Switched Wiring to/from the Power Supply?

[bitSync]

I've installed (not bolted in yet) a fanless Seasonic power supply in d8b #2 and I think I'm wired up correctly but the d8b won't power up from the front panel switch. I had this same experience with a couple of other power supplies I tried recently and I figured I was just overlooking something. The Mackie Linear Power Supply PC Board gets a dedicated Molex cable from the stock power supply (J200, I think). Is there something special about the way that cable is wired inside the power supply that enables power switching from the chassis front power switch? What am I missing?

[bitSync]

Well, I learned a little bit off of Tom's Hardware today. The original power supply that was in there was an AT style power supply with P8 and P9 plugs. This original power supply (AT/LPX) was designed to be able to be powered up and down with an AC switch on the chassis front panel whereas my new Seasonic fanless ATX style power supply uses pin 14/16 on the motherboard power connector for a momentary sense of the absence of the +3.3vdc to +5vdc PS_ON signal.

I need to figure out if and how this new power supply can be wired up with the obsolete AT/LPX front panel switch. Or, go with a momentary switch on the front panel that interrupts PS_ON. Anybody out there familiar with this?

[Edit 4/11/2014 - Updated "pin 14" to read "pin 14/16"; PS_ON is pin 14 on a 20-pin connector and pin 16 on a 24-pin connector]

[bitSync]

I did find Steve Hammond's excellent tips article -

http://www.sonido-7.com/d8b/tpis.html#Post15

but I need to investigate further.

[maxcz]

Simply connect the PSU wire at pin 14 (it is green, isn't it?) to ground (one of the black wires).

[bitSync]

Simply connect the PSU wire at pin 14/16 (it is green, isn't it?) to ground (one of the black wires).

I'm not sure I understand your post. Yes, momentarily grounding pin 14/16 will drop the voltage on that pin to switch on the ATX power supply. Also once running, grounding pin 14/16 for several seconds will shut down the ATX power supply. This presumes the presence of a momentary switch performing that function, a switch that is not found on the d8b CPU/PSU chassis. In neither of these scenarios is the front panel power switch playing a part in powering up or down the system.

Are you suggesting permanently grounding pin 14/16?

Also, in your wiring proposal, how is 115vac getting to the Mackie PSU in the chassis?

I think I have a wiring scheme sorted out but I have yet to test it. I'll report my results when I do. Thanks.

[Edit 4/11/2014 - Updated "pin 14" to read "pin 14/16"; PS_ON is pin 14 on a 20-pin connector and pin 16 on a 24-pin connector]

[maxcz]

Yes,ground it permanently. The PSU will be always on then. And I would get the mains power from after the front panel power switch and feed it to the atx PSU, so you have the original Mackie-switch-it-on-and-it-runs - experience...

[bitSync]

Yes,ground it permanently. The PSU will be always on then. And I would get the mains power from after the front panel power switch and feed it to the atx PSU, so you have the original Mackie-switch-it-on-and-it-runs - experience...

I see. Well, at the moment, my plan is a little bit different. I intend to install a new IEC mains entry on the back panel that will supply 115vac to to the power distribution board J1 and J2 plus ground (as the AT power supply pass-through currently does), then J5 and J6 plus ground will exit a new switched outlet on the back panel that will supply 115vac to the ATX power supply by way of a short IEC power cord. In this scenario, the power supply is powered up only when 115vac power becomes available at the back panel switched outlet, switched by the standard front panel DPST rocker switch.

I plan to take advantage of the BIOS Power Management 'PWR Lost Resume State' to cause the ATX to power up upon sensing 115vac present on the new switched outlet. I'm also planning on adding a momentary switch to be able to close ATX-SW (J9 header pins 7 and 8 (essentially the same function as pin 14/16 on the power supply cable)) on the mainboard to take advantage of that ATX feature, should the system ever get stuck with 'PWR Lost Resume State' not powering up on sensation of applied 115vac.

Please let me know what you think about this approach. I haven't started cutting sheet metal on the back panel yet

[Edit 4/11/2014 - Updated "pin 14" to read "pin 14/16"; PS_ON is pin 14 on a 20-pin connector and pin 16 on a 24-pin connector]

[maxcz]

Hmm, too many switches for my liking.
I don't see the advantage of using the mainboard to turn on the PSU. If you ground the green wire it will always be on.
I personally like my modifications to be invisible, so if I had to change the PSU it would have to emulate the old one as perfectly as possible - without additional switches and holes in the case

[maxcz]

As an addendum: I am no d8b psu expert - I just know as much of it as I have seen while switching from HDD to CF... but I do have a very good knowledge of ATX and old computer mainboards

[bitSync]

...And I would get the mains power from after the front panel power switch and feed it to the atx PSU, so you have the original Mackie-switch-it-on-and-it-runs - experience...

The front panel power switch mains are supplied (indirectly) by the original AT PSU through the power distribution board J1 and J2. By removing the AT PSU and replacing it with an ATX PSU, that 115vac to J1 and J2 is no longer available. The 115vac now has to come from somewhere. Where are you proposing the chassis get 115vac from? Just trying to understand your idea.

I'm warming up to your permanent pin 14/16 grounding approach. I think the only thing you lose is the ability to do a soft power down from a 4-second close on pin 14/16, but if you have the power switch, I guess it doesn't really matter.

I still need to get 115vac into (and back out for the PSU IEC pwer in) the chassis by some new means though. Thanks.

[Edit 4/11/2014 - Updated "pin 14" to read "pin 14/16"; PS_ON is pin 14 on a 20-pin connector and pin 16 on a 24-pin connector]