Dell Powervault 5U84 Flextronics UD-PSU01-2200w PSU

[luckylinux]

Does anybody know the connector for this PSU ?

I bough 4x on a sale since they could be used with some PicoPSU or Similar .

I couldn't find many Pictures online, but it seems to have some Toot/Pin Connector, NOT a PCB with several Lanes for Control & Power.

Does anybody know the Specification of the Connector ? I guess the Small one is some kind of IDC Connector with what I seem to count as 24 Pins (3x8 Pins).

Flextronics UD-PSU01-2200w


Thank you

 

[luckylinux]

According to this Auction (NOT where I purchased it)

FLEXTRONICS UD-PSU01-2200-AC 0992524-10 240V DELL 06JN28 SWITCHING POWER SUPPLY | eBay

Find many great new & used options and get the best deals for FLEXTRONICS UD-PSU01-2200-AC 0992524-10 240V DELL 06JN28 SWITCHING POWER SUPPLY at the best online prices at eBay! Free shipping for many products!

www.ebay.com

They Mention DELL 06JN28, which might also be DELL 6JN28 since sometimes the "0" in Front on the Part Number are suppressed.

It might be related to Dell PowerVault

This Indeed

Dell EMC 6JN28 PowerVault ME4084 2200W 200-240V AC Power Supply for sale online | eBay

Find many great new & used options and get the best deals for Dell EMC 6JN28 PowerVault ME4084 2200W 200-240V AC Power Supply at the best online prices at eBay! Free shipping for many products!

www.ebay.com

Mentions Dell 6JN28 PowerVault ME4084

Does anybody know the Pinout and Type of Connector ? I couldn't find anything on Aliexpress. Seems to be WAY of a Niche Product. Especially if I want to use it for something other than its original Purpose .

I'd like to use it as a Standard 12V PSU for some PicoPSU or Similar.

Thank you

 

[luckylinux]

I made some Progress but not quite there yet.

There is an Interesting Thread in another Forum about another PSU 10+ Years ago, but still good Procedure to follow I guess

RC Groups - View Single Post - A simple high quality 12Volt 100Amp Power Supply- Part1

RC Groups - the most active Radio Control model community: electric and fuel rc airplanes,rc helis,rc boats and rc cars. Features discussion forums, blogs, videos and classifieds.

www.rcgroups.com

I now managed to clear the Fault LEDs (AC Fault Amber LED, DC Fault Amber LED).

However the PSU is stuck in "Standby" Mode (Green Status LED Blinking) .

This does NOT make any sense, because PS_KILL (a Feature of Redundant Power Supplies) should be quite easy to identify (it's the shortest Control Pin, in my Case it's Pin 24) and connecting it to GND/Chassis managed to clear the DC Fault LED that would otherwise come up every 5 Seconds or so.

I could pull LOW 4 Pins to GND and the PSU doesn't show any Amber LEDs Faults. But still in Standby Mode .

 

[luckylinux]

Any Idea ? Or anybody having some kind of Schematic either of the PSU or the Dell PowerVault ME4084 where it's connected to ?

I think I could pull like 12 Pins LOW (GND) and still in Standby Mode (Green LED Blinking).

Strangely enough it seems that to enable clear the Faults (Amber LEDs) of the the PSU I need to bridge one of the 5V together with the nearby "0V" Pin (it's not a GND but probably a pull-down Circuit, since the Resistance between them is like 5kOhm ... 10 kOhm). Otherwise I will have 2 Amber LEDs (AC Fault & DC Fault) and no Green LED at all (neither stable nor blinking).

 

[luckylinux]

I'm starting to wonder if it's possible to get this working. Heck I even bought another 13 of them .

According to the LED Guide on Dell Website for Troubleshooting, there is ongoing Firmware Communication between the PSU and the Chassis and apparently that can trigger a Fault as well:


I can replicate the Firmware Communication Loss if I DISABLE the Active HIGH on some of the Pins ... Except that Power is Blinking Green, not Steady Green ...

I never managed to get a Steady Green LED so far .

 

[luckylinux]

It would be good if somebody owning a Dell Powervault could dump some Communication Signals, as I believe the Hardware-only Mod isn't sufficient for this "advanced" and very proprietary PSU .














The Last Picture shows the PCB of the Controller, since that's where the LEDs in the Front Panel are wired to.

There is a 4-Pin Connector (2.0mm ???) probably used for Debugging on the Left Side not populated.

