I have a similar project and will be tapping into those same pins. While searching for the jumpers with wires to connect securely I am having trouble finding 2.0mm pitch connectors. Most of what is sold is 2.54mm. Does Intel sell this accessory or do I need to order from china(currently in US). Any help would be appreciated. Thanks.
In automotive applications, the vehicle electrical circuits can experience a number of "transients". Please see the following white paper from Advantech: http://www.advantech.com.tw/eservice-applied-computing/newsletters/white%20paper/Power_Challenges_Faced_by_Vehicle_Applications.pdf
So they say it isn't a good idea to connect sensitive computer electronics directly onto a car battery. A good DC to DC Power Supply designed for use in automotive applications will buffer the incoming electrical to filter out the typical automotive transients and output a clean stable voltage which will not harm the sensitive computer electronics.
Some of the power supplies out there are expensive (think high amperage / high wattage police car power applications). I bought mine for about $75.00. It is a small price to pay for peace of mind and stable long-term operations. I justify the cost in my mind as a form of insurance.
The Intel specifications indicate the NUC runs on "DC Power 12 - 19 V, 65 W". At a voltage on the low end or even slightly below manufacturer's specification the electronics "might" work but if they do, they have to work harder and they put out more heat due to the inefficiency. (Not to mention possibly voiding the warranty.)
For my application, I would prefer not to "test" the limits of the NUC or risk shortening the life span of the unit. By using a "buck boost" capable power supply I am able to use any DC source ranging from 6 volts to 34 volts and still get clean, stable output power to the NUC at a single stable voltage the NUC prefers. (Also... by using an external DC to DC power supply, the heat generated by voltage conversion is kept outside the NUC case where it can be safely dissipated without overheating the NUC board.)
The intelligent power supply (DCDC-USB-200) paired with my NUC will also send a shutdown command to the computer if input voltages reach min or max levels I've set inside the power supply. (Because my DCDC-USB-200 also has a USB port, I'm able to use the NUC to monitor input power and output power as well as script the power supply response to various voltage level changes.)
With this setup, I feel pretty good that the computer will be kept isolated from issues that automotive power would normally introduce. My setup may be slightly overkill, but my tolerance for risk is also quite low in this application.
Anyway, hope this helps.
Here is what I have learned so far about the Intel NUC D54250WYB Front Panel Header:
- Pins 1 and 3 ( Hard Drive LED): Pin 1 is Positive 1.5 volts, Pin 3 is Negative
- Pins 2 and 4 ( Power LED): Pin 2 is Positive 4.8 volts, Pin 4 is Negative
- Pins 5 and 7 ( Reset): Pin 7 is Positive 3.27 volts, Pin 5 is Negative
- Pins 6 and 8 ( On/Off): Pin 6 is Positive 3.28 volts, Pin 8 is Negative
- Pin 9 (5 Volt): Pin 9 is Positive 4.8 volts.
These figures are as measured by multi-meter on the pins.
As I think I understand the way pins 6 and 8 work (and likely pins 5 and 7 as well), the pins are supposed to be shorted to ground in order to send the on/off signal to the motherboard? I could be wrong but I believe the 3.27/3.28 voltages are reference voltages that are measured by the board. When the board senses current flow on pin 6 (or pin 7) then the board triggers an event that is captured by software which then tells the computer to turn on or off depending on the current state ("On" if computer is off, "Off" if computer is on).
Alternatively, it could be that the negative pin 8 is being measured for voltage - causing the trigger when shorted to pin 6. I'm not real clear on how it works. What I do know is that the DCDC-USB-200 when activating the on/off switch is shunting the two pins to ground when it fires an "on/off" command.
I'm guessing a little here as there is very little if any documentation of how this works that I have been able to find so far and certainly no information specific to this board. It would be great if Intel could confirm for us how this works for D54250WYB?
I had the same trouble. You might find it easier to purchase the Schmartboard jumpers from Radio Shack for about $7. They work pretty well with the pins on the NUC.
Schmartboard 5" Male/Female Jumpers (10) with Headers (40)
Model: 920-0022-01 | Catalog #: 276-156
I was able to find them at my local store.
Hope this helps.
I am not saying that you don't need to do something to handle transients; I am simply saying that there is no absolute need (for this particular NUC anyway) to up-convert to 19V. Pick the output voltage (12V-19V) that is the least costly. This could be what you've chosen or maybe you want to look again (if not too late). Frankly, I don't think you are being too conservative. As Andy Grove said, 'Only the paranoid survive'...
Looking at the spec on this DC-DC converter and I am concerned that they have a low voltage shutdown at 11.4VDC, not a good value for deep cycling batteries as in my case I would lose my NUC and all of the functionality it provides, some critical, at roughly 60% remaining state of charge. Not good. As for pins 6 and 8, this is just a logic low that requires a momentary short to energize the NUC and a longer input to shutdown. I have designed much of my control logic around this simple fact. The DC-DC converter is nice and offers a lot of protection for the NUC however it is quite expensive and the additional protection could be effected with a few simple discrete components or modules for filtering and surge suppression. My power supply thankfully is massive in capacity and provides extremely clean straight line DC as observed on a 60MHZ scope under all source/sink conditions, some of my power control logic deals with a condition where deep cycles might be called upon to start a marine diesel engine normally isolated from the starting battery system and where high surge currents and electrical noise from mechanical energy might actually be an issue, in this case DC power to the NUC is opened for complete protection and self restoring post off normal start logic conditions. They can deal with a restart on the NUC which is less than 35 seconds for W8.1 Embedded Pro!