Here are some readings:
POWER_PB_B = 1.95v
PS_POR_PB_B = 3.3V
VSYS_IN = 12.273
VCC_5V0 = 0V
VCC_PSAUX = 0V
Pressing the on button raises VCC_5V0 to 5V and then it falls to 0.
Pressing the on button raises VCC_PSAUX to 0.8V and then it falls to 0.
These were just done with a meter, I can grab some scope readings from VCC_5V0 and VCC_PSAUX if that would be usefull?
How long have you had the board for? If you have had it for less then 60 days, I would suggest you contact your local Avnet office for an RMA as the product is still under warranty.
If you have had it for more then 60 days, I can do my best to help you trouble shoot.
Farnell in the UK have replaced it for me.
Haven't tried the replacement yet, fingers crossed!
Were you running anything intensive on the board when it failed? Was the fan running? This is the first failure like this I've heard of, my first thought is potentially thermal so I was curious about the run conditions.
Fan was running up to the point it died.
Nothing intensive was going on, linux was not running. I was testing out a PWM AIX component to one of the gpios and stepping through the code in the debugger over jtag.
there seems to be something similar maybe here: http://community.avnet.com/t5/Other-Avnet-Boards/Ultra96-board-won-t-power-up/td-p/19573
Is there no thermal cutoff ob the board?
I had in previous days used the gpio to turn the fan on and off, maximum fan off time would have been < 10 secs.
I am going to be attaching the design I am working on and to the fan signal in the coming days do I have to be aware that the board won't power down if temps get too hot?
Reviving this thread since I have the same problem, just with an Ultra96-V2. Reading 12.1 V on VIN, 3.32 V on 3V3_PRI, 0 V on 5V0. Curiously, I'm also reading 0 V on the KILL_N input to the ON/OFF controller and the same for the EN output of it (which is obviously why the 5V regulator does not turn on). Has anybody looked into if this might be a design flaw, particularly regarding the ON/OFF controller circuit? What can I do to bring my board back to life?
Also, does anyone know where to find the datasheet for the SLG4G42480V ON/OFF controller? Google tells me nothing beyond that it's probably manufactured by Dialog and distributed by Avnet...
Hi, bumping since I have the same problem. A coworker came to me after the Ultra96-V2 he is working on seemingly died.
The 3V3 LED and other indicators would light up for only a second after pressing SW4.
I traced the problem back to the interrupt controller SLG4G42480V ON/OFF controller. The signal KILL_N is pulled low as soon as VCC from the power controllers is pulled high, which effectively kills the power.
I found the datasheet for SLG4G42480V but have no idea how to pull out of this loop and be able to boot up again.
Anybody else had to deal with this?
I'm looking into this now, one of us will respond shortly. I'm digging through to see if we were able to root cause this failure. If the on/off controller failed we could possibly get around that, but if it is shutting down because something else failed then disabling it would just further fry the board. I'll get back to you asap
2 of 2 people found this helpful
ok, so in the past instances something failed on the board.
KILL_N is an output of the ZU+ on MIO34. If the user creates a design that mis-handles MIO34, then they could be inadvertently forcing the shutdown. They could try one of our known-good designs, like our out-of-box design to see if it works with that. If it does, then they need to fix their own design.
If it still fails with one of our known-good designs, then it’s likely a board failure.
That being said, I would check on when you purchased the board and see if it might still be under warranty. We have seen very few of these cases so it doesn't point to a specific issue that could be addressed. Sorry I couldn't be more help.
Thank you for your reply. It does still fail with known-good designs, so I'm going to look into the purchase date- I believe it might still be under warranty.