There is another 8-Pin Connector (2.0mm ???) on the Right Side also not populated, not sure if that's for some LCD in some of their other Products.

 

[luckylinux]

The 4-Pin Connector was actually a JST 1.25MM.

I had some Cables for that for another Project.

It has 3.3V (Red), 0V (Black) and 3.3V Pull-Up (Green, Yellow) through some quite high Resistance Pullups and/or Termination [differential] Resistors (~10kOhm ... ~100kOhm).

It's probably some I2C or UART Debugger / Flasher Interface (I haven't connected a Scope there yet).

These are NOT connected to any external Interface (well, except the Black 0V Wire that is connected to the Chassis GND), so it's probably some internal-only Header and not an Interface that is normally exposed. It could have been, because the Controller is like a Daughter Card/PCB.

I was hoping for something a bit more straightforward of a Power Supply though .

EDIT 1: Added some Details about the Resistances.

 

[mideel]

I've managed to get the main output turned on, but I haven't tested its behavior yet under load.

To achieve this I connected pin C7 (the shortened one, last one on the bottom row) to ground through a 10 ohm resistor (lowish value I had on hand, not important) AND pin B2 (third pin on the second row) through a 1.8k resistor (the value doesn't seem to be important, I just picked a higher value not to load the 3.3V too much). With these the PSU turns on and I get 13.38V on the output, which is a bit too high for my liking for a 12V rail. I suspect that there is feedback/voltage sense somewhere on the 3x8 connector and having it open causes the PSU to output the maximum voltage it can.

 

[luckylinux]

I've managed to get the main output turned on, but I haven't tested its behavior yet under load.

To achieve this I connected pin C7 (the shortened one, last one on the bottom row) to ground through a 10 ohm resistor (lowish value I had on hand, not important) AND pin B2 (third pin on the second row) through a 1.8k resistor (the value doesn't seem to be important, I just picked a higher value not to load the 3.3V too much). With these the PSU turns on and I get 13.38V on the output, which is a bit too high for my liking for a 12V rail. I suspect that there is feedback/voltage sense somewhere on the 3x8 connector and having it open causes the PSU to output the maximum voltage it can.

Thanks a lot for the Reply and for sharing .

Other Priorities took over so the Task was kinda paused for a while.

Are you counting from 1 or from 0 with your Pin Numbers ? I counted Pin01 ... Pin24 for my Tests, left to right, then top to bottom, always starting on the left-most Pin:

01​	02​	03​	04​	05​	06​	07​	08​
09​	10​	11​	12​	13​	14​	15​	16​
17​	18​	19​	20​	21​	22​	23​	24​

I assume you mean Pin C8 for the "Short One", right (Pin24) ?

In my testing GND seems to be only present on Pin C7 (or Pin23, depending how you count). The other Pins that look like GND (0V when Powered ON) actually have quite some Resistance (~ 1kOhm ... 10kOhm) to GND / Chassis, whereas Pin C7 has < 1 Ohm to Chassis.

So basically I always had Pin24 and Pin23 short-circuited together (PS_KILL Signal, to trip PSU if for whatever Reason you pull out the PSU during Load, the shortest Pin will disconnect First).

About Pin B2 (which I guess would be Pin11 for me), I actually didn't think it was much of anything (based on the linked RC Thread), since the Resistance seems to be quite high.

BUT ... in my Testing I needed to Pull UP some of the Pins to actually clear the Faults (DC Fault & AC Fault).

So I'm a bit surprised that connecting Pin11 to GND (Pin24) alone solved the Issue in your Case, as I'm pretty sure I already tried to connecting almost all Pins to GND during one Test at least, and there was still a DC Fault & AC Fault Present, and I could only clear those with a pull-up.

Did you already give it some Thoughts about the Busbar/Connector ? I tried looking for some Similar Parts but couldn't find one that matched the one used here .

 

[mideel]

Your 01 pin would be A0 for me. So with your naming scheme it's pin 11 (through 1.8k to ground) and pin 24 (through 10 ohms to ground).

The PSU still thinks it's in standby mode for me though. I get the main out powered on, but the green LED doesn't turn solid, it will still blink. What pins did you pull up to clear the faults? And did you use 5V or some of the 3.3V pins?

 

[luckylinux]

Your 01 pin would be A0 for me. So with your naming scheme it's pin 11 (through 1.8k to ground) and pin 24 (through 10 ohms to ground).

The PSU still thinks it's in standby mode for me though. I get the main out powered on, but the green LED doesn't turn solid, it will still blink. What pins did you pull up to clear the faults? And did you use 5V or some of the 3.3V pins?

