This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

TPS23754: TPS23754PWPR - Looking for the Two event hardware detection fail rate in the PD

Part Number: TPS23754

As we are using this TPS23754PWPR in the POE PD side. As we used, both of the Network switch and AP (designed with TPS23754PWPR)  are supports PoE+ (Type 2 at).  But, in some of the times PoE (type 1 at(af) ) only detected on the both of the Network switch and AP. 

So that, I want to know the any fail rate (non detection of PoE+ (two event hardware detection) in PD side/PSE Side.

If any tested fail rate there, please update. otherwise looking for a theoritical fail rate for this device

  • Hi Mariselvam,

    If the same PSE and PD is intermittently operating in Type 1 or Type 2 power in the same conditions, then it may be that the class current is marginal. Note a PSE will demote a PSE to Class 0 (13W, type 1) if it doesn't process a class current dictated by the standard.

  • Thanks for the Reply, Darwin.

    As I have already said, we used both the devices (PSE and PD) with POE+ Support.

    As we have done 30 cycle test by using the these devices for POE+ detection.

    3 out 30 iteration got detected in the PoE only. Remaining all the 27 iteration detected in the PoE+ mode. 
    So, just want to isolate the issue on the PSE side or PD side like PSE not able to detect(two event hardware Detection) the current hosted by the PD  or PD not able to detect the classification voltage from the PSE?

    As I gone through some forum, got the below information

    "Dynamic power management is the better approach. It enables real-time measurement and management of power on each port. This eliminates the need for fixed power allocations, since only the necessary power is delivered to each port. The system monitors the actual per-port power consumption, including what is consumed by Class 0 PDs (which would otherwise always be delivered full power)"

    Is it any feature there in the PoE?

  • Hi Mariselvam,

    Dynamic power management or power port management is implemented after negotiation so I don't think this explains the intermittentcy.

    Can you forward me your schematic of the PD?
  • Hi Darwin Fernandez,

    I have got the reason for this inconsistency.

    As I have verified, PD can able to detect all the two event hardware detection requested by PSE. 

    But, If we unplug and plug the cable very quickly (less than a second) in PSE side, it does not give the two event hardware detection signal on the PoE line, So, by default PD can indicate the Power as af mode (T2P pin active low indicates the at detection). This is the reason for the issue.

    Could you please clarify the below two points for understanding about this?

    1. Why PSE can not initiates the two event hardware detection if connect the cable back within very short period of time (less than a sec)? - Attached the image

    2. Because of the first point, PD cannot able to identify the class detection and this will cause any issue on PD side?

    3. I can see all the times it requires minimum of two times of Two event hardware detection required on the PoE line to classify the Class, Why its happening  and this is because of PSE or PD? - Attached the image

    Image - PSE not generating the two event hardware detection

    Image - Always required minimum 2 number of two event hardware detection happening

  • Hi Mariselvam,

    1. If the PSE sees a class current not within the Class 1,2,3,4 current range from the standard, then the PSE will default the PD to a Class 0 (13W). You can measure the current waveform as well during the class event to see what currents are present at that time.
    2. The TPS23754 should utilize the T2P pin for the load to know how much power was allocated to the PD. the load can then adjust the power whether its 13W or 25W. This is used so that an occasion like this one, where the PSE only allocates 13W, the PD won't try to output 25W because then the PSE will crash (overload).
    3. Majority of PSE controllers can be operated in different modes. This looks like semi-automode where a PSE will continuously do detection class but not turn ON. These PSEs like to know what's connected to it first and what power level so it can do power port management before turning ON and allocating a specific power level. In this case, the timing will depend on how the PSE is implementing the power management. This is NOT abnormal. Some designs I've seen do this cycle more than 2 times before powering up.