cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 
Highlighted
Observer
Observer
1,089 Views
Registered: ‎05-06-2013

KC705 FPGA_INIT_B always red

Jump to solution

I have a KC705 which I got 2 years ago. When it arrived there was a problem with the power supplies. I got the Ti power supply dongle and got the board working fine, until recently. The FPGA_INIT_B LED is always lit RED.

I've run the same script as before to test the power supplies. It indicates that everything is working properly ( the FPGA_INIT_B LED is RED as it reports this ). All of the power supply indicator LEDs are green and all of the supplies are working. I have no problem configuring the FPGA using either the Digilent programming port or the JTAG header with a Digilent HS3 cable. Regardless, nothing is working after configuration. I have an older KC705 board that I've blown the PCIe interface and I can configure it with the same bit file and have the expected operation. The programming switch is set for JTAG. Nothing should be trying to configure the device on power-up. I can't use the old board because I need the USB UART which also seems to be dead.

0 Kudos
1 Solution

Accepted Solutions
Highlighted
Observer
Observer
623 Views
Registered: ‎05-06-2013
So I've figured out what the problem is. The cheap SW14 Program button is stuck so that FPGA_PROG_B is always asserted. All voltages are just fine. I assume that it's a mechanical failure but I never use it so I have no idea why it would fail. I'd say that it's safe to say that these button issues plagues a lot of boards.

View solution in original post

0 Kudos
8 Replies
Highlighted
Moderator
Moderator
1,073 Views
Registered: ‎09-18-2014

Bacarter,

 

INIT lighting red would indicates CRC error. Read the device status register for more information. Your configuration was bad in someway. Have you run through the KC705 debug checklist AR? I have linked in below. Make sure to also run through the power supply section as well. In particular those related to configuration like VCCINT, VCCAUX, VCCO_0 etc. Have you tried using an example design bit-stream? Made sure all the jumpers, switches are set to default? Make sure your JTAG module on the board is not loose/cracked? Make sure you are not budging the JTAG cable while programming? Have you tried different JTAG cable? There other debug steps you can perform from the debug AR as well as other posts in the forums with similar issues. 

 

https://www.xilinx.com/support/answers/50079.html

 

 

Regards,

Tezz

-------------------------------------------------------------------------
Don’t forget to reply, kudo, and accept as solution.
-------------------------------------------------------------------------
0 Kudos
Highlighted
Observer
Observer
1,061 Views
Registered: ‎05-06-2013
Hi Tezz, Thanks for the quick reply. I don't think that you read my post very carefully. Yes, I checked all of the supply voltages. They are all there and the LED indicators are all green Yes, I did use a known good bitstream for that device (it works on another KC705 board) It's not a connector issue. I tried using the external JTAG USB connector and J60 using a JTAG cable. I am able to detect the correct device through either method, the application that I use configures the device without error, but it's clear that the device never gets configured. I've never seen this behaviour in 20 years of FPGA development. Yes, I went followed the debug page that you refer to. It doesn't address my issue. On top of that I ran through the Ti Power Supply Software using the Ti USB to GPIO Adapter that fixed the power supplies when I first got the board. Multiple times. It ran though all of the tests and when finished declared the power supplies to be "Manufacturing Passed' . Multiple times. Though all of this the FPGA_INIT_B LED is NEVER anything but RED.
0 Kudos
Highlighted
Observer
Observer
989 Views
Registered: ‎05-06-2013
Here's some more information about the issue which I still haven't been able to resolve: After letting the board sit for a few days, unconnected to anything, I am able to configure the board with a known good design. All the time the FPGA_INIT_B LED is lit solid RED. Then, if I try to reprogram the part it apparently fails; even though the configuration utility says it was successfully configured and is able to identify the device codes properly. I know when the board is configured because of the user LEDs, once of which is a heart-beat in the design. I spent some time trying to debug the board and there is nothing trying to initiate configuration, except when I use the JTAG to do it. Were there ever production Kintex devices released into the wild with this problem?
0 Kudos
Highlighted
Observer
Observer
753 Views
Registered: ‎05-06-2013
I still haven't resolved this issue. I can program and verify using Impact or Digilent Adept JTAG and there are no problems except that nothing works. I've verified that all of the voltages are correct. INIT_B LED is always RED and the config_done LED is never lit despite what the JTAG programmer indicates. I've attached the laste log using Impact.
0 Kudos
Highlighted
Xilinx Employee
Xilinx Employee
736 Views
Registered: ‎06-06-2018

Hi @bacarter ,

I believe only INIT_B is RED at power up. Since you are claiming that All Power GOOD LEDS are active. 

 

Since INIT_B is always RED. After power up, Pulse PROG_B and then try to configure the FPGA through JTAG. And share the results.

 

If again fails, Try to reduce the JTAG TCK frequency to 3 MHz and try configuring the FPGA and share the results.

 

 

I observed That there Bad packet error signal is HIGH in the Configuration Status Register. Thats the reason INIT_B pin is low and configuration is not complete.

Bad Packet error can happen due to :

1. The CCLK frequency is above the data sheet specification.

2. Power supply issues. If your configuration source rails droop or drift out of spec this can be seen as well.

 

To check the healthiness of the Board, perform BIST Test and download Related Files and share the results.

Note : You can download Vivado Lab edition and try BIST Test.

 

 

Regards,

Deepak D N

---------------------------------------------------------------------------------------

Please reply or give Kudo or Mark it as an Accepted Solution.

---------------------------------------------------------------------------------------

 

 

Regards,
Deepak D N
---------------------------------------------------------------------------
Please Kudo and Accept as a Solution, If it helps.
---------------------------------------------------------------------------
0 Kudos
Highlighted
Observer
Observer
722 Views
Registered: ‎05-06-2013

Deepak,

Thanks for the reply. I don't believe that you understand the situation. INIT_B is always asserted low from power on until power off. I can't run the BIST because it's impossible to configure the device. It doesn't configure from FLASH or from JTAG, even though Impact or the Adept utility says that configuration and verification was successful. I've tried the lowest JTAG clock and there is no change. Obviously the FPGA JTAG section is working since the onfiguration tools never indicate anything but success.

INIT_B is never high, the INIT_B LED is always RED when power is on regardless of any attempt at configuration, and config_done is never asserted. The board is quite useless. I managed to get about 10-15 uses out of the board before the issue started. If you read back through the thread you will see that I reprogrammed the power supply chain successfully a number of times.

0 Kudos
Highlighted
Observer
Observer
720 Views
Registered: ‎05-06-2013
I mean't to add that I checked the supply voltages with a volt meter.
0 Kudos
Highlighted
Observer
Observer
624 Views
Registered: ‎05-06-2013
So I've figured out what the problem is. The cheap SW14 Program button is stuck so that FPGA_PROG_B is always asserted. All voltages are just fine. I assume that it's a mechanical failure but I never use it so I have no idea why it would fail. I'd say that it's safe to say that these button issues plagues a lot of boards.

View solution in original post

0 Kudos