Why are USB-C cables with ferrites rare?

[liteyear]

There was a time, if my memory serves me, when adding a ferrite to a power supply cable was a simple mitigation against device EMI. In this USB-C era, despite the cable frequently being used simply as a power supply cable for a radiating device, I don't see products coming with or requiring a ferrite-equipped cable.

Is this:

• A figment of my imagination?
• A technical requirement, as implied here
• Unnecessary because compliant USB-C cables have an impervious shield and compliant USB-C power supplies have significant common mode impedance or something??
• Aesthetically driven?
• A sign engineers have got better at mitigating EMI at the source?
• Because adding a common mode choke to the product itself is cheaper/safer/better/more desirable?

[wraper]

My Сanon scanner came with USB-C cable with ferrite beads. USB-C is mostly used and first appeared in portable devices, so ferrite bead on a cable is a major usability downgrade.

[xrunner]

I have some USB cables in my junque box that have ferrites embedded to the cable. These all were included with equipment I bought such as perhaps a camera or some other common consumer device. All the USB cables I buy now don't have the ferrite on them, but I haven't specifically searched for the ones that do. I just tried a search on Amazon for "ferrite USB cable" and yes you can buy them if you choose to. If you don't include "ferrite" in the search you most likely won't be served up that choice, as many consumers will probably do, simply because they don't know what "ferrite" even means. But if you are knowledgeable about RF interference and need a USB cable with a ferrite - you can get it.

[T3sl4co1l]

USB-C is mostly used and first appeared in portable devices, so ferrite bead on a cable is a major usability downgrade.

It would still do, to put it on the desktop/hub end, where impedance is likely lower / stray capacitance to ground/surroundings higher.

But ah, that works fine with ye olde USB-A to -B/mini/micro cable, the ordering is well-defined.  With USB-C, it's whatever.

So it kinda makes sense that the common denominator might be "none".

The safer choice would be to put one near both ends (the lighter-loaded end would simply not do much), but that would add more cost and bulk, and they're probably just the plain unadorned cords, cheap, for mobile-to-mobile, etc.

But that's neither here nor there.  I wouldn't say there's much insight here; it's not a high-information question.  The more interesting questions go in different directions.

For example, I wouldn't assume they've solved issues; the safer assumption is a perpetual race to the bottom.  If they can get away with it, they will.  Indeed, chargers tend to have poor CM noise and ground leakage (interference with capacitive touch is a common symptom, as is that rubbery-stiction touch feel when modest AC voltages exist on a surface).  Whether these would be helped with beads on cables, maybe, but mostly it doesn't matter as it's a small device, small system (the charger-cable-cellphone system, give or take someone's hand also holding it), and transiently used -- it might be giving off obnoxious emissions, but good luck locating and prosecuting a source that's only on a couple hours at a time, etc.  And that if licensed spectrum users even notice and report the interference.

Regarding CMCs, the on-board equivalent is an array of data chokes, one side wired in parallel (carrying VCC/GND), the others passing the data signals.  (Data chokes are almost exclusively two-winding transformers, so you need a lot of them wired in parallel to handle all the data signals a USB cable contains.) That way all signals remain coupled to a reference plane, through the chokes, but the reference potential on the cable can be somewhat isolated from the main board ground plane.  Clearly this isn't a step they're going to be embedding in dense electronics like a cellphone, and the better option is to simply ground the connector shield to the board -- extending the reference plane onto the cable itself, maintaining signal quality but also carrying CM noise along it.

Tim

[Jeroen3]

I still have one on my USB A to C cable for my keyboard, on the A side. I don't think it has a purpose other than to feel premium.

I remember ferrites mostly on VGA/DVI cables. I suppose these analog carriers could actually be susceptible to common mode noise.

Today I rarely see one, not on USB, HDMI or Displayport. So you're right, something must have changed.
Though sometimes a power brick has one, there I can understand it as afterthough when it failed emc tests.

[Berni]

Ferrites on USB cables are just there to get the product past EMI compliance tests. So if they can get away with not having one, then adding one in is just extra cost that gives no benefit. Tho the silly thing is that the user can just swap it out for a non ferrite cable.

They don't really have much to do with actual signal integrity on the cable. For that using common mode chokes on your USB lines is much more effective.