You have Output when it's in Standby ? Are you sure ? I measured between different Busbars (especially between the Long and the Short ones, as I was expecting one to be the Positive and one the Negative) on the Power Connector and it was always 0V for me when Status was Green Blinking (Standby) with no other Fault Present.

I was bridging Pin17 and Pin18 in order to clear the DC & AC Faults. I think I had a zero Ohm Link.

About voltage Tuning, it's probably a matter of Putting a Potentiometer between 2/3 Pins and try to lower the Voltage on the Sensing Pin.

As you probably have already figured out, most of these are pulled up to 3.3V, mostly through some 10k ... 100k Resistors.

Likewise in my Case I'm wondering, based on what you are saying, if my Pin17 to Pin18 Bridge actually forces the PSU to Zero Volt for some Reason.

Too tired Today, I'll give it a try Tomorrow again, thanks for the Tips . In the last Batch of Tests I think I had a 470 Resistors in Series with all Pins possibly except GND which would fail otherwise.

 

[luckylinux]

@mideel: I'm just thinking ...



Did you try to put some load on it ? The Key (in one of the Red Boxes is): other PSU is providing Power.

So maybe below a given Output Current on the 12V Rail (and possibly 5Vsb Rail as well), the PSU thinks that another PSU is providing Power, since it itself doesn't seem to.

 

[mideel]

I didn't have a bigger electronic load on hand (ZPB30A1), but the power supply doesn't complain about delivering 60W even while its in the blinking standby state.

 

[luckylinux]

I didn't have a bigger electronic load on hand (ZPB30A1), but the power supply doesn't complain about delivering 60W even while its in the blinking standby state.

View attachment 44308

View attachment 44309

I just tried to replicate it on my Side.

If I disconnect the 5V Bridges (Pin1-2, Pin9-10 and also Pin17-18) then the PSU goes into DC Fault & AC Fault Mode.

If then I bridge:
- Pin24 (PS_KILL) & Pin23 (GND): either directly (0 Ohm) or through 470 Ohm Resistor
- Pin11 & Pin23 through a 470 Ohm Resistor

The PSU stays in AC Fault & DC Fault.

It's possible that I damaged something during my initial Testing as I didn't have any Resistors initially (couldn't find any and wanted to get started - bad Idea). So maybe I damaged something (or blew a Fuse), that's not impossible, but the 5V is still present so it must be a Resistor/Fuse on the other Pins if anything broke.

I can try with another PSU to exclude that maybe ....

EDIT 1: tried with a PSU I received a few Days ago and same Issue. PSU AC Fault & DC Fault. Bridge between Pin23&Pin24 was 470 Ohm or 0 Ohm, Bridge between Pin 11 and Pin23 was 2x470 Ohm in Series (approx. 1kOhm).

EDIT 2: @mideel can you please post a Picture of the PSU Naming Plate ? Just to make sure it's the same Model.

At this Point I don't know to be honest, I was also surprised I had to apply 5V to some Pins in order to clear the Faults, but at best it would still be in standby State, but without putting out any Voltage at all ...

 

[mideel]

EDIT 2: @mideel can you please post a Picture of the PSU Naming Plate ? Just to make sure it's the same Model.

Maybe the exact value of the resistances is important? Or where the current flows. I'm taking Pin23 and using it to provide the path to ground for both Pin24 and Pin11. Pin 24 to Pin23 is 10.5 ohms and Pin11 to Pin23 is 1.8k ohms. Here's the labels on the two PSU's that I've gotten to work:

 

[luckylinux]

Maybe the exact value of the resistances is important? Or where the current flows. I'm taking Pin23 and using it to provide the path to ground for both Pin24 and Pin11. Pin 24 to Pin23 is 10.5 ohms and Pin11 to Pin23 is 1.8k ohms.

I'm starting to think the same ...

Pin23-Pin24: 0 Ohm doesn't seem to work, 470 Ohm seems too much

and/or

Pin23-Pin11: neither 470 Ohm nor 940 Ohm worked ... I might try 4 x 470 Ohm in Series to get as close to 1.8 kOhm as possible, not sure where I have other Resistors.

But I repeat, in my Case, the ONLY way to clear the DC Fault & AC Fault and get the Green Status LED (blinking: Standby, NOT permanently ON) was to bridge 5V to the Adjacent Pin (so IIRC Pin1-Pin2, Pin9-Pin10, Pin17-Pin18), which is very weird.

