DP83TD510E: BACK TO BACK

您好，已经收到了您的案例，调查需要些时间，感谢您的耐心等待。

您好，

请允许我们的团队在下周审阅原理图。 我将回复此主题。 感谢您的耐心。

Hi , 

I wanted some additional information about your setup.

What exactly is the purpose of this "isolation repeater" placed between the 2 PHYs? The recommendation for back-to-back operation, as listed in the datasheet, simply has the MII lines directly connected to each other, ensuring a impedance of 50 Ohms on each line. I am worried that this component may be changing the parameters of the MII interface, such as affecting the slew rate, total impedance, etc which may be the reason for communication failure. 

Do you have 4 total 510 devices in this setup? Are you trying to ping the ethernet device that is connected after the last 510?

Can you check the register 0x0010 for the link status in all these 510 devices? I want to make sure that there is an active link.

Another thing we could try would be to use a MII Loopback on the last 510 device to ensure that the data is going through all the different 510s correctly. Like you mentioned, there are waveforms on the MDI lines according to your testing, but these could just be Auto-negotiation pulses, or idle symbols. Doing this loopback would give us more confidence that the datapath is stable. 

Best,

Lydia

你好，谢谢你的回复。我无法在此处发送附件。我已将回复邮件发送到您的电子邮件。你能看看电子邮件信息吗

您可以尝试私信发送。

Sorry, I couldn't find the entrance channel for the private message. Could you please send me a private message? I will reply to you via private message once I receive it

Hi

I had sent you a private message.

Hi,

Thank you for clarifying customer's setup. It sounds like this is being made for intrinsically safe system. My worry here is that the relay device may be interfering with the slew rate/impedance of the MII lines, which may be a contributing factor to the ping not working. 

My understanding is that the DP83TD510EVM(L) and DP83TD510EVM(R) are connected through the DP83822 PHY on the EVM to PC1 and PC2, and that the customer is trying to ping PC1 from PC2 or visa versa. Before validating this complete path, I wanted to see if the data is going through the product customer designed as expected. For this test, we can use the MII Loopback functionality of the 510 to send any data coming from PC1 back to PC1. 

This setting should be enabled on the right side 510 device on customer board such that the data first goes from PC1 to DP83TD510EVM(L), then to the input end DP83TD510 of customer board, and then the output end of the customer board, at which point it is looped back and transmitted back to PC1. This setting can be enabled using register 0x0000 bit 14. 

Doing this, customer could see network traffic on PC1 using a program such as wireshark and verify if they are able to see the packets sent being looped back. This would serve as a good basis for us to conclude that the datapath between the 2 510s on customer board is good. It is also important to check that the data within the packets being transmitted and received match.

Thank you for your reply. This design is indeed a part of the intrinsic safety system. I understand the verification test you mentioned, but I don't quite understand how to implement it specifically. How should the loopback function be implemented? Can you help me come up with a testing plan? There is one more thing that I designed this product without an MCU system. How should I implement register control? We look forward to your reply

Hi,

xianghe Wu 说：

How should the loopback function be implemented?

This setting can be enabled using register 0x0000 bit 14.

xianghe Wu 说：

There is one more thing that I designed this product without an MCU system. How should I implement register control?

If the customer has any provision for the MDC/MDIO pins on a jumper, they can use the USB-2-MDIO software  and one of our MSP430 devices to access the register. This information is also given in the user guide for USB-2-MDIO

For the first test, I just wanted to see if we can have data looped back after enabling MII Loopback on the output terminal of customers design.