[kucol]

I have a problem with installation of Pulse Eight internal hdmi-cec adapter in my Intel NUC (D54250WYKH).
I did proper modifications to wiring according to manual (https://www.pulse-eight.com/store/download/get/28),
all signals seems to be connected OK, but adapter is not being recognized in OS (openelec) - no messages in dmesg about finding adapter.
I read some posts of other users saying that it might be a problem with providing 3.3V from CSH header - in my case measuring voltage
on adapter pins gives only 2,6V. Has anyone managed to get this configuration to work??

[maber23]

Hi!

Did you solve your problem? I'm very interested in the internal adapter.
Can you post some photos of your installation? Where did you connect the white wire (the power button wire)?

Thanks!

[kucol]

No, I did not.
Power button wire goes to the 6th pin of Front Panel Header - pinout is described on page 45 of technical documentation
(http://downloadmirror.intel.com/23089/en...Spec06.pdf).
I'm still waiting for reply from Intel (I created support request) - Pulse Eight support wasn't able to help with this case, but their sales office was very helpful - i'll tryout external version of their adapter.

[tavoc]

I bought this adapter as well for my non-NUC Board. I had to update the firmware to v4 before I could use it in XBMC.

Do you see the adapter in dmesg (some USB device)? If yes then you need the firmware update.

[kucol]

No, there are no entries in dmesg about finding adapter - I already wrote that in my first post.

[maber23]

No, I did not.
Power button wire goes to the 6th pin of Front Panel Header - pinout is described on page 45 of technical documentation
(http://downloadmirror.intel.com/23089/en...Spec06.pdf).
I'm still waiting for reply from Intel (I created support request) - Pulse Eight support wasn't able to help with this case, but their sales office was very helpful - i'll tryout external version of their adapter.

Thank you for your reply! keep us update on the support request!

[tavoc]

I know that you could not find it in dmesg. The internal adapter has two connectors, one for hdmi and one for usb.
If you connect usb there should be something. Have you a spare computer where you can attach the usb part to an usb header?

A live ubuntu 12 OS should be okay for a quick dmesg and an update.

[kucol]

So you're saying, that connecting only USB part of wiring to USB header should allow me to see the device?
Can you confirm that??
If that's true then I should see it in OS regardles of lack of 3.3V from CSH header, even completely disconnecting CSH wires, and that's not what's happening. That would mean that I:
- have something wrong with my wiring (which i already checked 100 times...)
- have something wrong with USB header in NUC - this I can check connecting different device
- have something wrong with CEC adapter itself

Are you able to test the case, when internal adapter is only connected to USB header?
Is it really recognized and put entries in dmesg?

[tavoc]

I could try this tomorrow evening.

As far as I could remember I updated the device on a different PC where there was no HDMI Connector. Therefore only USB was connected

[kucol]

I'd appreciate if you try - that would help to explain a lot of things with my problem.

[kucol]

OK - so I performed more tests and now I have some answers:-)
1) I created additional USB wire/adapter to connect internal CEC adapter directly to USB port of my notebook.
tavoc - unfortunately your theory was wrong - providing only USB signals to internal CEC adapter (pins 9-12) is not enough - it does not allow to upgrade firmware nor use cec-client to test it.
So then I added two more wires to pins 1&2 with 3.3V from external power supply and.......it worked - I successfully upgraded firmware to version 4.



2) I got back to my NUC configuration with new firmware of CEC adapter.
Still no change - no entries about CEC adapter in dmesg, cec-client couldn't find any connected adapter
Digital multimetr still shows 2.6V on CEC adapter's pins 1&2

3) I slightly modified my wiring and provided 3.3V to CEC adapter from SATA power connector (orange wire)
and..........everything works! CEC adapter magically appeared in dmesg and XMBC's peripherials menu!





SATA power connector gives rock stable 3,3V measured on CEC adapter's pins 1&2, but it's not originating from standby rail and because of that, internal CEC adapter has lost one of its greatest advantages over external version - it can not wake NUC with CEC signal:-(
I'm not satisfied with this solution, because it's not very elegant, especially if you want to connect SATA drive and also because in this configuration it doesn't have functionality I bought it for (power up on CEC signal) - in this case external CEC adapter is in my opinion better and safer option to buy. I wasn't also able to find 3.3V Standby signal from any other header/connector on the motherboard, and I don't want to solder anything directly to it, because it would void my warranty from Intel :-(
There is certainly something wrong with 3.3V standby voltage from CSH header, so I'm still waiting for answer from Intel to my support request, but I found something unsetting in their technical documentation - although in this article (http://www.intel.com/support/motherboard...034631.htm) Intel claims that 3.3V standby pin from CSH/HTPC header "Can be used to power custom solution (such as daughtercard, etc.) with up to 2 A of current rating capability", but in technical documentation to NUC motherboard (http://downloadmirror.intel.com/23090/en...Spec06.pdf), on page 47 (2.2.4.5) you can only find statement that "3.3 V Standby: can be used to monitor the presence of 3.3 V standby power". In "Specification Changes or Clarifications" on page iv you can also find that in March 2014 there was "Spec Clarification" regarding section 2.2.4.5, but I don't know how it looked before this change:-(

[tavoc]

Nice find.