What do you mean "where the current flows" though ?

Did you do some test at NO LOAD whatsoever (not even the multimeter etc when you turn the PSU on) ?

Did you hear any "clicking" sound as soon as you plug the PSU to the AC Power ? I do, most likely it's just the Relays for the Fault Indicators.

In Terms of Grid, I am off-grid, so it really shouldn't be getting a huge inrush Current when I plug it in (the lights flicker a bit, but that's the same also with e.g. Table Saw etc), definitively not enough to Trip by Overcurrent. Unless it's tripping due to AC Undervoltage during the Inrush Event (although that would be a bad Design of the startup Circuit). I don't have a Inrush Limiter on Hand so best I could do to rule that out is to put a ~10 Ohm Resistors in Series on Phase and Neutral. Of course at the expense of voltage Drop during Operation, but mine seems to be unable to start at all .

I read a bit the Manual and got the Impression that some Faults require Power Cycling in order to clear them.

Needless to say, I already tried unplugging & replugging the PSU several Times, but without any Success .

Here's the labels on the two PSU's that I've gotten to work:

View attachment 44313

Sorry, I meant from the Front, where you have both the Serial Number and the Model Number ....

From the Looks of it and the similar Date Code, I'd say we got it from the same Seller in the Netherlands at an Auction though .

If that's the Case, that should also rule out them being broken. Also because the likelihood that I received 2 broken ones from 2 different Orders ... well ... but be Lottery Day .

 

[mideel]

What do you mean "where the current flows" though ?

Did you do some test at NO LOAD whatsoever (not even the multimeter etc when you turn the PSU on) ?

Did you hear any "clicking" sound as soon as you plug the PSU to the AC Power ? I do, most likely it's just the Relays for the Fault Indicators.

There could be a difference between using the case as the ground, the main 12V output ground as the ground or Pin23 as the ground. Kind of like how with a GFCI system all the grounds (neutral and safety ground) are connected together at some point, the GFCI trips because the current is returning through the wrong path. Having the current flowing out of Pin24 and into somewhere else than Pin23 might cause some kind of fault state.

At no load the +12V just turns on and overshoots due to the lack of feedback (my best guess). Loading it down with 4.5A doesn't the voltage.

The power supply clicks when the AC power cord is plugged in and you get the 3.3V outputs activated. These pins have voltages:

-- -- -- -- A4 A5 A6 A7
-- -- B2 B3 B4 B5 B6 B7
-- -- C2 C3 C4 C5 -- C7

And when I load each of them down with the 1.8k resistor I get the following:

A4=0.082V
A5=0.54V
A6=0.082V
A7=0.047V

B2=0.54V
B3=0.54V
B4=0.082V
B5=0.54V
B6=0.082V
B7=0.047V

C2=0.54V
C3=0V (not sure about this one, I have way too many notes).
C4=0.5V
C5=0.5V
C7=2.1V

The higher the voltage the more current the pin is capable of providing. This is how I narrowed the pins down. C7 (Pin24) being physically shorter than the rest had to be the "Am I plugged into the server?" pin. I reasoned that the PS_ON had to be one of the 0.54V pins, so I tested each of them with different value resistors until I hit B2 and the PSU turned on.

I also did the table of resistances for each pin to ground:

A0=4.8k A1=11.4k A2=3k A3=600 A4=99k A5=12k A6=99k A7=O/L

B0=4.8k B1=8.9k B2=11.9k B3=3.6k B4=99k B5=11.9k B6=98k B7=O/L

C0=4.8k C1=10k C2=11.9k C3=O/L C4=15M (value lowers slowly when measured) C5=15M (value lowers slowly when measured) C6=0.02 (GND) C7=2.9k

Next I'm going to be looking at the C4 and C5 pins, the suspiciously high resistance kind of looks like it's going to an op-amp, which could mean voltage feedback.

If that doesn't work I'll have to look for the I2C bus with an oscilloscope.

 

[luckylinux]

There could be a difference between using the case as the ground, the main 12V output ground as the ground or Pin23 as the ground. Kind of like how with a GFCI system all the grounds (neutral and safety ground) are connected together at some point, the GFCI trips because the current is returning through the wrong path. Having the current flowing out of Pin24 and into somewhere else than Pin23 might cause some kind of fault state.

Except the current is flowing from Pin24 to Pin23 ...

Based on my Resistance Measurements, ONLY Pin23 is GND, so I assume for both 5Vsb and 12V.

