• State Resolved
  • Locked Locked
  • Replies 6 replies
  • Answers 1 answer
  • Subscribers 33 subscribers
  • Views 530 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

DP83822IF: Strap configuration, fiber interface, power consumption

Oba
Genius 17635 points
Part Number: DP83822IF


Hello,

My customer has some questions for DP83822IF.

 

Q1)

On page 5 of the datasheet(SNLS505E), there is the below note, (1)-(f)

Type: S-PU, S-PD - Strapping Pin (All strap pins have weak internal pullups or pulldowns. If the default strap value is needed to be

changed then an external 2.2-kΩ resistor should be used)

 

This notes says just 2.2Kohm should be used.
But the strap resistors are specified in Table 9 on page 47 and it is not just 2.2Kohm and is more complicated.

So can the user just ignore note(1)-(f) on page 5?

 

Q2)
On page 6 of the datasheet, type of LED_1 / GPIO1 is defined as S-PD.
But on the table 11 on page 48, there is no “LED_1 / GPIO1” pin strap information.
What is this pin used to configure?

 

Q3)
LED_1/GPIO1 is act as “signal detect” in IF device.
Is this signal detect pin a single end logic input with VIH_min and VIL_max specified on the page 10?

 

Q4)
There is no information on the datasheet about TDP/TDM and RDP/RDM signal level.
Are these pins LVPECL?
There signals are coupled with capacitance and it looks like shifting the voltage.
But they would like to know how to judge whether these signals are within the specification.

 

Q5)
In the Figure 34 on page 93, these pins as pulled up to 3.3V through 49.9ohm.
Does this mean AVD must be 3.3V in fiber case?

 

Q6)
There are power consumption data in the table 145 on page 97-99.
They think these values are taken when the MDI driver is in Class-B (default)
Do you have the values when the MDI driver is in Class-A?

Regards,
Oba

  • 0
    TI__Mastermind 40545 points

    Hi Oba,

    Here are my answers to your customer's questions:

    1) This question was previously asked on E2E here. In short, yes, they can ignore the note(1)-(f).

    2) This question was also previously asked on E2E here. These straps are used for internal purposes.

    3) This question was also asked on E2E here. Because this is an IO pin, they would follow VIH_max and VIL_min on page 10.

    4) This pin is CML. We would like to recommend that they follow Figure 34, where they have the series capacitor as well as ensuring they have the terminations for the type of transceiver they are using on that side.

    5) This is a typo, AVD can also be 1.8V as well as 3.3V.

    6) They are correct in thinking these are Class-B data. I am following up with the design team to see if Class-A values were measured. We are also wondering what their potential use case is for Class-A.

    Thanks,

    Gerome

  • 0 in reply to Gerome Cacho
    TI__Genius 17635 points

    Hello Gerome,

    Thanks for your reply.

    >2) This question was also previously asked on E2E here. These straps are used for internal purposes.

    Does this mean the user doesn't need to take care of the status of this pin at power-up and reset?

    >4) This pin is CML. We would like to recommend that they follow Figure 34, where they have the series capacitor as well as ensuring they have the terminations for the type of transceiver they are using on that side.

    I understand this is CML. But don't you have any electrical specification for these pins? 
    Without specifications, they can't check whether the signals on the pins are really within the spec.

    Regards,
    Oba

  • 0 in reply to Oba
    TI__Mastermind 40545 points

    Hi Oba,

    Oba said:
    Does this mean the user doesn't need to take care of the status of this pin at power-up and reset?

    If the user is not using those pins, they may leave them floating. If they will be using them, they will need to strap them.

    Oba said:
    I understand this is CML. But don't you have any electrical specification for these pins? 
    Without specifications, they can't check whether the signals on the pins are really within the spec.

    The electrical specifications for these pins are dictated by the ethernet standard (in this case, 10BASE-Te and 100BASE-Tx).

    Thanks,

    Gerome

  • 0 in reply to Gerome Cacho
    TI__Genius 17635 points

    Hello Gerome,

    Sorry, I wrongly pushed "This resolved my issue".

    Anyway, thank you very much for your repy.

    Regarding LED_1/GPIO_1 pin strap question,  I can't still understand your answer.

    My understanding for this LED_1/GPIO_1 pin is that this is TI internal-use strap pin.

    This pin is LED or GPIO pin. So any external circuit could be connected to this pin.

    My concern is whether there is any possibility to enter the TI internal-use mode if this pin is connected to a certain external circuit.

     

    Regards,

    Oba

  • 0 in reply to Oba
    TI__Mastermind 40545 points

    Hello Oba,

    I understand your concern that external circuitry may affect the initial bootstrapping configurations. If the pin in its regular operation is being unused, the designer may leave that pin floating as the internal pull-up (or in this case, pull-down as specified in the datasheet) will set the default mode. Section 8.5.2 of the datasheet has more information on if a LED was hooked up to that pin. In that case, depending on the configuration of the LED, you would need to strap it low or high.

    Sincerely,

    Gerome

  • 0 in reply to Gerome Cacho
    TI__Mastermind 40545 points

    Gerome Cacho said:
    6) They are correct in thinking these are Class-B data. I am following up with the design team to see if Class-A values were measured. We are also wondering what their potential use case is for Class-A.

    Hello,

    The design team was not able to find measured values, but stated that in Class A mode, the 100Base-TX mode power consumption is estimated go up by around 20mA typically.

    Sincerely,

    Gerome