[dantheman]

Need help. I bought a used iMac DV 400 motherboard. Now need to put into an ATX case with ATX pwr supply etc... From web search, I know the ATX PFW signal control is reversed for the iMac. Simple solutions on the web range from single transistor/opto jobs to pull line low but this means iMac can start PS, not shut it off. XLRYOURMAC has a simple inverter circuit with LS04 that's more logical, should be able to pull ATX line high to shutdown. But Takahi Imai's site <http://www.bekkoame.ne.jp/~t-imai/think/w032.html> has a inverter circuit with an '04 too but with additional bits. What are they for? I'm not an electronic engineer. I've copied the circuit here temporarily
<http://www.danpercival.com/images/atxpwr1.gif>
What's that part circled in red for? A latch? Timing circuit? Also what's that bit shown by the arrow? Is it a switch? A connector? Why isn't it shown connected?

Sorry for the impolite way I've wriiten all this. I'm just curious about whether I really need to use the more complicated circuit - I don't have Mr Imai's email address to pose this question to him directly. Thanks in advance for an help - Dan

[vmarks]

Actually transistor and two resistors to pull line low mean to convert PS_ON# to PFW work to make the iMac shutdown, restart, and power on as normal.

In my circuit, I used a NPN transistor, a 1k resistor and a 2.2k.

Works perfectly.

Why is any of this necessary?

Well, to be perfectly honest, it isn't- you just lose the ability to shutdown. Without these circuits, either mine or the one you hosted, what happens is that the iMac will restart instead of shutting down.

The reason for this is that when you shut the iMac down, if the rate of decay is too quick, it perceives this as a power failure, and reboots. PFW- power failure warning. The transistor and resistors smooth out that decay so that it shuts down normally.

[dantheman]

I thought the ATX PS_ON had to be pulled to a TTL high for a soft shutdown. And a transistor between PS_ON and ground can only pull it to ground. (My layman understanding;-) Unless floating the PS_ON also has the effect of shutting the PS down. Well I guess I have to start building the project to find out.. starting with the one transistor circuit. Thanks

[dantheman]

I had an electronics engineer look at the circuit by T. Imai and his reply below might be of interest....

----
The part you arrowed that is after the inverter is a jumper which allows you to select the inverted signal or direct signal to go to the Power On sensing of the PSU. I suppose you set this according to what your PSU needs to turn on.

The circuit with the transistors I think is to prevent a spurious signal from turning on the PSU when the main switch is turned on. I think it holds the PFW low for a couple of milliseconds while the +5VTR comes up. Should be no harm building the whole circuit to turn on the system.
----