For something like VGA it can make sense to have a ferrite for signal integrity because the signals inside are not differential and are very sensitive to noise due to being analog. So if you can reduce some common mode crap going over the grounds you can get a actual signal integrity improvement.

[T3sl4co1l]

For something like VGA it can make sense to have a ferrite for signal integrity because the signals inside are not differential and are very sensitive to noise due to being analog. So if you can reduce some common mode crap going over the grounds you can get a actual signal integrity improvement.

I mean... it depends.  Usually the cable is triple coax inside braid, so the signal quality can be quite good indeed.  The coaxes may go through bead(s) on their way down to the video board -- more the case for classic CRTs (Trinitrons and such) I think, than for single-board digital panels (connector direct into PCB).  There's definitely the possibility or opportunity there, to treat it differentially, and reap the improvement in signal quality -- but also one which may be left out by many, and which is rapidly degraded by connectors (DE-15 does not have coax pins on it; they should've never let DB13W3 go ), or discarded entirely (i.e., a novice designer might simply see all the ground pins and think, I'll just tie all these to ground plane and not even think about CMCs).

Cable quality itself varies widely.  In the last years I had a CRT, I no longer had a VGA card at all, and had to use a DVI adapter; the stub length and impedance mismatch (or connector quality, or..) introduced a reflection, making a subtle after-image to the right of sharp vertical details (so, most noticable on window outlines for example).

It's all very much more precise and critical than a full-digital or coded scheme is, and for HDMI it's pretty much there or it's blank.  So it's no accident we've moved away from [lightly/un-coded] analog signaling in many places.

But anyway, the differential comment -- just to emphasize that better for those that don't know about the techniques -- you can, and absolutely should, treat coax differentially, where possible.

The key point is, you're after the signal between core and shield, so read that signal in such a way that CM currents do not affect it.  For example, tying shield to circuit ground through an impedance (there's inevitably some series impedance, shielding is rarely perfect), but terminating the signal to circuit ground, returns that signal current through the series impedance, and thus CM ingress (and emission) occurs.  You can terminate the signal to the shield first, so that it flows in a local loop free of CM interference.  Then tie shield to circuit ground, so that CM current is sunk separately.  The signal voltage still rides on top of that CM noise voltage, it's superimposed upon it.  So you use a differential amplifier, or CM filtering, or isolation transformer, or whatever suits the signal, to subtract out the CM error and get clean undisturbed signal referenced to circuit ground.  This works anywhere from DC to RF, given that we may want different ground-return impedance in those cases (high at LF, where shielding effectiveness is low, to avoid introducing voltage drop on the shield and therefore the signal reference too; low at RF, for CMRR).

And really, it makes sense, coax is unbalanced to its surroundings, so too should its impedance be unbalanced.  The signal line should never terminate towards the surroundings, but only towards its immediate reference plane -- inner shield.  Shield outside, common mode, can terminate with whatever impedance, but probably a low impedance to shunt CM current away from the circuit.

Tim

[liteyear]

USB-C is mostly used and first appeared in portable devices, so ferrite bead on a cable is a major usability downgrade.

I think you might be on to something here. There are forces at work, but a consumer product trend could trump them all.

[Harrow]

I tested a few USB-A plug pack power supplies to see how much noise they were introducing into my USB-powered oscilloscope. I had a couple of unbranded USB-A plug packs that were quite noisy, but when I used my Samsung ones, the noise was reduced by a factor of about 20 and my old Apple USB-A plug pack was even better. I tried adding a ferrite bead from an old laptop supply to the lead of the unbranded USB-A plug packs and the improvement was negligible.

[radiolistener]

Adding ferrite beads to cable cost additional money, so cheap cables don't have it. The same cheap Chinese switching mode DC/DC converters and power supplies don't have filters and produce very high EMI noise, but people still buying and using it.

[wraper]

Adding ferrite beads to cable cost additional money, so cheap cables don't have it. The same cheap Chinese switching mode DC/DC converters and power supplies don't have filters and produce very high EMI noise, but people still buying and using it.

The most expensive USB-C cables (including $100+ thunderbolt stuff) don't have them either. And I'd never buy one with ferrite to use with portable devices. HDMI and Displayport cables usually don't use them too, including expensive 48Gbps rated stuff.

