Dell RPS-720 Cable Pinout
If you own a Dell Redundant Power Supply unit (e.g. RPS-600 or RPS-720), hopefully you have the original 4 cables it came with because they are very difficult to find and you can’t even order them from Dell.
There are not many details about these cords online which adds to the mystery of them. Interestingly, they are not as proprietary as you might think, they are very close to ATX 14-pin power cables.
The cable is straight through on both sides with 0 Ohm resistance on any of the pins. The plug is actually a 14-pin Molex Micro-Fit 3.0 series but I can’t find one with the exact keying that Dell uses. (The chamfered pins aren’t identified by Molex as keying, so they might fit into the mating plug without any issue)
Some have suggested using the ~$5 Dell 83181 20″ internal ATX power cable for their PowerEdge server as it is a 14-pin molex cable from Dell. I have not tried this yet, but I did order one and it does fit the plug.
The stock DELL cable for the RPS-720 is the G192F
The stock DELL cable for the RPS-600 is 3080417900 (UPC 646437461863)
Pretty sure the RPS-600, MPS-600, RPS-720 and MPS-720 all use the same cable. (RPS are for non-PoE Switches, MPS are for PoE switches.) The PoE cables might have thicker wires, but based on the RPS-600 cable, it feels overspec-ed for its 180w rating.
You can find used cables on ebay for around $30 each. I ended up getting the RPS-600 cable and have been using it on the RPS-720 with no issues. Try googling all the search terms to find the best deal as the cable seem all be the same.
Dell Switches
According to Dell, this is the pairing of power units to switches. Unless you are running near the upper limit, you can probably use either for any Dell switch.
RPS-600 (4 x 150 watt = 12v * 12A)
- PowerConnect 3524 – 24 port 10/100 switch
- PowerConnect 3548 – 48 port 10/100 switch
- PowerConnect 5224
- PowerConnect 5424
- PowerConnect 5448
- PowerConnect 5524 – 24 port 10/100/1000 switch
- PowerConnect 5548 – 48 port 10/100/1000 switch
- PowerConnect 6224
- PowerConnect 6248
- PowerConnect 7024
- PowerConnect 7048
RPS-720 (4 x 180 watt = 12v * 15A)
- N1500 Series
- N1524 – 24 port 10/100/1000 + 4 SFP+ 10G
- N2000 Series – Campus Managed Switch Series
- N2024 – 24 port 10/100/1000 + 2 SFP+ 10G
- N2048 – 48 port 10/100/1000 + 2 SFP+ 10G
Xcode to Swift Versions
Retrofitting USB port on older APC BackUPS with RJ50
I'm a 35 year old UIUC Computer Engineer building mobile apps, websites and hardware integrations with an interest in 3D printing, biotechnology and Arduinos.
I think I found the exact connector, I my RPS hasn’t been delivered yet tho:
https://www.molex.com/molex/products/part-detail/crimp_housings/0039012145
The connector is actually a Molex Mini-Fit Jr (same thing linked in the earlier comment), and your drawing of the key pins is a little bit off which made interpreting it a little difficult – it’s drawn looking into the pin receptacle side of the cable, with pin 1 at the bottom left corner. An interesting feature of these connectors is that the key pins are a repeating pattern with a cycle of 4 – any Mini-Fit Jr connector will mate as long as you line up pin 1 with pin 1, 5,9, etc. You can totally trim down an ATX connector to 14 pins and it’ll fit – although you might need to remove the clip too.
Your pinout appears to be correct, at least for some of it. The pins in your diagram are officially numbered like this:
8 9 10 11 12 13 14
1 2 3 4 5 6 7
(sorry about formatting)
So we end up with the following:
1: GND
2: NC
3: 12V
4: 12V
5: 12V
6: 12V
7: GND
8: GND
9: NC
10: GND?
11: 3V3?
12: NC
13: 3V3?
14: GND
Testing my rescued Powerconnect 6224 with a multimeter indicates that pins 1, 7, 8, and 14 are connected internally, as are all of the 12V pins. The other pins aren’t connected directly to any others, including the GND at pin 10. It’s likely that this and the two 3V3 pins make some form of power status signal.
The switch runs with 12V from my bench supply if I leave pin 10 (signal ground?) and the 3V3 pins disconnected, I just get red light indicating a power fault. I could check that supplying 3V3 makes the fault go away, but then I’d have to dig out a different PSU… 😀