RMA Request RoboRIO 2

[Wired8516]

Hello NI Support,

We are team 8516 that has already posted with issue with the red power light issue on a RoboRio 2.0.  I am the head coach as well as a CSA that work champs this past week.  Seeing many RoboRio 2.0's fail with seem issues we came back from champs and have done through inspection of our bad RoboRio.  Below are the notes of what we have done and pictures of the board.

After exhausting all possibilities of any metal shavings inside the rio we are still unable to clear read power light that indicates shorts.  Boots in safe mode with solid green light and boots fine without SD card pulled with green power light and flashing status light.  All normal.  But with SD card installed with new image, pull up all he isolation pads around the pins up to blow out and inspect for metal shavings the board never boots up without red power light.  Even removing he 3.3vdc / 5vdc jumper to the DIO etc power rails results in red power light after booting up.  We did notice that the U81 chip heats up quickly once the power has been powered up for a couple minutes.  It's hard to tell if it's just that chip or the components around it that is heating up as well.  The back of the power is really warm with in 3-4 minutes after being powered up to the touch.  I suspect this is not normal.

After seeing multiple issues with communication issues, red power lights and no robot code on the new RoboRio's it would be great if NI would take some of these boards back and perform some root cause analysis to determine if there are manufacturing issues or other issues with SD cards being corrupt etc.  Please advise how we can get an RMA replacement for our current unusable RoboRio. 

U81 Chip heats up very quickly

[Yeoj]

Hey Wired8516, U81 is an NH245 IC, which is a bi-directional logic level shifter for data buses. Given that this device is connected to multiple power buses, it is possible it has an internal fault that is causing the issue, but is less likely than other causes (listed below).

The NI documentation in the manual, as well as tables published in the community forums and as answers to questions point to multiple fault conditions that can result in a solid RED Power LED (so its a higher level diagnostic indicator). So far it is claimed that the solid RED power LED can be caused by:

Shorts to external power buses
Input Voltage too low (brownout)
Damage to an external voltage bus (my experience - shorting 12V to a DIO ground and +5V caused our 5V bus to fail and the RED Power LED)
Overcurrent condition on any of the external power buses
And finally "Fault Condition Detected" on a power bus


Only NI can confirm (I am a mentor on another team and not an NI employee), but when booting into safe mode the manual specifies that only communication with the device and software installation is possible. Your Power LED staying GREEN is likely because higher level diagnostics have not been instantiated and are not running, leaving the LED in its default GREEN state even when a fault is still present. This is logical, as in safe mode you want the absolute bare minimum running so you can attempt to recover the device given bad software or image.


For your particular issue, can you power on the RIO and connect to it with a driver station and search for which power bus the fault is being detected on (as described in this post by BoKnows)? Alternatively if you have a DMM you can probe any of the 5V, 6V, and 3.3V pins to their respective grounds when the RIO is powered on to figure out which if any voltage bus is not present which could be triggering the fault and RED Power LED.


Yeoj2131

[Wired8516]

Yeoj2131,

Everything you have stated is true.  However, we seen these have red power lights on and removing the jumper for the 3.3v and 5v takes voltage to zero on DIO rails.  Making the power to go green when jumper was removed. Therefore, if there was a short it would now read zero and the power light goes to green as we discovered at champs while troubleshooting some issues there last week.  Our board doesn't clear the red light with this jumper removed.  So, I agree there is component on the board causing this issue.  Because safe mode booting only starts up with minimum configuration load for further diagnostics.  Which Greg from NI was able to communicate to a couple boards at champs via RS-232 port using some I assume NI driver and Putty software.  I didn't get a chance to follow-up with him on what he found while connected.  I assuming there is a way to read the boot up messages while startup to get further information.

Additional another CSA onsite was also using a multimeter looking for shorts as well as inspection of the solder joints on the board but we didn't find any smoking gun.  Thanks for your feedback.  We really need NI to step and issue RMA's on these rio's for replacements as well as other testing.  

[Yeoj]

Wired8516,

Yup, I see your other post about removing the jumper and agree it points to a fault between the DIO/Relay ports themselves and the jumper header. Those boards need further in person analysis to determine root cause.

Your issue however could be similar to mine in that a bus is permanently damaged. Is it possible for your team to determine WHICH power bus is causing the fault to be thrown by powering up and via the driver station reporting which bus is at fault, or use a DMM to determine which bus is not operational? (i.e. no 5V, no 6V or no 3.3V when powered on).


Regarding comm's via RS-232, the manual shows the configuration settings to accomplish this (snapshot below). I haven't done it so I dont know if there is a menu system or if we need to know commands to effect communication.

Yeoj2131

[mshafer]

To follow up here, @oscarfonseca , has told me he is in email chain with Wired8516 and will continue to work with him there.

[Wired8516]

After further investigation we found the component that is heating up causing the short condition we believe.  This board is being sent back to NI for further follow up.  The U21 component is heating up to 130 deg C very quickly during power applied.