honored Contributor I
50 Views

The conf_done pin needs to see a 10k resistor to VCCIO of the respective IO-Bank. 

Maybe the resistor is broken? If the aforementioned circuit is not present (respective damaged) you will see the FPGA constantly cycling through it's configuration scheme which should be easily measured with a scope on the aproppriate pins.
Altera_Forum
Honored Contributor I
50 Views

Maybe strange idea, but have you checked, that the device in QuartusII matches the target device on the board? (in case it's an iteration of an existing design it should). 

I had similiar stuff with prototype units I'm working with after several reprogramming and USB-Blaster connect/disconnect cycles. Unfortunately all devices showing the issue you describe were physically damaged, most likely by voltage peaks by connecting / disconnecting the USB-Blaster... 

Hope your's problem has other than physical (broken device) reason... 

 

BTW: I do not think it's an installation issue - maybe you can check the JTAG I/F in QuartusII (try to have Quartus find any connected devices).
Altera_Forum
Honored Contributor I
50 Views

i have tryed some of the sof that come with development board kit and the same error has occured. 

For the resistor it may be the problem i will check it . 

 

thx for your help
Altera_Forum
Honored Contributor I
50 Views

i have tested the resistor and found no problem in it any other ideas 

 

thx
Altera_Forum
Honored Contributor I
50 Views

Hi there,  

does the development board support programming of any onboard configuration eeprom e.g. by active serial I/F? (I'm not familiar with this board). That case - does the FPGA starts when loading from the ROM and "only" fails for the JTAG? 

Is the FPGA detected in the programmer and does the JTAG I/F seems to operate? 

 

Well.. this does not really sounds fine - do you have warranty on the board?
Altera_Forum
Honored Contributor I
50 Views

i have tried to use the Ethernet method to download the program into the flash but failed and i tried to reset the MAXII cpld on the board to factory via usb-blaster/jatag successfully 

even the conf_done problem is the same when i try to program the FPGA .
Altera_Forum
Honored Contributor I
50 Views

i can auto detect the device using the Quartus II programmer but i can't configure it

Altera_Forum
Honored Contributor I
50 Views

Hi, 

this indicates the JTAG interface being ok while there seems to be something defect with the configuration "part" or interfaces of the Cyclone... Last thing (while very unlikely to be the reason) may be the resistor being ok but not connected to VCCIO...  

Well seems for me - if there are no other ideas - that the FPGA is really broken to whatever reason...
Altera_Forum
Honored Contributor I
50 Views

I've seen this kind of thing before. If you see the config done light blink momentarily, are there outputs that, once configured, are driving a short? Or causing the supply to droop? Check the all the supply voltages right at the moment that you see config done blink. I bet one of them is drooping, causing the FPGA to reset and look for another configuration.

Altera_Forum
Honored Contributor I
50 Views

See whats under the board!  

Lately I had a board lying on top of an electrostatig shield bag which worked as a big GND-Layer shortcutting many pins to GND which prevented the CONF_DONE pin to go high. 

Good Luck