TPS23753: Design Review support

Hello, 

Have you used our Poe PD schematic and Layout review guidelines?

If this post answers your question, please indicate so by marking this thread as resolved. Thank you.

Regards, 

Michael P.

Applications Engineer

Texas Instruments 

Hi Michael Pahl,

Yes, we have used TI - TPS23753 PoE PD Controller, considered TI guidelines in our design.

But we have implemented few changes in our design, based on our testing.

We would like to get our PoE PD Controller section reviewed.

Hi Michael Pahl,

We are using current limiting resistors in the VDD and VDD1 pin of TPS23753.

To decide the resistor values, please share the current consumption for the pin VDD and VDD1

Hello, 

Can I ask why you would use current limiting resistors on these pins? 

The power in the design is limited in a number of ways. First is the internal pass MOSFET has a current limit that will shut off if it is too high. The other current limit is the Current Sense (CS) pin that measures peak current through the primary winding of the transformer. 

Finally, the PSE should also limit power consumption and will shut off power if more than 15W is consumed. 

Additionally to this, if you want to implement a finer tuned power limiting circuit, we typically limit it on the CS pin using the guidelines from the following paper: Designing PoE switching supplies for input power sharing and with power limiting

For the VDD current, I really don't think you want to be trying to limit this one. This supplies the power the PoE PD circuitry, like detection and class. These functions depend on voltage and current accuracy, so interfering with that is not recommended. Additionally, these functions will turn on and off, depending on where in the PoE handshaking process we are. 

Instead, when I think about power consumption, and the two ones to watch are power across the internal pass MOSFET and the power needed for the DCDC converter (VCC*ICC + power needed to drive the FETs). 

If this post answers your question, please indicate so by marking this thread as resolved. Thank you.

Regards, 

Michael P.

Applications Engineer

Texas Instruments 

Hi Michael,

Our Poe design got failed in the pre compliance surge test. When, Surge enters the Poe section obviously the diode across the Poe inputs react in order to protect the circuit, but it’s not in our case. Pd controller IC is getting short when the surge enters the section. We tried placing high impedance on the Vcc and Vcc1 pin during the surge, the Pd controller works fine. We suspect during the surge event, reaction time of tvs diode may be not quick enough to safe the controller, which is getting damaged in vcc and vcc1 pin. The current limiting resistor we are denoting here not for load current limit, it is to protect the Vcc and Vcc1 pin during the high surge event.

As you mentioned the vcc has lot of internal power requirements, we are able to understand that from the internal block diagram in the datasheet, but in order to place proper current limiting resistor values for pin vcc and vcc1 without any interruptions on handshaking and other normal operation, we required current consumption of vcc and vcc1 pin.

please provide the current consumption details for the same 

Hello Michael,

Please review the below MOSFET current consideration and calculated Rvdd1 and Rvdd resistor values based on that.

According to IEEE 802.3at specification the voltage range at PD side will be 37V to 57V. We have considered 37V as the minimum input voltage and 57V as the maximum input voltage at pin VDD and VDD1. We are using an MOSFET which has 21nC (gate charge max), below is the link to MOSFET datasheet - https://www.vishay.com/docs/71356/71356.pdf 

The switching frequency of the TPS23573 is 248KHz (typical), time period will be 4.03uS.

Considering 33% dutycycle (@37V for 12V output), the time period will be (ton - 1.329uS, toff - 2.7uS)

Igate = Qgate/switching time = 21nC/1.329uS = 15.8mA is the maximum inrush gate current.

VDD1 Current = 0.85mA + (Mosfet Gate current) = 0.85+15.8 = 16.65mA

Power = 12V * 16.65mA = 199.8mW (test condition, Vc = 12V)

Power (@37V), = 37V * input Current

VDD1 input current max = 199.8mW / 37V = 5.4mA

VDD1 input current min = 199.8mW / 57V = 3.5mA

RVdd = RVdd1

The resistor value Rvdd1 = 6.85Kohm (@37V, 5.4mA), Rvdd1 = 16.28Kohm (@57V, 3.5mA)

We are having two designs one is with class0 and other design is class4 compatible. Please confirm, Will the internal negotiation current ratings will be same as 0.85mA for both the ethernet classifications ?

If yes, we can use the resistor value between 6.85K and 16.28K for both the designs, right ?

Also please confirm on the Schematics and layout review requested for the PD controller section. I will share the design files in person, please update on the review process ASAP.  Share your mail ID to send the design files for review.

Or you can send me files through the messenger. 

Hi Michael,

The classification current values - 44mA for class4 PD, its with respect to 2.53V as per the internal block diagram. I believe its within the limit of resistor values selected. It should not have any impact as we have considered the maximum Vcc current (0.85mA @ 12V).

Thanks for your support in clarifying the query. Will definitely test the design with these resistor values, will get back to you for further support.

Regarding the design review, its better to share the files through local FAE to maintain confidentiality of our design.

Hi Michael,

For class 4 device, we are using TPS23754.

Thanks for reviewing the schematics and layout files. please share your comments on layout and placement, if any.