I did NOT check Pin23 to Main_POS and Main_NEG though.

At no load the +12V just turns on and overshoots due to the lack of feedback (my best guess).

Alright, so there is NOT a minimum Load required, thanks.

Loading it down with 4.5A doesn't the voltage.

Well, for 183A Nominal Current, 4.5A is Peanuts .

The power supply clicks when the AC power cord is plugged in and you get the 3.3V outputs activated.

I don't think there are 3.3V Outputs, those are at least NOT advertised on the Nameplate. It's only 12V and 5Vsb.

I think that the 3.3V are more for Input/Outputs, rather than supply.

I agree that there are lots of 3.3V, BUT they all have high Internal Resistances to GND.

To estimate the Output Resistance (and NOT the Parallel one) we would have to put a variable Load on it.

Or measure the Resistance between Pins, NOT between the Pin and GND, as that would be the Parallel Resistance instead.

These pins have voltages:
-- -- -- -- A4 A5 A6 A7
-- -- B2 B3 B4 B5 B6 B7
-- -- C2 C3 C4 C5 -- C7

I assume you are only listening the 3.3V, not the 5Vsb.

C3 (Pin20) is 0.00 V in my Case.

And when I load each of them down with the 1.8k resistor I get the following:

A4=0.082V
A5=0.54V
A6=0.082V
A7=0.047V

B2=0.54V
B3=0.54V
B4=0.082V
B5=0.54V
B6=0.082V
B7=0.047V

C2=0.54V
C3=0V (not sure about this one, I have way too many notes).
C4=0.5V
C5=0.5V
C7=2.1V

What do you mean with "load each of them down" ? Do you mean you place the Resistor between the Pin listed and GND ?

For Reference here is my Resistance Measurement to GND:

2,00E+03​	1,14E+04​	3,12E+03​	6,00E+02​	9,75E+04​	1,09E+04​	9,78E+04​	2,00E+08​
3,84E+03​	5,70E+03​	1,10E+04​	2,68E+03​	9,82E+04​	1,09E+04​	9,65E+04​	2,00E+08​
3,63E+03​	1,00E+04​	1,10E+04​	6,00E+06​	3,75E+07​	1,80E+07​	0,00E+00​	1,93E+03​

The higher the voltage the more current the pin is capable of providing. This is how I narrowed the pins down. C7 (Pin24) being physically shorter than the rest had to be the "Am I plugged into the server?" pin. I reasoned that the PS_ON had to be one of the 0.54V pins, so I tested each of them with different value resistors until I hit B2 and the PSU turned on.

PS_KILL was the easy one being shorten . And the Pin nearby had nearly 0 Ohm to Chassis .

PS_ON on the other Hand, to be honest, with my 470 Ohm Resistor I tried to connect each and every one of them to GND. It made no difference at all.



So in Theory you could still double the Voltage (so your 1.8kOhm Resistance could still be doubled), so roughly 3.6 kOhm should be the absolute maximum.

Voltage Divider:

3.3V * (Rload) / (Rout + Rload) = 0.54V with Rload = 1.8 kOhm

3.3 * Rload = 0.54*Rout + 0.54*Rload

Rout = (3.3-0.54)/0.54 * Rload = 5.11 * 1.8 kOhm = 9.2kOhm


So I still don't see why it would NOT work with a short Circuit, that should obviously be current limited by the series 9.2 kOhm.

Or 470 Ohm or 2x470 Ohm in Series for that Matter ...

I also did the table of resistances for each pin to ground:

A0=4.8k A1=11.4k A2=3k A3=600 A4=99k A5=12k A6=99k A7=O/L

B0=4.8k B1=8.9k B2=11.9k B3=3.6k B4=99k B5=11.9k B6=98k B7=O/L

C0=4.8k C1=10k C2=11.9k C3=O/L C4=15M (value lowers slowly when measured) C5=15M (value lowers slowly when measured) C6=0.02 (GND) C7=2.9k

Quite a big Difference on A0 (Red) and some differences in Orange. See my Measurement Table above.

Next I'm going to be looking at the C4 and C5 pins, the suspiciously high resistance kind of looks like it's going to an op-amp, which could mean voltage feedback.

If that doesn't work I'll have to look for the I2C bus with an oscilloscope.

So you think the Header I mentioned in one of my Posts is for I2C ? I was kinda hoping UART to be honest ...


