Computer Case front panel power switch connector

[LFaWolf]

Hey all, I recently came across a problem that I am still baffled. I have built over 100 PCs for various people and reasons but have never seen anything like this. Perhaps my knowledge is wrong so it would be neat to learn something today.

I have a test machine that has been running without any issue. I recently decided to take the case, a nice Antec P280 for a different build with a larger motherboard, and move the test hardware to a Corsair 400R, a new case that is not used. After everything has been moved, I turned on the 400R. The CPU fans spun up, as well as the lights, DVD drive, GPU fan (a GTX 680) for a second and then the machine is off. I was puzzled as the test machine was running just fine previously. I proceeded to do all kinds of troubleshooting and finally, remove the motherboard and placed it on a test bench. Lo and behold it booted right up without any issue. At this point the only difference between the test bench and the Corsair 400R is the power supply and the case. I put the motherboard back into the 400R, and the same issue happened. I thought about replacing the power supply as it is getting old (a Silverstone Decathlon 800w). However, I thought the symptom of immediate shut off is reminiscing of a power "short", so with nothing else from the front panel connected, I used a screwdriver to "connect" the power pins of the motherboard, and lo and behold, it booted and posted!

At this point I figured the front panel power switch connector from the 400R is bad. I tried one more time but this time I reversed the polarity of the front panel power switch and again, the board booted up! So I figured, why not switch back to the previous polarity and see? Well, that didn't work!

My question is, and my understanding has always been, that the front panel power switch connector has no polarity. Its job is to "short" or connect the 2 power pins of the motherboard. This is similar to the reset switch, and unlike the HDD and power LED light that has + and - polarity. Am I wrong? What could be the problem with the 400R?

Since I switched the "polarity", the 400R has been running fine, turn on and off without any issues.

 

[Bill_Bright]

My question is, and my understanding has always been, that the front panel power switch connector has no polarity. Its job is to "short" or connect the 2 power pins of the motherboard. This is similar to the reset switch, and unlike the HDD and power LED light that has + and - polarity. Am I wrong? What could be the problem with the 400R?

No, you are not wrong. In fact, it is not "similar" to the reset switch, but exactly like it (electronically speaking) such that I have on many occasions were the power switch wore out, simply moved the wires from the reset switch to the power pins on the motherboard and told the users to use the Reset button from now on.

So what is happening here? My "guess" is the switch is broken and one side is already damaged, not making contact, or constantly making contact. Can you easily see the backside of the switch? If so, see if damaged or perhaps pushed out of its retainer.

 

[sneekypeet]

All front panel wiring has polarity built in. If you look closely to the connectors, one side has an arrow on it for the positive connection. Also if you look at the motherboard they note positive and negative. As many cases as I have gone through over the past decade, only a handful matter with the power switch orientation, but there are a handful where it matters, and why it is best to address polarity when connecting them.

 

[jboydgolfer]

Reverse direction of power header plug

Its likely a header related issue

 

[DeathtoGnomes]

All front panel wiring has polarity built in. If you look closely to the connectors, one side has an arrow on it for the positive connection. Also if you look at the motherboard they note positive and negative. As many cases as I have gone through over the past decade, only a handful matter with the power switch orientation, but there are a handful where it matters, and why it is best to address polarity when connecting them.

This. Pretend there is a diode built in.

 

[Bill_Bright]

That's not how it is on this Fractal Design or the two Corsairs here. In fact, I have not seen a case where this matters. Got an example?

All front panel wiring has polarity built in.

If it mattered for the case, it would matter for the motherboard. And if it mattered for the motherboard, how would shorting the two pins in the motherboard header work? Adding a diode just adds costs.

Those switches are like a wall light switch. They just make, or break a circuit. Switches don't care about polarity.

Now the Power and Drive Activity LEDs certainly are polarity sensitive which makes sense - since the D in LED stands for diode.

 

[coonbro]

sounds like you got the pin out wrong on the board to the switch [ I done it there small and hard to see what the correct 2 needed bunched up with the rest in line with it ] or maybe the switch is now bad or shorted seeing it fires right up and runs when you pop start it . so I would think you plugged the switch to the wrong 2 pins or the switch is bad

remove the case switch from the case and try it on the test bed correctly plugged in the right 2 pins - if it works you goofed that if not then maybe a bad switch after all

I guess I should a be sure you don't have a rogue stand off shorting the board or something making contact that should not in that case

 

[jboydgolfer]

Got an example?

my Enthoo ProM was directional. incorrect installation caused power on, & then immediate off. switched around ,and viola! if thats what you mean by example, atleast in my case.i havd the Power Bttn reversed iirc.

