Sharing Ethernet Magnetics

[Pseudobyte]

I have an application where I have two external 100BaseT ports.

Can I use a ethernet magnetics designed for 1000BaseT and use two channels for each port respectively?

Any issues doing something like this?

 I have little doubt it will work, my concerns are largely with EMC/EMI issues.

[nctnico]

I don't see why it would be a problem. I think I did something similar in one of my own designs. EMC issues with ethernet a more likely to stem from crappy phys which have too much harmonics on the ethernet signal lines. Make sure you are able to insert series resistors or ferrite beads (0402 size) in the ethernet lines connected to the phy. At first you can use 0 Ohm resistors but if EMC testing fails, these resistors may save your from needing a PCB respin.

[temperance]

I don't see why it would degenerate EMI performance. Each diff pair has it's own transformer and CM inductors.

Sources which degenerate EMI performance are for example asymmetric termination, power and ground noise,...

Make sure you are able to insert series resistors or ferrite beads (0402 size) in the ethernet lines connected to the phy.

I second that. Those ferrite beads can be a source of nasty resonances when combined with MLCC capacitors. A few ohms can safe your life at the test facility. (and make you look like a a genius.)

Or if unsure about the ferrite beads you can test the filter circuit with small current sink able to source the same current as the Phy in terms of di/dt.

Some like to sprinkle magic ferrite beads everywhere. When not implemented properly then for example when an EFT burst test is performed on the IO cables, the resonances caused by those resonating filters will disturb the power network and make you fail those tests.

[Pseudobyte]

Thanks for the insight and the tip about resistors.

I actually failed my prescan on this design at 250MHz so I am doing everything i can to improve the design.

I am changing from 8 to 12 core magnetics which is why my original question came up.

Apparently ethernet is often a source of failure according to my test house. Well this is what prescans are for, and hopefully i learned my lesson.

[T3sl4co1l]

One concern, the dielectric breakdown between channels.  Gig doesn't need it, two 100's does.

They're generally rated for 1.5kV anyway, but, you may not have independent BS terminations, which would be a problem (galvanic connection between ports).  If they are independent, and they are rated for the voltage, then that should be fine.

You can also just get dual/quad sets, or smaller pairs so that two side-by-side takes up about the same space.  YMMV.

Tim

[wraper]

You don't need a 1000BaseT part though https://uk.farnell.com/epcos/b78476a1889a003/transformer-lan-dual-10-100-base/dp/1644266 as multiple port versions exist for either type. It's just how many replication of the same thing in the package.

[T3sl4co1l]

Make sure you are able to insert series resistors or ferrite beads (0402 size) in the ethernet lines connected to the phy.

I second that. Those ferrite beads can be a source of nasty resonances when combined with MLCC capacitors. A few ohms can safe your life at the test facility. (and make you look like a a genius.)

Keep in mind where this advice is coming from (I think).  nctnico has worked with, LAN- or KSZ-something PHYs from Microchip, which emitted significant wideband noise not rejected by magnetics.  Remember the magnetics are only rated to 100MHz or thereabouts, if that; and with fairly middling specs, at least for the minimum IEEE 802.3-whatever requirements they have to meet -- how much any given set exceeds the standard, they never tell, though.  You can't expect balance (CMRR) at such frequencies (100s MHz, GHz), or filtering.

(There will generally be a notch in the transformer's transfer function due to the TLT construction, but we can only know here that it's above some minimum, and in lieu of filtering, there will be plenty of pass bands above that cutoff/breakup frequency, making it likely that harmonics go right on through.)

FBs with MLCCs is another matter.  This probably refers to the AVCC bypasses around the PHY and magnetics.  Indeed, they should be placed judiciously, and enough loss present in suitable locations (large ceramic + explicit resistor, tantalum, electrolytic, etc.) to dampen the system.  I hope it goes without saying that large MLCCs (values that would be prone to resonate with FBs) do not belong on the signal lines, which was the earlier context.

Note that midding-value FBs with small Cs (100s pF say) are generally safe, the resonance being in the 1s to 10s MHz, and the Q being small (<2?).  Such values would a possible further improvement (as L section, reverse-L section, or pi network) to the mentioned problematic PHY outputs, with suitable values of course (cutoff above 100MHz; 10s Ω FB, 10s pF cap).

Tim

[Pseudobyte]

You don't need a 1000BaseT part though https://uk.farnell.com/epcos/b78476a1889a003/transformer-lan-dual-10-100-base/dp/1644266 as multiple port versions exist for either type. It's just how many replication of the same thing in the package.

I couldn't find magnetics with auto transformers that were not designed for 4 pairs.

[temperance]

I hope it goes without saying that large MLCCs (values that would be prone to resonate with FBs) do not belong on the signal lines, which was the earlier context.

I didn't read that carefully it seems. I was referring to ferrite beads in power lines.