Thread: All-in-one Atom330/ION project

Here are some high-res pictures of the power button:
Front:
Back: There are 3 contacts. 2 for trigger, 1 for securing it to the PCB?
Underneath: not much here. Those two big contacts are way too far apart to be the trigger:

I know the contacts look pretty big in these pictures, but here it is, a bit zoomed out. It's that tiny black element in the top right corner, next to the LEDs:

Any advice? I think I can see the tiny metal thing that's holding the button housing together, but it'd be nicer not to destroy the button permanently.

Cut off the Nano-ITX stands in the black box:


Peeled off the wifi antennas from the nettop's case (I used a small metal pick):

Originally Posted by nivanov

Here are some high-res pictures of the power button:
Front: *omitted pic*(no reason to repost them-- imo, takes up too much space..)
Back: There are 3 contacts. 2 for trigger, 1 for securing it to the PCB? *omitted pic*
Underneath: not much here. Those two big contacts are way too far apart to be the trigger:*omitted pic*

I know the contacts look pretty big in these pictures, but here it is, a bit zoomed out. It's that tiny black element in the top right corner, next to the LEDs: *omitted pic*

Any advice? I think I can see the tiny metal thing that's holding the button housing together, but it'd be nicer not to destroy the button permanently.

ok, it is a really really small switch..

you might need to try to check the switch with a ohm meter-- 2 contacts should short every time you press the button. i am thinking the center contact should be one..

the switch might be a single pole, double throw(commonly called spdt), so you might need to use a relay to trip the contacts to fool the pc correctly...

could you please elaborate? It's not enough to just trip it with the mobo signal switch? The way I see it, a button closes a circuit and thats pretty much it - why would i need a relay? Sorry if this is a dumb question - electronics aren't exactly my specialty

Edit: I just measured the resistance. The side contacts as well as the metal shroud around the button are one of the poles. The middle one reports no contact when the button is depressed and full contact when the button is pressed. So the good news is I only have to solder one tiny contact. The bad news is that that contact is in the middle of the other 2 tiny contacts. I really don't think this kind of soldering job is doable with my current tools. Unless there's a needle-thin soldering tip, I think I have to try something else.

Originally Posted by nivanov

could you please elaborate? It's not enough to just trip it with the mobo signal switch? The way I see it, a button closes a circuit and thats pretty much it - why would i need a relay? Sorry if this is a dumb question - electronics aren't exactly my specialty

Some buttons use an spdt design which means that the button closes two circuits at the same time...... not the case in any motherboards i've ever seen. I would avoid soldering unless you really have to-lets look at a software solution.

In the system bios is there any option for what to do after a power failure? (like automatically power on)

In the bios is there any option to turn the pc off after the power button is held for 3 seconds? (and can that option be disabled)

most motherboards have 2 pins-- you short those pins, and the motherboard turns on...

your board has a 3-pin switch-- so the motherboard could be normal, and just be using 2 pins for a power-button signal.. but it could also require all 3 pins-- possibly for other features(maybe something that needs power when the button is not pressed?)

i would probably try to power it by shorting 2 pins first, after figuring out which pins are normally open..

but with the 3pin switch, you might need to remove, or relocate the switch to allow for other wires(i would probably extend it a couple of inches [50-80mm ] off the board-- to make it easier to wire other things to it..)

but all of this needs to be done after determining what contacts get shorted when the switch is in the normal position, and when it is pressed down..

and another thought-- it could also use ground at certain contacts as well, so you'll need to check that as well..



you can forget about the relay thing-- that's me thinking too far ahead.

assuming the switch is a spdt, you would need a relay to short the 2 normally-closed contacts, and then, when you want to turn the pc on, you would energize the relay so it closes the other way..

but enough theorizing , i could be completely wrong-- you'll need to check the switch first..

So the fact that the shroud as well as side contacts are all pretty much the same contact (I touched all of combinations of side contacts as well as the shroud and they are all reporting conductivity) means that it's not a SPDT, right? It's not like for contacts 1 2 3, 2-3 are shorted when the button is off and 1-2 are shorted when the button is pressed (that would be an SPDT scenario, if I understand it correctly)

JustChat_1 - there are power options. I'm going to look up exactly what they are tomorrow. Worse comes to worst, I can get a fusion brain chip to monitor the Off signal and shut down the computer that way and try to set it to resume on power restore. Is that what you were thinking? Is there any other way you can think of to shut down the computer when the car gets turned off?

EDIT: haha, we are all posting/editing way too fast. I'm gonna let the dust settle until tomorrow

i'm gonna wait for justchats non-hardware ideas before commenting further on hardware/soldering solutions(a software solution might be easier and less destructive.)

Originally Posted by nivanov

So the fact that the shroud as well as side contacts are all pretty much the same contact (I touched all of combinations of side contacts as well as the shroud and they are all reporting conductivity) means that it's not a SPDT, right? It's not like for contacts 1 2 3, 2-3 are shorted when the button is off and 1-2 are shorted when the button is pressed (that would be an SPDT scenario, if I understand it correctly)

Yea 3 contact switches are more secure (3 points securing it to the board) but motherboards never need more then an on signal.

Originally Posted by nivanov

JustChat_1 - there are power options. I'm going to look up exactly what they are tomorrow. Worse comes to worst, I can get a fusion brain chip to monitor the Off signal and shut down the computer that way and try to set it to resume on power restore. Is that what you were thinking? Is there any other way you can think of to shut down the computer when the car gets turned off?

I'm usually the first to recommend a fusion brain for anything....but in this case you can do it much cheaper with either a simple relay adc, serial to usb adapter or even a relay on the cpu fan sense wire if you want to get creative.