 

[sneekypeet]

That's not how it is on this Fractal Design or the two Corsairs here. In fact, I have not seen a case where this matters. Got an example?
If it mattered for the case, it would matter for the motherboard. And if it mattered for the motherboard, how would shorting the two pins in the motherboard header work? Adding a diode just adds costs.

Those switches are like a wall light switch. They just make, or break a circuit. Switches don't care about polarity.

Now the Power and Drive Activity LEDs certainly are polarity sensitive which makes sense - since the D in LED stands for diode.

Just calling it as I see it man. I don't design them I just use them by the hundreds. I get what you are saying about jumping the pins, but as jboyd so eloquently answered, I am not the only one who has run into this in the wild.

 

[Jetster]

All front panel wiring has polarity built in. If you look closely to the connectors, one side has an arrow on it for the positive connection. Also if you look at the motherboard they note positive and negative. As many cases as I have gone through over the past decade, only a handful matter with the power switch orientation, but there are a handful where it matters, and why it is best to address polarity when connecting them.

I didn't know that. Thanks Makes sense they would mark it for a reason

Its also easy to get confused about the direction of the plug. Older boards have a different direction the plug goes. And of course in proprietary cases who knows?

 

[sneekypeet]

I didn't know that. Thanks Makes sense they would mark it for a reason

Its also easy to get confused about the direction of the plug. Older boards have a different direction the plug goes. And of course in proprietary cases who knows?

Best advice is to flip the connectors over and look for the molded arrow on them. It is always the positive lead.

 

[Bill_Bright]

I think it should be pointed out that IF cases or motherboards are requiring these front panel power and reset buttons be connected according to some polarity requirement, that would be a clear, non-compliance violation of the ATX Form Factor standard for PCs.

So if this is happening with factory made computers where "proprietary" modifications are not uncommon, then I can accept (though not happily) this is happening. But for any case that is sold through Newegg, Amazon or anywhere as an ATX compliant case or motherboard, then that is wrong and they need to be taken to court - and boycotted by everyone who builds computers for themselves or as custom computers builders like me, as a business owner, who builds PC computers for clients.

Again, I sure would like to see a link to an example where the front panel power and reset button of an "ATX" case are somehow polarity orientated, and for any motherboard claiming to be "ATX" (including micro, ITX, or EATX) that requires it.

 

[LFaWolf]

sounds like you got the pin out wrong on the board to the switch [ I done it there small and hard to see what the correct 2 needed bunched up with the rest in line with it ] or maybe the switch is now bad or shorted seeing it fires right up and runs when you pop start it . so I would think you plugged the switch to the wrong 2 pins or the switch is bad

remove the case switch from the case and try it on the test bed correctly plugged in the right 2 pins - if it works you goofed that if not then maybe a bad switch after all

I guess I should a be sure you don't have a rogue stand off shorting the board or something making contact that should not in that case

No short anywhere else, and no I did not have the pin out wrong.

That's not how it is on this Fractal Design or the two Corsairs here. In fact, I have not seen a case where this matters. Got an example?
If it mattered for the case, it would matter for the motherboard. And if it mattered for the motherboard, how would shorting the two pins in the motherboard header work? Adding a diode just adds costs.

Those switches are like a wall light switch. They just make, or break a circuit. Switches don't care about polarity.

Now the Power and Drive Activity LEDs certainly are polarity sensitive which makes sense - since the D in LED stands for diode.

I would like to ask the same thing, why would connecting the 2 pins with a screwdriver be able to turn the machine on? The motherboard in question is a ASRock Fatal1ty 990FX Killer.

 

[sneekypeet]

I think it should be pointed out that IF cases or motherboards are requiring these front panel power and reset buttons be connected according to some polarity requirement, that would be a clear, non-compliance violation of the ATX Form Factor standard for PCs.

