V5 White Screen of Death


#21

Today in the BC provincials, we had the WSOD 4 times!!! We are on the latest firmware and as far as I know where using different batteries, so it couldn’t have come from that. I noticed that it always happened after a bump into something metallic (another robot or the field boarder). We tried adding a grounder (a little piece of metal attached to the robot that drags along on the field) to attempt to ground the robot to the field, but it did not work (the next match a robot bumped into us and it WSOD’d again).

This issue did not only happen to us but other robots as well. Interestingly, it happened a lot on one certain field and never (as far as I’ve noticed) on any other.


#22

I am pretty Shure that, at least one cause of this, is a static shock


#23

I had a team report repeated “white screens of death” to me today at the BC Provincial Championship.

The fields had been static sprayed (given the high-ish relative humidity of the west coast we haven’t sprayed for the past ten years and it hasn’t been a problem… but V5)… and I only heard about it from one team, and it happened to them multiple times… they were a relatively junior team. Our senior teams (and the elimination rounds, to my knowledge) proceeded without such disruption.

It suggests to me something robot-dependent, such as cabling or firmware version.

Jason


#24

Is white screening caused by the same thing that kills brain ports when you use long custom length wires?


#25

Well we have experienced WSOD on sprayed fields with stock cables. So we are focused on impacts being the cause. The team mounted the brain with foam tape in between the brain and the structure. Went through a lot of impacts, and did not suffer as man WSOD, only one during regionals. Team reported it was after being hit the hardest ever. So my thinking is still along the lines that WSOD is a result forceful impacts.


#26

I remember being told that it is caused by a memory leak in your code.
If that is correct, than it is probably a good idea to look through your code and not have memory leaks.


#27

Memory leaks - Never saw that mentioned.

Will code review and see where it could happen …

doubtful


#28

What’s a memory leak?


#29

Oh please…


#30

Not your code, I mean code written by and used by a user. I think its important for me to note that I’m not sure if that’s the actual cause.


#31

Impacts seems to be the most probably cause of WSOD, but what is being effected by the impacts seems unknown. battery connections perchance? It would be nice if a team that often WSOD’s could try and recreate this issue.


#32

The only way I have ever been able to reproduce WSOD is with momentary loss of power to the V5 brain, and even then, creating a test condition to reproduce that was difficult. ESD may also be a cause, I’ve not been involved with any testing in that area.

I was at one of the CA state championships yesterday, I didn’t see any teams with this issue, the majority of teams were using V5 (probably more than 80%). I did see one team suffer from a disconnect that required reboot of the controller, ironically it was my team :disappointed:


#33

Epic answer :). Epic Question :slight_smile:

Given fields were sprayed, indoor humidity ok, the only causes we have seen are probably due to impact.

I am open that I write awful code that continually malloc() without free()… but given the code the middle schoolers have is not that complex - pretty sure memory leak is pretty far fetched …


#34

I couldn’t see any code my kids write to where there could be memory leaks… don’t see any memory allocation or anything beyond basic primitive data type usage


#35

We have the WSOD. State is on Saturday, we have tried everything and it won’t go away. Any suggestions?


#36

Pull out the battery and put it back in. If you want to prevent it, touch a metal part of your robot base to a field wall before every match to reduce static.


#37

some things you should make sure you do if possible:
switch brain
switch battery
switch radio
switch controller (why not, you never know)
switch battery cables
switch all cables
if all else fails then Idk what do say… except you can’t WSOD on a cortex…


#38

We can still drive but the screen stays white no mater what.


#39

odd. I’ve never heard of that happening before, from what I’ve heard usually the robot turns into a brick when that happens.


#40

I have had WSOD’s and sometimes the brain completely turning off when my robot takes impacts. In my experience, it seems to be caused by the battery cable connectors.