DP83867E: LED Operations

joe lindula

Part Number: DP83867E
Other Parts Discussed in Thread: 66AK2H14, DP83867CS

Hello, I am using the DP8367 PHY to interface with the 66AK2H14 processor. I have wired the PHY to the ARM+DSP processor through the SGMII interface. I have not programmed the DSP/ARM with any boot code. When I plug in an ethernet cable to the RJ45 connector I don't see any LEDs toggling on LED_0 (LINK established) or LED_1 (1000BASE-T) link is established. I am connecting my board to a laptop so I thought I would see some leds flickering. When I plug the ethernet cable into the 66AK2H14 EVM it's leds flicker. Should I expect some flickering? LED_2 is always high.

Some say I won't see any action until the port on the ARM+DSP processor is configured but I'm not an expert in that area and we are working to configure/initialize the port now.

I have probed the CLK_OUT pin and I do see the 25Mhz Reference clock. I have also probed all the boot strap pins.

I just thought by simply plugging in an ethernet cable that is connected to a laptop that I would see seem of the LED toggling or stay on.

Here is my schematic

ARM_DSP_MOD 37.pdf

If anyone can help me please send me your comments.

Thank you,

Joe

over 6 years ago

0 Ross Pimentel over 6 years ago

TI__Mastermind 23675 points

Hi Joe,

I am reviewing this right now.

0 Ross Pimentel over 6 years ago in reply to Ross Pimentel

TI__Mastermind 23675 points

Hi Joe,

A few things about your schematic:

1. RBIAS should be 11k ohms

2. Center taps of the RJ45 on the PHY side need to be isolated from each other. The one you have shorts all the center taps together.

3. There are parallel pull-down resistors on the LED pins. LED_2 circuit is fine because you are bootstrapping the pin low so the PHY will sample that pin for MODE1 bootstrap. However, LED_0 and LED_1 are pulled HIGH. You need to add a parallel pull-up so that the PHY goes into MODE4 bootstrap. LED_0 and LED_1 will not blink by default. They will remain solid ON when there is link and 1000Mbps speed. LED_2 is set to blink for RX/TX activity.

0 joe lindula over 6 years ago in reply to Ross Pimentel

Mastermind 6940 points

Ross,

Hello, thanks for catching my errors. On number 2, whould you suggest that I remove C275 and C276?

Thanks,

Joe

0 Cecilia Reyes over 6 years ago in reply to joe lindula

TI__Genius 9920 points

Hi Joe,

Which capacitors are you talking about exactly on the schematic? I don't see C275 or C276 near the center taps.

Thanks,

Cecilia

0 joe lindula over 6 years ago in reply to joe lindula

Mastermind 6940 points

Okay, so correct me if I'm wrong here:

LED_0 needs to me in MODE4? MIrror Enable and SGMII. Ooops I should have got that.

I have LED_0 floating.

LED_1 is is MODE4 and is pulled high, that is correct in the schematic, is that right?

Thanks,

Joe

0 joe lindula over 6 years ago in reply to joe lindula

Mastermind 6940 points

Oops, I should have said C770 and C771.

0 joe lindula over 6 years ago in reply to joe lindula

Mastermind 6940 points

When I did the design I missed Figure 26 in the DP83867CS datasheet.

0 Cecilia Reyes over 6 years ago in reply to joe lindula

TI__Genius 9920 points

I believe you are using the incorrect magnetics. For the PHY to work properly, you're going to need magnetics where the center taps do not all short together. They need to be isolated. Please refer to our datasheet and some of our TI Designs as reference.

Thanks,

Cecilia

0 joe lindula over 6 years ago in reply to Ross Pimentel

Mastermind 6940 points

Ross,

Hi, in the schematic R89, R84 and R91 are not installed and the LEDs are all pulled high. Is that correct?

Joe

0 joe lindula over 6 years ago in reply to joe lindula

Mastermind 6940 points

I believe my Rlo boot strapping resistors should be 10K and installed for MODE4 for all the boot strapping pins and I'm operating in SGMII mode. Does that sound correct?

Joe

0 Cecilia Reyes over 6 years ago in reply to joe lindula

TI__Genius 9920 points

Hi Joe,

Yes it is pulled high. Are you selecting mode 4 for the straps? If so, then idea Rhi should be 2.49k. Please take a look at our datasheet for the recommended resistor ratios.

Best,

Cecilia

0 joe lindula over 6 years ago in reply to Cecilia Reyes

Mastermind 6940 points

Cecilia,

Hello, we are operating using the SGMII interface and on page 36 under Table 5 it has a note.

For SGMII Mode 4 strap, TI recommends using Rhi = 4 kΩ and Rlo = 10 kΩ on RX_D0 and RX_D1 , RX_D2 and RX_D3.

Is this not correct? Am I missing something. I have to admit that I'm new to using this device.

Thanks,

Joe

0 joe lindula over 6 years ago in reply to joe lindula

Mastermind 6940 points

Hello, I left pin 44 floating, could that be a problem? I am seeing a 25-mhz clock on pin 18, not sure if I would see that if the device was in a power down state. Thank you very much, Joe

0 Cecilia Reyes over 6 years ago in reply to joe lindula

TI__Genius 9920 points

Hi Joe,

Leaving pin 44 floating is ok. Were you able to follow up on Ross' point on #3?

Thanks,

Cecilia

0 joe lindula over 6 years ago in reply to Cecilia Reyes

Mastermind 6940 points

We decided to replace the RJ45 connector and see if that fixes the problem. Will get back with you when we do. Thank you very much for all your suggestions.

Respectfully,

Joe

0 joe lindula over 6 years ago in reply to Cecilia Reyes

Mastermind 6940 points

Cecilia,

I also forgot to say that after closer inspection of my Boot Strapping resistors, I found that I was using the wrong values on the GPIOx and LEDx pins. As usual my mess.

Joe

Interface

Interface forum

DP83867E: LED Operations