So if this is happening with factory made computers where "proprietary" modifications are not uncommon, then I can accept (though not happily) this is happening. But for any case that is sold through Newegg, Amazon or anywhere as an ATX compliant case or motherboard, then that is wrong and they need to be taken to court - and boycotted by everyone who builds computers for themselves or as custom computers builders like me, as a business owner, who builds PC computers for clients.

Again, I sure would like to see a link to an example where the front panel power and reset button of an "ATX" case are somehow polarity orientated, and for any motherboard claiming to be "ATX" (including micro, ITX, or EATX) that requires it.

How is it not in compliance when both the connection and motherboard are both clearly labeled for polarity?

 

[Jetster]

There is no standard for the front header

 

[jboydgolfer]

Again, I sure would like to see a link to an example where the front panel power and reset button of an "ATX" case are somehow polarity orientated, and for any motherboard claiming to be "ATX" (including micro, ITX, or EATX) that requires it.

Search motherboard manuals, im sure you'll see something like this. But I have to assume there is some type of miscommuniation happening here, since this is pretty common knowledge, i think we are not talking about the same subject possibly.

here you can see the header pins for a Fractal Deign Meshify, clearly have both + & -, and under that, a Asrock motherboards header pin ports, also, + & -

Edit: The more I think of it, I feel like I do recall in the past it not mattering (or if not existing) in the power switch and in the reset switch in regards to polarity, but my memory has never been great

i found this .pdf Document pertaining to ATX FF guidlines/spec's, but there is no mention of specific implemenatation outside of placement on the PCB, and conforming to size and location restrictions/expectations.

Spoiler: ATX form factor standard (front I/O)

The exact location of the front panel I/O connector is not specified. It is recommended that it be placed along the front edge of the board to the right of the expansion slots. When placing the connector, the designer should keep in mind that proper clearance must be provided for the chassis peripheral bays. Locating the front panel I/O connector along the left edge of the board is not recommended due to limited clearance with a full length add-in card. Locating it along the front edge of the board under the expansion slots using a right angle header may be acceptable, provided that clearance for the add-in cards and mechanical retention of the mating connector is properly accounted for.

 

[DeathtoGnomes]

I think it should be pointed out that IF cases or motherboards are requiring these front panel power and reset buttons be connected according to some polarity requirement, that would be a clear, non-compliance violation of the ATX Form Factor standard for PCs.

So if this is happening with factory made computers where "proprietary" modifications are not uncommon, then I can accept (though not happily) this is happening. But for any case that is sold through Newegg, Amazon or anywhere as an ATX compliant case or motherboard, then that is wrong and they need to be taken to court - and boycotted by everyone who builds computers for themselves or as custom computers builders like me, as a business owner, who builds PC computers for clients.

Again, I sure would like to see a link to an example where the front panel power and reset button of an "ATX" case are somehow polarity orientated, and for any motherboard claiming to be "ATX" (including micro, ITX, or EATX) that requires it.

Not that I trust your word here, I dont, but what is "your" source for this non-compliance violation? In all PC's I built, polarity mattered for about 80% of the time, especially true if its a combo switch used as a reset.

 

[dorsetknob]

getting all heated up over a case switch
The header cable is probably marked + & - " for health and Safety as its an electrical connection
Probably mandated somewhere in California law and so the rest of the world follow's so they can Sell in California.

Does the Screwdriver i use to bridge the pins to jump start a breadborded Motherboard need to be + or -.

 

[jboydgolfer]

im not saying its carrying any serious current or anything, im just pointing out that these plugs and ports are indeed marked +/- (& sometimes GRND), and they do seem to serve some function beyond simply briging a connection , atleast in the cases ive built in.

 

[Ferrum Master]

If the switch got moist, worn out and then corrosive process and oxidation process is done some salts inside a switch membrane.

In my experience I have seen passive elements to gain active element behavior when damaged, it acts as a diode... ie semiconductor.

So it is normal... just replace it.

 

[trog100]

irrespective of what its says on the plug all the power switch does is bridge two contact pins.. exactly like a screwdriver does.. its a momentary connection and should break the connection when the switch is let go..

trog

 

[eidairaman1]

Search motherboard manuals, im sure you'll see something like this. But I have to assume there is some type of miscommuniation happening here, since this is pretty common knowledge, i think we are not talking about the same subject possibly.

here you can see the header pins for a Fractal Deign Meshify, clearly have both + & -, and under that, a Asrock motherboards header pin ports, also, + & -

Edit: The more I think of it, I feel like I do recall in the past it not mattering (or if not existing) in the power switch and in the reset switch in regards to polarity, but my memory has never been great

i found this .pdf Document pertaining to ATX FF guidlines/spec's, but there is no mention of specific implemenatation outside of placement on the PCB, and conforming to size and location restrictions/expectations.

