Dual PHYTER DP83849IFVS Media Converter Mode

Michael Wagner72

Hello,

I have some questions about the Dual PHYTER DP83849IFVS.

We tried to use the DP83849IFVS as a Media Converter by putting the Pin EXTENDER_EN on high level (2.2k pull up to 3.3V). We also did some configurations with the Pins AN_EN, FX_EN and so on. Like its described in the Application Note / Datasheet.

My question is: Is it necessary to do some configuration of the registers to let the PHY work as a media converter? We have a MSP430 LaunchPad where we can read/write the registers of the PHY with the software USB2MDIO. Or is it enough to do the right configuration with high/low levels at the input pins of the PHY?

In addition some information about our application:

- We want to use the PHY as a media converter to convert the ethernet signals from IEE standard “TX” (electrical) to IEE standard “FX” (optical)

- On PORT A (TX), a RJ45 is connected

- On PORT B (FX), an optical transceiver is connected

Thanks for your help.

Michael

over 8 years ago

0 Ross Pimentel over 8 years ago

TI__Mastermind 23675 points

Hi Michael,

No register access is required for Media converter operation. All you will need to do is configure one to FX operation via strap and then enable the Extender strap. The other one can be left as default since default is copper.

Kind regards,
Ross

0 Michael Wagner72 over 8 years ago in reply to Ross Pimentel

Prodigy 120 points

Hi Ross,

thanks for your answer. I will attach a picture of our schematic. The resistors have a value of 2.2k. We tried to configure Port B as FX operation (Pin 59) and enable the Extender Mode (Pin 56), by putting it over a pull up to 3.3V. Port A was left by default, so TX (copper).

When putting the board to supply voltage, the voltage at EXT_EN (Pin 56) is 0 (GND). Also the voltage at Port B: FX_EN (Pin 59) is 0 (GND).

If we give a reset signal to the PHY, the voltage at both, EXT_EN (Pin 56) and FX_EN (Pin 59) is 3.3V. So the layout is correct and the pull up resistors work correctly, but the PHY pulls down the voltage to GND if we stop the reset.

Maybe we did something wrong with the leds? Or do you have any other idea where we could have made a mistake? If it's helpful we can also send you the whole schematic.

0 Ross Pimentel over 8 years ago in reply to Michael Wagner72

TI__Mastermind 23675 points

Hi Michael,

So even with a reset, the PHY does not strap in the correct setting? What are the register readings for registers 0x00 to 0x1F after power-up and after reset?

Kind regards,
Ross

0 Michael Wagner72 over 8 years ago in reply to Ross Pimentel

Prodigy 120 points

Hi Ross,

Even after reset or after power on, the PHY has no high level (3.3V) at Pin 56 (EXT_EN). Only if reset is active, we can measure 3.3V at Pin 56, so pull up works then.

An overview of the register readings is shown below.

After power-up:

PHY ID 00:

Register 0000 is: 3100
Register 0001 is: 7849
Register 0002 is: 2000
Register 0003 is: 5CA2
Register 0004 is: 0141
Register 0005 is: 0000
Register 0006 is: 0004
Register 0007 is: 2001
Register 0008 is: 0000
Register 0009 is: 0000
Register 000A is: 0000
Register 000B is: 0000
Register 000C is: 0000
Register 000D is: 0000
Register 000E is: 0000
Register 000F is: 0000
Register 0010 is: 0000
Register 0011 is: 0000
Register 0012 is: 0000
Register 0013 is: 0000
Register 0014 is: 0000
Register 0015 is: 0000
Register 0016 is: 0100
Register 0017 is: 04C1
Register 0018 is: 0000
Register 0019 is: 8020
Register 001A is: 0804
Register 001B is: 0000
Register 001C is: 0000
Register 001D is: 6011
Register 001E is: 003F
Register 001F is: 0000

PHY ID 01:

Register 0000 is: 2100
Register 0001 is: 7849
Register 0002 is: 2000
Register 0003 is: 5CA2
Register 0004 is: 0101
Register 0005 is: 0000
Register 0006 is: 0004
Register 0007 is: 2001
Register 0008 is: 0000
Register 0009 is: 0000
Register 000A is: 0000
Register 000B is: 0000
Register 000C is: 0000
Register 000D is: 0000
Register 000E is: 0000
Register 000F is: 0000
Register 0010 is: 0004
Register 0011 is: 0000
Register 0012 is: 0000
Register 0013 is: 0000
Register 0014 is: 0000
Register 0015 is: 0000
Register 0016 is: 014B
Register 0017 is: 04C1
Register 0018 is: 0000
Register 0019 is: 0021
Register 001A is: 0804
Register 001B is: 0000
Register 001C is: 0000
Register 001D is: 6011
Register 001E is: 089E
Register 001F is: 0000

After reset:

PHY ID 00:

Register 0000 is: 3100
Register 0001 is: 7849
Register 0002 is: 2000
Register 0003 is: 5CA2
Register 0004 is: 0141
Register 0005 is: 0000
Register 0006 is: 0004
Register 0007 is: 2001
Register 0008 is: 0000
Register 0009 is: 0000
Register 000A is: 0000
Register 000B is: 0000
Register 000C is: 0000
Register 000D is: 0000
Register 000E is: 0000
Register 000F is: 0000
Register 0010 is: 4000
Register 0011 is: 0000
Register 0012 is: 0000
Register 0013 is: 0000
Register 0014 is: 0000
Register 0015 is: 0000
Register 0016 is: 0100
Register 0017 is: 04C1
Register 0018 is: 0000
Register 0019 is: 8020
Register 001A is: 0804
Register 001B is: 0000
Register 001C is: 0000
Register 001D is: 6011
Register 001E is: 003F
Register 001F is: 0000

PHY ID 01:

Do you have any idea where the mistake could be? Thank you very much for your help.

