Optiplex 7000 SFF proprietary 8 pin pinout to use ATX supply

the 7000 8 pin out is completely different from old Dell 8 pin mentioned in the other thread.  nice pic you posted for the new Dell 8 pin

Dell proprietary version of ATX12VO old discussion

"RED, GRAY, and BROWN: These stay at 3.3v high regardless of whether the power supply is on or off; based on what I saw with a scope there's not even a brief dip on them when the front panel power button is pushed.  these don't change, and are TTL level."

Re: there are two 3.3v returns (brown and grey) from the motherboard, but the Dell PSU does not need them to operate  

The 3.3V on the red wire is not the same function as the ATX power good signal, as it is always on when the AC is connected.

Gray used to be PWR_OK: a signal provided by psu to indicate to the motherboard that all of the voltages are okay.   Dell in the past has always used grey in its proprietary connector (old 8 pin and old 6 pin) for PWR_OK.

@redxps630​ With the wires cut it could be seen that the 3.3V on the red wire was from the PSU, and the 3.3v on the grey and brown were coming from the motherboard.
I did not observe any flickering of the voltage indicating a data signal, but I was only using a DMM. It remains a possibility

Re:

Replacing the Dell PSU with a standard ATX was accomplished and worked almost normally. A small 5v DC to 12v DC convertor was used to get 5v from the ATX standby rail to 12v on the purple wires. This was done through a small diode so that the DC-DC convertor will not struggle against the 12v output of the ATX PSU when it is on. The PS_ON wires were connected directly, as were the grounds and the other two 12v wires.

The only oddity is that the fan of the ATX PSU does not switch off when then PC shuts down or goes to sleep, it keeps spinning. 

the second oddity is known in older Dell 24-6 pin adapter.  it is expected

are you able to test whether you can get 12v directly from 24 pin ATX without diode adapter?

@redxps630​ Yes, the supply starts normally when the PS_ON line is grounded. The 12V from the DC-DC was removed and the PC kept running, the 12V was coming from the ATX PSU for sure.
Like this (no 12V from the DC-DC) when the PC is shut down, the PC releases PS_ON line and the ATX PSU shuts off as well.

If the DC-DC is left connected the ATX PSU looks like it stays on. (it is hard to tell, since if you open the connection between the DC-DC and the purple wires the ATX PSU shuts off. The only clue is the fan). Something is latching it on, I suspect the +12V from the DC-DC convertor.

I have thought about it some more and I believe one solution could be to use the PS_ON line being released on shutdown (or the 3.3V from the motherboard) as a trigger to momentarily open the connection between the +12V from the DC-DC and the purple wires. A couple of seconds is likely all that is needed for the ATX PSU to power down. Then the +12V from the DC-DC can be reconnected to the purple wires. A MOSFET and a 555 perhaps. I don't think sleep mode will be possible like this, as the "12V standby" will get interrupted, that's ok hibernate is better

https://edc.intel.com/content/www/xl/es/design/products-and-solutions/processors-and-chipsets/alder-lake-s/atx12vo-12v-only-desktop-power-supply-design-guide/2.0/dc-connectors/

in the standard ATX12VO

I_​PSU% is a signal coming from the power supply that reports the proportionality of Power being delivered by the +12VDC rail with the Output-Load rating of the PSU. Which is represented as a unitless percentage of the total capacity using a current mode. If multiple +12VDC rails are implemented (for example, +12V1DC, +12V2DC) then I_​PSU% must report the utilization ratio of the combined total capacity.

I_​PSU% Signal Characteristics

Remote sensing is optional. Remote sensing can accurately control motherboard loads by adding it to the PSU connector. This is for the power supply to monitor the 12V Voltage to the motherboard connector, through remote sensing, and then compensate the voltage if there is excessive cable Voltage drop. The default sense should be connected to pin 10 of the main power connector. The power supply should draw no more than 10 mA through the remote sense line to keep DC offset voltages to a minimum.

so the guess now is

one of the 3.3V from PSU is a signal to motherboard reporting I_​PSU%.  

Re: To determine the function of the red, brown and grey wires, theses were each cut and the system operation tested. None had any effect and the system ran normally without them. The PC could be turned on and off from the power button, the standby/sleep function worked normally.