Spoiler: ATX form factor standard (front I/O)

The exact location of the front panel I/O connector is not specified. It is recommended that it be placed along the front edge of the board to the right of the expansion slots. When placing the connector, the designer should keep in mind that proper clearance must be provided for the chassis peripheral bays. Locating the front panel I/O connector along the left edge of the board is not recommended due to limited clearance with a full length add-in card. Locating it along the front edge of the board under the expansion slots using a right angle header may be acceptable, provided that clearance for the add-in cards and mechanical retention of the mating connector is properly accounted for.

Yeah not all have +- on them, i just follow the color wire, if it doesnt work initially i flip it over. Nothing critical here

 

[Bill_Bright]

There is no standard for the front header

Sure there is. If there were no standard, how could you put any ATX motherboard from ASUS, Gigabyte, MSI, EVGA, ASRock, BioStar, XFX, or Foxconn, and install it in any ATX case made by Corsair, NZXT, Fractal Design, Antec, Cooler Master, Be Quiet!, TT, or Lian LI and expect them to be physically and electronically compatible?

The + or - you may see in manuals and on the connectors are there JUST to minimize "amateur" builder (no disrespect intended - I simply mean people not "formally" trained as "electronics technicians") from inundating case and motherboard tech support centers by calling in and asking, "Which wire goes where?"

IF the motherboard required, electronically, the proper polarity, using a screwdriver to short the two pins could result in unexpected results (the voltage is too low to cause damage). Or it could mistakenly dump the voltage to ground at the switch itself - depending on switch design.

How is it not in compliance when both the connection and motherboard are both clearly labeled for polarity?

Sorry, my error. I should have been more clear. What I mean is if the motherboard's circuit "design", from an electrical stand point of the circuit in question, required that current must flow through the switch in only one direction so the switch can only be installed one way, that would not be in compliance with the ATX Form Factor standard.

Here is an image of the circuit.

All you are doing when you press the button is "closing" that switch, connecting those two points in that PWR_SW switch together momentarily. That dumps a "low" on the circuit signaling the system to "change states" - power up or power down. You do the same thing with a flat tip screwdriver. What happens if you flip the screwdriver over? The same thing happens!

Got a multimeter? Disconnect the two wires from the motherboard front panel header leading to the switch. Stick the two multimeter leads into the wire connectors, observing the labeled polarity and set the meter to measure resistance. You should see an "open" (infinite Ω). Press the button and you should get a "short" (0 Ω). Flip the multimeter leads around and you should see the same results. IF polarity through the switch mattered, you would see an open regardless if the switch is open or closed.

And again, this only applies to the Power and Reset button switches. For sure, the LED power and drive activity connectors are polarity dependent.

Is it good practice to ensure you follow the labeling? Sure.

 

[Jetster]

Bill show me where ATX specification says to place the pin assignments? It doesn't, it only says where to play the I/O and its just a recommendation. It's just an agreement between manufactures. This is also why case manufactures do not group the Front I/O plug as one plug. They're separated. The point is you need to be sure which is which

 

[Bill_Bright]

Bill show me where ATX specification says to place the pin assignments? It doesn't, it only says where to play the I/O and its just a recommendation.

Exactly!

Note the ATX Form Factor standard only dictates pin assignments when it matters. For example, USB pin assignments use an industry standard pin layout. So does the main 24 pin power connector. What is standard are the voltages.

This is also why case manufactures do not group the Front I/O plug as one plug. They're separated

Right! But only partially. For example, the case speaker pinout is standardized and grouped even though the use of a case speaker is not required. But should you decide to add one, like one of these System Speakers, note the molded connector designed to connect in a specific pin layout of that front panel header. That is, the 4 (really 2) pins are laid out in a standard configuration, though they may be located in different places in the header.

BTW, the entire ATX Form Factor standard is an agreement between the manufactures. Too bad we don't have such an agreement for notebooks.