Kind regards,

Michael

0 Rob Rodrigues over 8 years ago in reply to Michael Wagner72

TI__Mastermind 19780 points

Hi Michael,

Is the DP83849 not working in extender mode? You shouldn't see a high level on EXT_EN after the power up because the EXT_EN pin is shared with RX_D2. After power up the pin role changes from EXT_EN strap to RX_D2 function. While in RESET it will be EXT_EN strap.

If you are in extender mode, the MAC interface isn't likely to be active but your device should be in extender mode.

You could test this with a PC connected to a router easily. The extender will be totally transparent to the PC and router.

Best Regards,

0 Michael Wagner72 over 8 years ago in reply to Rob Rodrigues

Prodigy 120 points

Hi Rob Rodrigues,

thanks for your answer. I think we see our trouble now. The PHY is connected to a transceiver (Firecomms FB1M2LPR, data sheet attached). Five pins of the DP83849 are connected to the transceiver (RD+-, TX+- and (FX)SD).

Pin FXSD is even pin AN1_B (42). The transceiver causes a voltage of about 1.4V at the PHY pin 42, so AN1_B is neither 0 nor 1. That's why the DP83849 doesn't work in extender mode. If we disconnect the transceiver's SD pin, the PHY works in extender mode.

The transceiver is connected like it's described in the data sheet (see attached file), so it should work with the DP83849. I have no idea why the transceiver generates a voltage of 1.4V for the pin 42 (FXSD and AN1_B) of the DP83849.

FB1M2LPR_Data_Sheet.pdf

Do you have any idea how to fix this problem? Can you recommend any transceiver type the DP83849 works fine with?

Thank you very much for your support!

Best Regards,

Michael

0 Rob Rodrigues over 8 years ago in reply to Michael Wagner72

TI__Mastermind 19780 points

Hi Michael,

I am pleased to hear that you have found the problem.

I believe you are running into an issue interfacing the single-ended PECL signal from the fiber transceiver with the LVCMOS buffer of the DP83849. This could be fixed as easily as including a buffer between the DP83849 and the SFP.

Best Regards,

0 Michael Wagner72 over 8 years ago in reply to Rob Rodrigues

Prodigy 120 points

Hi Rob Rodrigues,

thanks for your answer. You mentioned that this issue could be fixed easily as including a buffer between the DP83849 and the Transceiver. I will attach a principle schematic of how the PHY and the transceiver are connected right now. Could you please explain what you mean by "including a buffer" ? Do you mean putting a capacitor between both? A little picture of your idea would be very helpful.

Thank you very much for your help.

Best Regards,

Michael

0 Ross Pimentel over 8 years ago in reply to Michael Wagner72

TI__Mastermind 23675 points

Hi Michael,

The SD pin for the transceiver you are using shows the need for a 22k pull down, However in the schematic you show a pull up on this pin.

Can you show me how you are connecting the SD pin from the fiber transceiver to the PHY SD pin?

Regards,

Ross

0 Ross Pimentel over 8 years ago in reply to Ross Pimentel

TI__Mastermind 23675 points

Hi Michael,

Can you check with your fiber transceiver vendor what interface type the SD pin is? It appears to be CMOS, which means you can remove that 22k resistance and just keep the pull up you have. This should eliminate the voltage issue.

Regards,
Ross

0 Michael Wagner72 over 8 years ago in reply to Ross Pimentel

Prodigy 120 points

Hi Ross,

thanks for your answer and sorry for the late reply. Here is a picture that shows how the PHY SD pin is connected to the SD pin of the fiber transceiver.

The datasheet says that the transceiver has the following features:

• Integrated CMOS driver IC for RCLED

• High sensitivity CMOS receiver IC and PIN diode

I removed the 22k resistor but the voltage issue is still there (1.2V at Pin 42). Do you have any other idea how to fix that problem?

Thank you very much for your help.

Kind regards,

Michael

0 Ross Pimentel over 8 years ago in reply to Michael Wagner72

TI__Mastermind 23675 points

Hi Michael,

Please also remove the LED (H106) and current limiting resistor (R146).

Also remove the parallel pull-up resistor (R143) since AN1 is not used for FX operation.

Please stuff the 22k pull-down that is specified by the fiber transceiver.

The only components that should be on the line are thus the SD fiber transceiver pin, 22k pull-down resistor and the 0 ohm series resistor.

Kind regards,

Ross

0 Michael Wagner72 over 7 years ago in reply to Ross Pimentel

Prodigy 120 points

Hi Ross,

first of all, sorry for that very very late reply. I am working as a student on this project and the last months I had to write my final exam. But now I'm back and want to complete this topic.

The DP83849 seems to work fine, the clock is okay, the different levels are correct and some diagnostic leds are shining. I just don't know why there is no output at the TX+- pins of the PHY (PORT B: pin 35/36).

There is a similar circuit (that works) and on this there is a output signal on these pins (data signal with around 60 MHz). I have no idea why the PHY of the other circuit does not output a signal.

Does anyone have a idea how to fix this problem?

Best Regards,

Michael

0 Michael Wagner72 over 7 years ago in reply to Ross Pimentel

Prodigy 120 points

Once again, with some Schematic attached:

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

Hi Ross,

first of all, sorry for that very very late reply. I am working as a student on this project and the last months I had to write my final exam. But now I'm back and want to complete this topic.

There is a similar circuit (that works) and on this there is a output signal on these pins (data signal with around 60 MHz). I have no idea why the PHY of the other circuit does not output a signal.

Does anyone have a idea how to fix this problem?

Best Regards,

Michael

Schematic_IPE.pdf

Interface

Interface forum

Dual PHYTER DP83849IFVS Media Converter Mode