- there is a 3.3v output (red wire), but it does not seem to have an effect 

3.3V on the red wire is always on when the AC is connected.

if the red is psu output to motherboard, it may be the I_​PSU%

- there are two 3.3v returns (gray and brown) from the motherboard, but the Dell PSU does not need them to operate

these may serve as remote sensing pins.

@redxps630​ certainly a possibility that there is a signal there, though I did not observe anything but 3.3V on the red wire, plus this was present even when the Dell PSU was off. (a 3.3v SB rail, if you like). The other two 3.3V lines (grey and brown wires) get their voltage from the motherboard, not the PSU.

The Dell PSU did not implement a 12V standby rail, which surprised me. Seems to be combined with the normal 12V, which makes sense I suppose

From what I can tell the two purple 12V wires are common inside the Dell PSU, though I have not opened it to confirm. There is no resistance between them at least.

Re: The other two 3.3V lines (grey and brown wires) get their voltage from the motherboard, not the PSU.

that is very good observation.

Re: 3.3V on the red wire was present even when the Dell PSU was off

seems as soon as you connect AC power to psu, even when it is totally disconnected from motherboard, you get a 3.3V on red.   this may be a I_​PSU% pin.

The I_​PSU% signal is routed to the Psys pin on the CPU’s IMVP controller on the motherboard.

The system input power monitor, PSYS is designed to provide information necessary for the processor to dynamically allocate power between different system components, such as the CPU, memory, and the rest of the platform, within a designated power budget. This helps in optimal use of total available platform power by facilitating the dynamic employment of surplus system power to assist the platform turbo. 

@redxps630​ Next opportunity to have it on the bench I will connect the oscilloscope to the 3.3V from the Dell PSU and see if there is anything else present.

Indeed all the 3.3V and 5V lines for example SATA drives all come from the motherboard VRMs, it is only 12v input to the motherboard

3.5" and 2.5" SATA device do not require or need 3.3v power, but mSATA ssd still needs 3.3 v power.

ultimate question is can user make a 24 to 8 pin adapter easily if they do not want to use proprietary Dell oem psu.  your experiment has made large progress in the understanding of Dell engineering.

at this point, a crude adapter seems to only need 5 essential pins out of 8 pins to turn on psu and power on motherboard, that the red, grey, brown pins of motherboard 8 pin socket do not have to be connected, although system may not operate in optimal condition.

The PWR_OK signal lets your circuit know that the power coming from the PSU is OK and running within tolerance - not under or over voltage.

It does not need to be connected to anything if you don't need it, so you can just leave it disconnected.

Re: The other two 3.3V lines (grey and brown wires) get their voltage from the motherboard, not the PSU.

I would be surprised if Dell psu skips PWR_OK.  so if red is definitely from psu independent of motherboard, it may be the PWR_OK.

the gray and brown are from motherboard feed back to psu, which are optional based on OP experiment.  

Examined the inside of the Dell PSU and got a number of nice clues and general info from the PCB silkscreen.

- The EPS 12V connectors (brown wires) are referred to as +12VADC, and the 8pin 12V (purple wires) are referred to as +12VBDC. Each of these, and the grounds are common on wires of the same colour.
- The other wires are all on CN1 and have names: Red=ALERT, brown=I2C_CLK, grey=I2C_DAT, and blue=ON_OFF

So RedXPS you're right and the 3.3V wires are some data connection. I2C is a birectional multidrop bus so it would allow exchange of information between the motherboard and the PSU. Not sure what exactly, and for my application that's about as much as I care.

The red 'ALERT' wire probably serves a similar function PWR_GOOD. If it's not 3.3V then the PSU is not OK? However the PC seems to run fine with out it ¯\_(ツ)_/¯
For usage with an ATX PSU not much has changed. You could convert PWR_GOOD from 5V to 3.3V and feed it into ALERT, but so far what I have seen indicates this is not needed. Maybe there's a pull up on the motherboard and it defaults to OK if the wire is open.

As for my PSU running all the time, this is indeed due to the 12VSB from the DC-DC. I'll need to come up with another plan, maybe some of those P-channel MOSFET diode replacement things to isolate the 12VSB from the +12V, without a large voltage drop. Will report on what I come up with

this is very good revealing and breakthrough in understanding red, brown, grey, blue

OP has confirmed

red: alert

blue: ON_OFF

brown: 12C-CLK

grey: 12C_DAT

the purple 12VA and 12VB refer to two separate 12VC rails.