[tszaboo]

Here is a list of ferrites from Würth:
https://www.we-online.com/en/components/products/pbs/emc_components/ferrites_for_cable_assembly/we_snap_ferrites_cable
They go from 1MHz to 1Ghz.
USB 3.1 signaling rate starts at 5Gbps.
So the ferrite wouldn't work at those speeds.

[T3sl4co1l]

Signaling rate is only a tiny consideration out of everything that ferrite beads are used for here...

Tim

[tszaboo]

Signaling rate is only a tiny consideration out of everything that ferrite beads are used for here...

Tim

It wouldn't attenuate ie common mode noise, which is a big the deal with the radiated emission testing. So what's the point of having them then?

[T3sl4co1l]

Immunity, conducted, whatever it's plugged into.

The primary purpose is reducing cable resonances; they're there regardless of signal rate, and exacerbate emissions when signals conspire to put peaks in that band (for USB, mostly by SE0 symbol, emitted at a much lower rate than bulk signaling rate anyway).  But it could also be PSU or other emissions from something merely connected, or something unconnected and transmitting, or nearby interference like ESD or EFT.

Emissions are mostly controlled by the shield (and this is your [readers'] daily reminder to readers that USB is mostly but not completely balanced, and receivers have narrow V_CMR (supply rails or less) so you can't run it bare wires or unshielded, not under EMC conditions), and the bead is take-it-or-leave-it for signal quality purposes.  At least until the shield wears down to nubs and stops making connection.

Tim

[tszaboo]

I meant to write immunity testing then had a brainfart. Big antenna in the anechoic chamber sending 5GHz signals into your 5gbit/s serial interface, very likely to cause issues. That's when you add the ferrite on the cable, and pinky promise to always use it with your product.

[T3sl4co1l]

Ah, yes indeed 

Tim

[Berni]

Cable ferrites are only about common mode noise suppression and that's it.

The high speed wires going trough it are traveling together and twisted, so they have no significant differential mode field around it. They just help more tightly couple the other non twisted wires and shielding together so that a common mode upsets have a tougher time transforming into differential mode upsets between the wires.

It doesn't magically improve signal integrity of high speed lines. It just makes it harder for interference to get in, be it get into the differential mode signal, or swinging the common mode out of a workable range.

They also help with passing EMI certification tests if the device is spewing out too much common mode noise on that port.

[AndyC_772]

They also help with passing EMI certification tests if the device is spewing out too much common mode noise on that port.

This, 1000 times over.

The first, last, and only reason to have a ferrite on a USB cable is because the device it's shipped with requires one to pass EMC compliance testing. If it's cheaper to add a ferrite to the cable that comes in the box than to delay the product launch for a redesign, then a ferrite is the solution.

If you buy a USB cable as an accessory, of course it won't have one - they're bulky, heavy and add cost.

Yes, your device may then radiate more EMI than it should, at some particular frequency and under some operating conditions that you may or may not use. It might - just might! - interfere with the analogue radio you no longer listen to, or be the reason your phone signal drops from 4 bars to 3 when it's in the same room. Your neighbour who's into ham radio will moan at you, but life will go on.

[coppercone2]

I like that quality first mentality.

[radiolistener]

Your neighbour who's into ham radio will moan at you, but life will go on.

I heard stories when they do revenge like using transmitter or magnetron to stop FM radio/music center/WiFi/GSM device operation.

[EE-digger]

I'm going to go with door #5.

As time goes on, SMPS designs have improved, moved higher in frequency, and the designers have learned better layout, component placement, and choice of the actual components.

A lot of failures are from the USB cable radiating power supply noise while others are from (mostly) digital sources, layout and decoupling.

ALSO, as geometries of components have continually shrunk, as have product sizes, it's very common now that for a given emitter, there aren't any traces long enough to be a good radiator, hence compliance failures go down.

[EE-digger]

I'll add torture to the list.

Ignorance in design for EMC leads to lost schedules, possibly halted production, loss of a critical market window due to delays in fixing.  All can be expensive to VERY expensive.

If the engineer responsible hasn't been flogged for one or more of the above, he/she is in a nice company    You learn how to do it better.