1

I've been looking for a good, definitive reference example for MIPI C-PHY signal routing and have come up short. Bear in mind that C-PHY uses a set of 3 signals per lane, with multi-level coding between them. This is different than D-PHY which uses regular differential coding.

Generally I get that it should have 50 ohm single-ended / 100 diff, but is it as simple as having 3 lines run together side-by side (microstrip) with 100 ohm diff between each outer one (A,C) and the 'middle' one (B)? If so, can someone explain the theory of why this works, in detail?

Related question - if a system is laid out for D-PHY (microstrip diff pairs), how well can it support C-PHY, if it all?

hacktastical
  • 49,832
  • 2
  • 47
  • 138

1 Answers1

1

A couple of thoughts then.

In the case where the run is very short it probably doesn’t matter.

Longer run, each C-PHY trio member needs to be its own 50 ohm transmission line with little or no coupling to its neighbors, even within the trio. (They can however be treated as a common-mode set for filtering. More here: https://www.murata.com/en-us/products/emc/emifil/library/pickup/mipic-phy-immunity)

So the only practical way to support both D-PHY and C-PHY in the same layout is to route the signals separately, as single-ended. In fact C-PHY requires it; D-PHY doesn’t care so much.

What won’t work is trying to run C-PHY (single-ended) over a D-PHY (differential) layout. The trio members will interfere with each other too much.

hacktastical
  • 49,832
  • 2
  • 47
  • 138