How about taking the 3.3V from USB through a 5v to 3.3v resistor?
Sometimes it is possible to enable USB power in standby.

Have a look at this:
http://www.missingremote.com/news/2014-0...h-33v-flaw

you are not the only one with this problem. Seems related to your NUC.

[kucol]

Yes, I'm thinking about converting 5.5V stb to 3.3V stb - CSH header provides 5V Stb signal.
But rather than using resistors I was thinking about linear regulator + maybe some capacitors.

So apparently problem is related to this specific model of NUC :-(
I'm still waiting for answer from Intel support in this matter.

[Stereodude]

How about taking the 3.3V from USB through a 5v to 3.3v resistor?

You could use a regulator, but you can't use a resistor to turn 5V into 3.3V unless the current draw is absolutely constant.

[maber23]

... but in technical documentation to NUC motherboard (http://downloadmirror.intel.com/23090/en...Spec06.pdf), on page 47 (2.2.4.5) you can only find statement that "3.3 V Standby: can be used to monitor the presence of 3.3 V standby power". In "Specification Changes or Clarifications" on page iv you can also find that in March 2014 there was "Spec Clarification" regarding section 2.2.4.5, but I don't know how it looked before this change:-(

HI!

I have the revision 1 of this document, here is the specification:


revision 1:

2.2.3.5 System ID / Custom Solutions Header (2.0 mm Pitch)
The System ID / Customs Solution header is provided to aid customers in developing custom applications.
• Prog_LED#: general purpose signal output that indicates when an event was triggered by the operating system. Signal is amplified by a transistor. Intel can provide sample code for customers who may want to write their own applications leveraging this signal.
• SMB_CLK and SMB_DATA: SMBus interface, reserved for future support of All-In-One chassis detection. General SMBus information can be found on the platform EDS and at http://smbus.org/specs/.
• 3.3 V Standby: can be used to monitor the presence of 3.3 V standby power.
• PWRBT#: power button signal (functions in the same manner as the power button pin on the front panel header).
• HDMI Consumer Electronics Control (CEC): standard communication signal from the Mini HDMI connector (http://www.hdmi.org/) - the signal is exposed through this header for third party solutions to monitor/control CEC activity between multiple HDMI devices.
• 5 V Standby: can be used to monitor the presence of 5 V Standby power or provide power from the 5 V Standby rail (up to 2A current rating).
• SCI/SMI GPI: input signal for direct connection to a front panel push-button to trigger a Windows command. Intel will be adding BIOS support and accompanying Windows utility to enable Direct Application Launch* feature. General information about Direct Application Launch can be found at:
http://msdn.microsoft.com/en-us/windows/...63078.aspx

revision 6:

2.2.4.5 System ID / Custom Solutions Header (2.0 mm Pitch)
The System ID / Customs Solution header is provided to aid customers in developing custom applications.
• Prog_LED#: general purpose signal output that indicates when an event was triggered by the operating system. Signal is amplified by a transistor. Intel can provide sample code for customers who may want to write their own applications leveraging this signal.
• DMIC_CLK and DMIC_DATA: clock output and data I/O for digital microphone interface.
• WDTO#: Watchdog timer that provides a general purpose implementation that can be wired to external systems (i.e., industrial controls) or to the rest pin on the board. Interested customers usually write their own software for watchdog timer support.
• 3.3 V Standby: can be used to monitor the presence of 3.3 V standby power.
• HDMI Consumer Electronics Control (CEC): standard communication signal from the Mini HDMI connector (http://www.hdmi.org/) - the signal is exposed through this header for third party solutions to monitor/control CEC activity between multiple HDMI devices.
• 5 V Standby: can be used to monitor the presence of 5 V Standby power or provide power from the 5 V Standby rail (up to 2A current rating).
• SCI/SMI GPI: input signal for direct connection to a front panel push-button to trigger a Windows command. Intel will be adding BIOS support and accompanying Windows utility to enable Direct Application Launch* feature. General information about Direct Application Launch can be found at:
http://msdn.microsoft.com/en-us/windows/...63078.aspx

So no change in 3.3 V. Stanby.