To be honest I don't know where it's going wrong for me. It could be a crappy Breadboard, some bad Connection on some of the Pins (are these Jumper Wires reliable at all on that 24 Pins Signal Headers ???), but I definitively don't get any Output ever .

EDIT 1: and of course the PSU shows Amber LED on AC Fault & DC Fault, in my Case. And the Green Status LED is permanently OFF.

The ONLY way to get rid of those Amber Lights, weirdly enough, was to bridge the 5Vsb to the adjacent Pins ...

 

[mideel]

C3 (Pin20) is 0.00 V in my Case.

C3 is probably 0V when the PSU is first plugged in. I had same weird measurements on some of the pins, so the Pin3 is probably an error on my part. On one of my notes I had:

C4 and C5=3.5V (jumping around, goes up to 15V momentarily, double checked with a second multimeter).

You'd think that there should be no voltage higher that goes to 15V on the DC side, which is why I did double check with a second multimeter.

What do you mean with "load each of them down" ? Do you mean you place the Resistor between the Pin listed and GND ?

Yup, provide each pin a path to ground through a resistor, then measure the voltage across the resistor to get an idea how much current the pin is providing.

Quite a big Difference on A0 (Red) and some differences in Orange. See my Measurement Table above.

The discrepancy is probably due to multimeter providing voltage to a capacitor (or the power supply having been plugged in), which changes the measurement. Measuring again I get:

A0=2.2k (rapidly climbs to 2.5k when measured)
B0=2.2k (rapidly climbs to 2.5k when measured)
C0=2.5k

These provide the +5V power, which measure 0.5 ohms when measured against each other (any combination of A0/B0/C0 measures 0.5 ohms). So both of our resistance measurements are incorrect. Initially the value should be around 2k and they should all give the same value when measuring resistance to ground.

As for the other discrepancies, here's my double check. Both units have sat powered off for hours:

B1=12k
B3=11.8k (on date code:1912) 10.9k (on date code:1910)
C7=2.96k (date code:1912) 2k (date code:1910)

 

[luckylinux]

C3 is probably 0V when the PSU is first plugged in. I had same weird measurements on some of the pins, so the Pin3 is probably an error on my part.

Fair enough .

C4 and C5=3.5V (jumping around, goes up to 15V momentarily, double checked with a second multimeter).

You'd think that there should be no voltage higher that goes to 15V on the DC side, which is why I did double check with a second multimeter.

I never noticed that. Heck I didn't see anything above 5V to be honest ...

When you say "goes up to 15V momentarily", how long are you talking about ?

Must be still relatively long if you could measure that with a Multimeter (> 1s ?) and without needing an Oscilloscope.

Although technically speaking:
- If there are floating Supplies in that PSU, it could get as high as Output 1 + Output 2 + Output 3 + ... (supplies in Series)
- Switching Transients / Resonances, especially at no load & low damping

Yup, provide each pin a path to ground through a resistor, then measure the voltage across the resistor to get an idea how much current the pin is providing.

The discrepancy is probably due to multimeter providing voltage to a capacitor (or the power supply having been plugged in), which changes the measurement. Measuring again I get:

A0=2.2k (rapidly climbs to 2.5k when measured)
B0=2.2k (rapidly climbs to 2.5k when measured)
C0=2.5k

It's indeed possible.

These provide the +5V power, which measure 0.5 ohms when measured against each other (any combination of A0/B0/C0 measures 0.5 ohms). So both of our resistance measurements are incorrect. Initially the value should be around 2k and they should all give the same value when measuring resistance to ground.

Don't quote me on the exact Value, but yeah, the 3 Pins are tied together (common 5V).

As for the other discrepancies, here's my double check. Both units have sat powered off for hours:

B1=12k
B3=11.8k (on date code:1912) 10.9k (on date code:1910)
C7=2.96k (date code:1912) 2k (date code:1910)

Interesting that they have different Resistors based on the PSU Date Code . I guess they changed something in their Design.

@mideel: can you please try Pin23 to Pin24 with either:
- 0 Ohm (short-Circuit)
- 470 Ohm

And if possible try 470 Ohm between Pin11 and Pin23 as well.

I'd just like to narrow down the Issue why mine won't start at all.

I'll try 4x470 Ohm in Series between Pin11 and Pin23 to try to match your 1.8kOhm.

Based on the Resistances and PS_ON / PS_KILL Thresholds, it SHOULD start as long as the Voltage on the Pin is < 1.0 VDC.

That being said, this NOT being a "standard" ATX PSU, it might have different Thresholds too ...