GS116LP - PoE Mode B

The PD is a custom board that I designed, it is for a specific application and did not need to adhere to PoE standard.  It was better for me to save the cost and space of only implementing one and B made sense because I am running 10/100 and thus do not need the other wire pair.  I just didn't want to use passive PoE in case someone plugged a laptop into it or something.  Unfortunately I did not realize that the great majority of switches are mode A only.  I can buy a mode B switch, but majority of them are much more expensive. 

You are correct on mode B.  Pin 4 is shorted to Pin 5 internal to RJ45 and both carry DC+.  Pin 7 is shorted to Pin 8 and carry DC-.  However, what I am seeing when I connect a cable from my switch to my device is that both pairs become shorted to eachother, which I would not expect.  Thus, Pins 9 and 10 on the electronics side are shorted going over to my diode bridge.  Suspect it was something to do with the fact that negotiation between my PHY and the switch hasn't occured yet, but still can't make sense of that since RJ45 PoE pinout for gigabit would have those pins as data separate data pairs. 

Ultimately I am ok since I luckily had the foresight to diode-or in power from the PDs main connector, but if there was an equivalent 16 port switch that did Mode B for around the same price it would be better.  GS116LP is unmanaged so I knew it was unlikely to be reconfigured to Mode B.  Looked around on its PCB hoping to find a jumper or something.  I do see some DNP'd UARTs......