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.
Hello
We are using DP83640 Ethernet PHY with a custom board, based on am335x soc with RGMII connection (PHY is in master mode - i.e. recevies input from 25Mhz crystal to produce 50Mhz output) and we experience issues.
What we have so far:
1. Internal loopback test between MAC and PHY works (i.e. no errors, link is ok, all packets are received back)
2. External loopback does not work at all (phy does not detect itself - i.e. does not link to itself)
3. When connected to host PC PHY, the link is established (from MDIOLINK register and leds on the phy), PHY is able to receive data from pc (based on CPSW_STATS registers), but host pc does not receive anything (again CPSW_STATS registers on the board show that the data is being sent).
We use Transformerless Ethernet connection similar to the one described in SNLA088A TI application report.
I would appreciate if you could provide any help on how to resolve this issue.
Best regards,
Denis
Hi Denis,
To comment on each scenario:
1. It looks like internal loopback is working. No issues here, correct?
2. What is the hardware configuration? The DP83640 cannot be configured for external loopback through register writes.
3. Let's solve item 2. before moving on to this item.
Thank you,
Nikhil
Hello Nikhil,
1. yes, everything works correctly without issues
2. there is no any specific configuration for external loopback, I just cut the ethernet cable and shortened TX+ with RX+ and TX- with RX- (this setup seems to work with other PHYS, like with the one on my PC, or with the one from Beaglebone Black, but I suppose they use trasformers unlike us, so probably I'm missing something here), the problem seems to be that PHY does not react to its own link pulses and thus can not even establish a link (i.e MDIOLINK register shows zeros, no link led)
Also, I mistyped in previous post about interface, it is not RGMII but RMII.
Forgot to add, we have several boards with same setup and the issue is present on all of them.
Thanks and Best regards
Denis
Hi Denis,
For item 2. You are correct, this cable setup should work. Can you share a schematic? If you have register access can you read back the following registers:
Thank you,
Nikhil
Hello, Nikhil
Here are the values of registers:
Reg 0x0: 0x1100
Reg 0x1: 0x78c9
Reg 0x10: 0x4000 (it says that auto-mdix swapped tx and rx, I'm wondering if it is normal for a loopback?)
Reg 0x17: 0x4021
Reg 0x18: 0x0
Reg 0x19: 0x8001
Reg 0x1c: 0x0
Regarding, schematics, we normally are not allowed to share them, so could you specify exactly what kind of schematics you would like to see and also if it is possible to share them in private?
Thanks in advance
Best regards
Denis
Hi Denis,
I have sent you a request to connect, can you share the PHY portion of the schematic via private message?
Thank you,
Nikhil
Hi Denis,
As we have moved this discussion offline, I will be closing this thread.
Thank you,
Nikhil