Can the NUC8i7BEH USB ports be disabled until boot?

[RLope16]

I have a system that uses a NUC that turns on using WOL and it senses +5VDC on the USB port when the NUC boots to activate other equipment.

When first applying power to the NUC rear connector, the NUC7i5 initially sends a quick +5VDC pulse to the USB ports and then goes to 0VDC.

When the NUC front panel power button is pressed and the unit boots, the USB ports go live with +5VDC.  

When the OS (Ubuntu 16.04LTS) executes shutdown -h now, the USB ports go back to 0VDC.

I now have a new NUC8i7BEH (BIOS 0056) to replace a NUC7i5.  

When applying power to the NUC rear connector, the NUC8i7 USB ports go live with +5VDC. 

When the NUC power button is pressed and the unit boots, the USB ports continue with +5VDC. 

When the OS (Ubuntu 16.04LTS) executes shutdown -h now, the USB ports go back to 0VDC.

Is there a way for the NUC8i7 rear USB ports to be dead (0VDC) until the NUC is turned on, either by WOL or front panel button?

Seems the OS can command such a state after shutdown, but it is not possible when powered, but not "ON".

Thank you.

[n_scott_pearson]

You need to explain yourself better.

First of all, saying NUC7i5 is simply not specific enough; there are multiple NUC models that this could be referring to (NUC7i7BNxx or NUC7i7DNxx). Either use the complete string or use the portions that really matter. For example, you can refer to the NUC7i7BNxx using its 'Baby Canyon' code name or simply using 'BN'. If pointing out which processor is being used is important, then use i5BN. If pointing out what chassis type you have, you can include the 'K' or 'H' designations.

You can control whether the USB ports are powered in the S4 and S5 states from within BIOS Setup (the Visual BIOS app). Also, by enabled Fast Boot, you can have USB be ignored until the O/S is booted (though this makes it more difficult to get into BIOS Setup).

...S

[RLope16]

Scott, thank you for responding.

It is a 7i5BNH.

I should be able to "control whether the USB ports are powered in the S4 and S5 states from within BIOS Setup."

With the S4,S5 USB power button marked off and Fastboot enabled, this 8i7BEH rear USB ports latch powered on +5VDC when 12VDC is initially applied to the unit. 

The NUC remains off (LEDs off, I'd assume at S5) with the ports powered on.   When the NUC is booted and shutdown, the ports then become unpowered. 

Is there another BIOS option that would disable power to the rear USB ports until the unit boots?

Thanks.

Roberto

[n_scott_pearson]

If you mean the case where power has been freshly applied, my understanding is that the answer is no. Unfortunately, the hardware powers up before the BIOS has a chance to kick in and disable it. It will be properly controlled only on the subsequent transitions down into S4 and S5.

Wait - are you saying that, when power is first applied, the ports are powering on and staying powered? You should have only seen a short powering on and then the port should have been powered off. It this is the case, they may have an issue to look at.

Sorry,

...S

[RLope16]

No problem, Scott. Appreciate your response.

Yes, it doesn't sound right, doesn't it? It shouldn't latch the power on. Hoping to get some Intel staff to confirm.

Regards,

Roberto

[n_scott_pearson]

For analysis, the team is looking for answers to these questions: How are you measuring power? What is the problem with power being there?

...S

[RLope16]

NUC rear USB port is connected to 7-port USB 2.0 hub. One USB cable from the hub is connected to a relay circuit with a 555 timer for a 1.5sec delay and a 5VDC relay (Fujitsu FTR-LYRA005V). The relay is used to send a trigger signal and power a custom 3 rail Acopian LPSU. The delay filters the initial voltage pulse from the NUC USB port when power is first applied to the NUC (not turned on, just power applied to the rear connector). So there is enough current to light the LED on the external hub, power the timer circuit, and close the Fujitsu coil (>40mA). 

Issue is, the system can operate with direct manual switch operation on the NUC as well as remotely using WOL. If the voltage on the rear port latches on upon initial power application, the rest of the system becomes powered, except for the NUC. That is not a desirable state.

We have workarounds. We'd change the 8i7 BIOS to boot upon power application. Instead of using the NUC's power button to start sequencing the power to the other units via the USB, we'd use the power button to the NUC power supply instead. The NUC can then be immediately disabled via http/php exec commands and this places in a state with USB power at 0VDC and then the remote WOL works like it did on the 7i5. Manual operation reverts back to the NUC power button.

Question we have is, is it a defective unit? If NUC8i7 continuous USB power upon initial power application is a valid state, we'll make it work. It just behaves differently than the 7i5 unit. Would like some confirmation before fielding the unit in case it is defective. 

Thank you, Scott, and all involved.

Regards,

Roberto

[Wanner_G_Intel]

Hello RLope16, Please let us know if you are still experiencing the issues described on this thread. We can try to address this concern as soon as possible. Wanner G. Intel Customer Support Technician Under Contract to Intel Corporation

[Wanner_G_Intel]

Hello RLope16, If you have further questions, we will be glad to help you. Wanner G. Intel Customer Support Technician Under Contract to Intel Corporation