0

I'm using this KSZ8041MLLI device and this is its Hardware checklist

I'm using the MII Interface between this PHY and the MAC Controller (STM32 MCU) on the same board.

My question is whether we require length matching of the MII signals between these two devices? It is not mentioned in the datasheet? Also, it would be helpful for me if you can tell for which (x)MII interface signals do we require length matching?

  • Have you searched e.g. https://electronics.stackexchange.com/questions/392368/impedance-matching-resistors-on-mii-ethernet-lines https://electronics.stackexchange.com/questions/630429/mii-inpedance-matching-for-lan9313i https://electronics.stackexchange.com/questions/437982/what-is-the-maximum-recommended-routing-distance-between-mac-and-phy – Tim Williams Feb 16 '23 at 01:13
  • I did go through the link. But I am not clear on how it is helping? I am using a 33ohm series resistance on the MII lines between PHY and MAC. Is it enough? –  Feb 16 '23 at 01:17
  • The links say length matching is not required... – Tim Williams Feb 16 '23 at 01:21
  • Thanks @TimWilliams. So, why is required for some signals and not required for other signals? Can you tell me what draws a line between them? –  Feb 16 '23 at 01:25
  • What effect does length have on a signal? Delay. MII signals are parallel data with a clock; the data must arrive (stabilize) before the respective clock arrives at the receiver. Precise length matching is not required because the allowed mismatch is massive: almost a clock cycle (40ns, half for RMII). For my part, I would match them within a cm or so, just because. Which is likely trivial when the MCU and PHY are placed adjacent, which is preferred. – Tim Williams Feb 16 '23 at 01:31
  • So, if the MCU and PHY are placed near by length matching is not required. Any ball park distance you can suggest? –  Feb 16 '23 at 01:39
  • Within a few cm I suppose. Up to 10s of cm is still fine, at a somewhat increased risk of EMC issues. – Tim Williams Feb 16 '23 at 03:08

0 Answers0