Real-Time Measurement and Control

cancel
Showing results for 
Search instead for 
Did you mean: 

Power up state on sbRIO-9607

Solved!
Go to solution

I'm developing a board set for the sbRIO-9607.  When power is applied to the RIO, it appears that all the FPGA pins are driven high for ~50-100ms.  Is this the expected behavior?

 

I'm using pull-down resistors.  And after the bitfile is loaded (~45 seconds after power-up), all the pins that are designated as outputs work correctly.  Thanks.

 

-Joe

0 Kudos
Message 1 of 12
(3,902 Views)

Any thoughts?

0 Kudos
Message 2 of 12
(3,824 Views)

Hi JoeLesker

 

On page 40 of the following document, you can see that all DIOs are floating before and during FPGA configuration. You need to use a pull-up or pull-down resistors to ensure start-up values. What value are you using on pull-down resistors?

 

NI sbRIO-9607 Single-Board RIO OEM Devices: http://www.ni.com/pdf/manuals/375463a.pdf

 

Besides that, what concerns me the most is that you mentioned that the bit file lasts ~45 seconds to load. Are you loading the bit file from the flash memory? 45 seconds sounds a lot to load a bit file.

 

Regards.

 

 

0 Kudos
Message 3 of 12
(3,812 Views)

Thanks for the reply.

 

I understand that the FPGA-IO will be tri-stated upon between power-up and the loading of a bitfile.  We are using 10k pulldown resistors on all outputs.

 

Sorry if I wasn't clear about the timing.  It takes 45 seconds for the OS and RT-EXE to load, which then flashes the bitfile as one of the first operations (the bitfile itself takes less than a second).  I was just trying to make the point that after normal loading, everything seems to function fine.

 

Our issue is with the first 50-100ms after power is applied the sbRIO board.  All outputs seem to assert themselves high during this time, and then go into tri-state.

0 Kudos
Message 4 of 12
(3,794 Views)

Hi JoeLesker,

 

What kind of accessories are you using to access the board outputs? I will try to reproduce this behavior. Are you using an oscilloscope to verify this?

 

Best regards,

 

0 Kudos
Message 5 of 12
(3,783 Views)

I'm using a NI-9694 RMC breakout board --> ribbon cables --> custom PCBs.  The PCBs have various functions -- 24V digital inputs and outputs, 24V PWM outputs, high frequency inputs, etc.  All pins are buffered and opto-isolated. I can send you some of the PCB schematics if you think that would be useful.  Thanks.

 

-Joe

0 Kudos
Message 6 of 12
(3,766 Views)

Hi,

 

Have you tried using lower resistance pull-down resistors? I would suggest trying 220Ω or something around. Maybe 10k is a lot of resistance and it is not being able to pull the output down.

 

regards!

0 Kudos
Message 7 of 12
(3,757 Views)

I can try a lower valued resistor, but I'm not sure how that would make a difference.  The pin should be tri-stated on power-up, so any reasonable resistance should pull the pin down to ground.

 

Any other ideas/explanations?  Thanks.

0 Kudos
Message 8 of 12
(3,716 Views)

Hi,

 

Due to the available power on the output when it is tri-stated, a 10 kiloohm resistor might not be pulling it down to ground. In order to provide you with a more efficient and dedicated support please give us a call at 866-275-6964 (866-ask-myNI) and one of our engineers will create a Service Request to continue assisting you.

 

Regards.

0 Kudos
Message 9 of 12
(3,708 Views)
Solution
Accepted by topic author JoeLesker

After some more digging/troubleshooting, it looks like the issue lies with the powerup sequence of one of the breakout boards.  Sorry if I wasted any of your time, and thanks for the help.

 

-Joe

0 Kudos
Message 10 of 12
(3,701 Views)