I just verified that the SRW224P is using an end span (802.3af) type of power sourcing. I am having hard time checking for the details for the POES5. Tried checking that on the Linksys knowledge base and different search engines, I figured out the info for the WAPPOE12 but I the POES5 is still in question.
I don't know if that supports end span or mid span.
End span for your reference requires using the pairs 1-2 and 3-6 on the Ethernet cable carry power to the device. If a mid-span implementation is being used, then power is sent along pairs 4-5 and 7-8. The SRW224P is using the pairs 1-2 and 3-6 for the data and power supply transmission and the rest of the spare pins are just idle. I never got info yet for the POES5.
Anyone who has an idea is welcome and kindly give me the link for the website.
Considering the date of the message, maybe it is a bit late to reply, but it may help other users reading this. The POES5 splitter is an end-span device. I know it because I had to directly measure it with a voltmeter to troubleshoot a connectivity problem and found that it gets power via the data pairs (1-2, 3-6), with the remaining wires being idle.
By the way, this PoE splitter does not work as expected when combined with the SRW224P PoE switch. When the powered device is properly earthed (via the negative ring of the POES5 power connector), sometimes the switch does not properly complete the initial PoE detection protocol and fails to detect the attached device, thus preventing us, for instance, from remotely rebooting a device by disabling and re-enabling it via the switch web management interface. You need to physically unplug the device and then plug it again in order for it to be detected and powered again.
Oddly enough, this problem does not happen when the powered device is not earthed. This odd behavior led us to start measuring voltages and we found that the 48V carried through the data pairs are a floating voltage, i.e., not bound to earth, and there is an approximately -1,5V offset at the 5V DC end of the splitter, that seems to confuse either the switch or the splitter (we do not know for sure whom to blame about it).
We hope we will be able to find a solution with the help of linksys or other users. Out system needs to work unattended at a very distant site, so remote reboot is an indispensable requirement for us, because unplugging and re-connecting the powered device whenever a reboot is required is not an option at all.