You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Thoughts pop up after having created an STM324F-based board with a LAN8720A PHY. I have selected this PHY because its compatible part 8742A is being used in a NUCLEO-F429ZI board. As such I wouldn't risk that ethernet won't work.
In general, I would prefer to feel free to make a choice out of different PHYs based on the knowledge that different PHY's share standardized registers and a bunch of vendor specific registers. Without a PHY driver, it makes extra sense to carefully strap PHY pins to my needs to avoid the need to initialize (vendor specific) registers.
If anyone could confirm that I do not need to hesitate about using PHYs that are untested in boards supported by Zephyr (when I practice careful strapping rules), then I'll consider other compact RMII-only-devices like KSZ8081 and DP83825i.
This topic is also related to the fact that drivers (which should open the door to using extra features) are mostly in modules/hal and hence with availability dependent on used SoC vendor.
--
The reason for my thoughts: 1) answers maybe helpful for other designers as well, and 2) I just encounter two nullseries boards with PHY problems: blinking LED where no link is active. When I measure an eye-diagram, it looks not so nice like sine wave as compared to another board equipped with KSZ8051 - even on TX pins. I fear that this is related to the 8720/42 requirement of secondary transformer taps connected via resistors to 3.3V instead of floating while DC-level is controlled by the PHY itself. I'd consider this scheme also more robust (while also saving four termination resistors).
reacted with thumbs up emoji reacted with thumbs down emoji reacted with laugh emoji reacted with hooray emoji reacted with confused emoji reacted with heart emoji reacted with rocket emoji reacted with eyes emoji
-
Thoughts pop up after having created an STM324F-based board with a LAN8720A PHY. I have selected this PHY because its compatible part 8742A is being used in a NUCLEO-F429ZI board. As such I wouldn't risk that ethernet won't work.
In general, I would prefer to feel free to make a choice out of different PHYs based on the knowledge that different PHY's share standardized registers and a bunch of vendor specific registers. Without a PHY driver, it makes extra sense to carefully strap PHY pins to my needs to avoid the need to initialize (vendor specific) registers.
If anyone could confirm that I do not need to hesitate about using PHYs that are untested in boards supported by Zephyr (when I practice careful strapping rules), then I'll consider other compact RMII-only-devices like KSZ8081 and DP83825i.
This topic is also related to the fact that drivers (which should open the door to using extra features) are mostly in modules/hal and hence with availability dependent on used SoC vendor.
--
The reason for my thoughts: 1) answers maybe helpful for other designers as well, and 2) I just encounter two nullseries boards with PHY problems: blinking LED where no link is active. When I measure an eye-diagram, it looks not so nice like sine wave as compared to another board equipped with KSZ8051 - even on TX pins. I fear that this is related to the 8720/42 requirement of secondary transformer taps connected via resistors to 3.3V instead of floating while DC-level is controlled by the PHY itself. I'd consider this scheme also more robust (while also saving four termination resistors).
Beta Was this translation helpful? Give feedback.
All reactions