Thanks for your comments. The problem with this solution is I now have a management problem as I will have over 100 camera's in various locations throughout a campus environment and I would have to know which ports each of these camera's was on to disable the POE and also assume they would not get moved around at times to other ports which is highly unlikely.

Is there any issue with potential port damage on the Cisco switch if these mid span devices are used and the POE on the Cisco switch is enabled?

Thanks!

Hello Curtis,

I remembered that in one campus we have some outdoor Access points that have their own power injectors

see

http://www.cisco.com/en/US/docs/wireless/access_point/1520/power/guide/1520pwrinj.html#wp49664

it provides some suggestions

in our design document about wireless says that switch port connected to the power injector has to be configured with POE disabled.

Tomorrow I will check if it is so, but in your scenario with 100 devices this can be almost unmanageable we have 30 outdoor devices with power injector.

I would think of putting them in dedicated vlans so that a normal PC connected to it cannot access the intranet

(I imagine cameras are deployed for video surveillance so I would consider to put them in a VRF)

>> Is there any issue with potential port damage on the Cisco switch if these mid span devices are used and the POE on the Cisco switch is enabled?

This is exactly what I'm afraid of.

The switch will try to detect if a POE device is on the port: if the power injector is well built it should not expose the resistance signature (25 Kohm) to the switch so POE should not be used on the switch port.

if the cameras also doesn't speak CDP or LLDP to ask power the switch under these two hyphotesis would be safe even if poe capable.

Cisco has started to introduce their own ip cameras

see

http://www.cisco.com/en/US/prod/collateral/vpndevc/ps6918/ps9692/ps9716/data_sheet_c78-492032.html

they don't mention a power injector.

on the slides of CIPT they show the midspan with a switch port non POE.

Sorry I cannot say yes you can do it.

I think in theory is possible to build a power injector that presents itself as a non PD device to the switch if has two ports and the correct electrical design.

searching with google I've found someone speaks of

PoE splitter (PD-AS-401/12).

see

http://www.electronicsnews.com.au/Article/PoE-splitter/170800.aspx

and for electronic details

http://www.powermanagementdesignline.com/howto/185300535;jsessionid=XVUWX4L5EYPROQSNDLPCKHSCJUNN2JVN?pgno=6

The unpowered CAT-5 cable plugs into the input (lower) jack (J1), and the injected 48 V power is placed on the unused (Alternative B) pair of wires that connect to the output (upper) jack (J2). This maintains the current midspan PSE convention, and prevents the 48 V from being connected to the unpowered switch or router. This solution does not include the detection and classification functions specified for a PSE by 802.3af, but that does not matter, since it is a temporary solution that only powers a single port.

however the author says it will not provide 48V to the switch port but nothing about what if some DC power comes from the switch to J1 (input).

This should at least protect from electrical damage the switch port but needs to be verified on your power injector data sheet.

Hope to help

Giuseppe

Hello Curtis,

I'm looking at an outdoor access point with a power injector

the POE is in auto mode but the operational state is off

SW-RM-SXR000-F-C4-1#sh power inline g3/24

Available:1400(w) Used:652(w) Remaining:748(w)

Interface Admin Oper Power(Watts) Device Class

From PS To Device

--------- ------ ---------- ---------- ---------- ------------------- -----

Gi3/24 auto off 0.0 0.0 LAP1510 n/a

Interface AdminPowerMax AdminConsumption

(Watts) (Watts)

---------- --------------- --------------------

Gi3/24 15.4 15.4

This confirms that the switch port can be left with auto configuration and doesn't need to have POE disabled

Hope to help

Giuseppe