Other Parts Discussed in Thread: DP83869
hello
I am using your DP83869HM chip media converter. It works fine in almost all situations, but there are times when it malfunctions.
First of all, the strap settings are as follows.
PIN 22 : OPEN ( See Understanding different modes of operation in DP83869.)
pin 32 : mode 0
pin 35 : mode 1
pin 36 : mode 0
pin 38 : mode 0
pin 45~47 : mode 1
1. I want to know the operation principle of 1000M media converter mode.
A. After the optical signal is input and LED1 is lit, Cupper link is up and in operation, if the SD is removed, Cupper link is also disconnected, is this normal?
Shouldn't Cupper link be connected regardless of SD?
2. When applying the above strap in 1000M media converter mode, AUTONEGO seems to be activated, can it be set to 1G Full duplex by changing the strap or register value?
A. When the phy chip and Ethernet switch (autonego) are connected and link up, there is a phenomenon that intermittently sticks to 10M_HD (half dulpex) when the cable of the optical transceiver is removed and reconnected. This phenomenon disappeared after fixing the setting of the Ethernet switch to 1G_FD. This is expected to be a 'Duplex miss match' phenomenon.
It happens when both sides are set to auto nego and the SD signal comes in late.
According to the datasheet, when applying this strap, the settings are as follows.
- Fiber Force mode
- Copper : Auto Negotiation ( 1000 Advertised), Auto MDIX
Can you explain the above 2 modes?
In particular, Auto Negotiation ( 1000 Advertised), Auto MDIX What is the difference between these modes and 1G_FD?
In the datasheet,
1000base-t <-> 1000base-x only
100base-t <-> 100base-x only
It says above, but it seems like it is actually not using AUTONEGO.
Please let me know which strap pin or REGISTER I can change the AUTONEGO setting.
3. Does pin 24 receive LOS or SD signaling?
A. I am using Coretek's optical transceiver (1.25Gbps /1310nm). This optical transceiver outputs SD signal as HIGH (3.3VDC) when it receives optical signal. When I first designed the DP83869HM, the chip did not work when I connected the SD pin of the optical transceiver and pin 24 of the chip directly, so I connected the SD signal inverted and it worked. I couldn't find this in the datasheet, so if I missed something, I would be very grateful if you could let me know.
4. Using the MCU, I designed to keep the DP83869HM in RESET state when the SD signal is LOW through the SD(HIGH) signal of the optical transceiver.
A. In the process, there is a phenomenon that the phy chip does not work in a situation where the receiving sensitivity of the optical transceiver is -16dBm. Furthermore, the optical transceiver does not work at all, could there be any setting on the SD pin? The problem is obviously my fault for not following the reference design, so there's nothing wrong with the TI chip, but I can't figure out what's going on between the PHY chip and the optical transceiver when the optical signal is attenuated.
As always, thank you for your kind response.
