Hacker News Reader

Comment by userbinator | Hacker News Reader

This means there’s a small microcontroller inside the cable with information on the cable’s capabilities.

More evidence that USB-C is an insanely overengineered spec. Cables should be dumb pipes, not devices with their own active circuitry. IMHO Ethernet, while not perfect, got this part right.

The only reason the signals don’t exactly match the one above is just the orientation of the cable.

This is one of the more perplexing design decisions. Was simply mirroring the contacts too simple for them? Did someone imagine a use-case where they wanted to detect the orientation of the cable? WTF.

View on HN

So how’d you safely solve having cables of (widely) varying resistance?

I appreciate having the option to choose between a thin cable for charging only, and a pretty thick (and also heavy, inflexible, and relatively expensive) one for high-speed data transfer and charging.

The only alternative to that would seem to be having different ports per device, which is infeasible in many cases, and inflexible in most others.

> Was simply mirroring the contacts too simple for them?

That would require twice the number of contacts for the same number of physical wires, which would make the plugs even larger.

So how’d you safely solve having cables of (widely) varying resistance?

Smartphones already do this. They essentially measure the resistance of the cable by measuring the voltage and correlating it with the current drawn, and regulate the current accordingly.

That would require twice the number of contacts for the same number of physical wires, which would make the plugs even larger.

No. Look at the pinout. It's not symmetrical. They could've made it perfectly symmetrical.

That's not true, see this random pinout diagram: https://acroname.com/sites/default/files/shared/alt_mode_-_d...

> It's not symmetrical. They could've made it perfectly symmetrical.

Without internally connecting all corresponding pins to the same wire? To what benefit?

The parts that need to be symmetric for passive legacy devices already are.

> Smartphones already do this. They essentially measure the resistance of the cable by measuring the voltage and correlating it with the current drawn, and regulate the current accordingly.

Not sufficient. It’s easy to have a cable where the long conductors can safely carry a lot more current than the terminations. Or one could have a 1 foot thin cable connected in series with a 10 foot thick cable.

For that matter, a short low-ampacity cable would look just like a long high-ampacity cable.

On top of that, a slightly imprecise voltage source could also easily look like a very low resistance cable.

It's differential impedance that they measure, not the absolute voltage.

What would they measure that for?

To determine how much current they can draw.

They can only measure combined voltage sag of the source and resistive losses in the cable though that way, no?

Doesn’t help with the concern of high-resistance spots somewhere in a low-current-rated cable.

They aren’t trying to deal with poorly manufactured out of spec cables.

But resistance in series is additive - the resistance of the combination is dominated by the resistance of the worse cable: 0Ω + 1Ω = 1Ω

Not sure what problem you're thinking of.

Most of the resistance occurring in a small part of the entire cable, potentially causing a fire, seems possible, no?

No. You could also suppose a dead short in the cable that causes a fire, or just a cable with a length of detcord in the middle.

It's not resistance per se that you care about, it's how much the cable is heating up which is mostly dependent on resistance per meter.

Imagine 2 cables with the same resistance, one 0.5m long and the other 2m. At a set amount of amps, the 0.5m long cable would need to safely dissipate 4 times as much heat than the 2m cable over the same distance. And you don't know how long the cable actually is (and they make 10cm usb cables after all) so you can't make any decision based on resistance that doesn't risk some cables going up in flames.

Fortunately copper and aluminium, the two most common metals for cables, have resistance that increases with temperature.

Ultimately what I'm saying is that the endpoints have or can measure all the information they need to adapt, and this is even more accurate than requiring a cable to self-identify which is worthless if it's lying (and that can happen unintentionally if it's damaged or the connector is worn.)

The copper and aluminum will not become resistive enough to make a difference at temperatures low enough to prevent the rest of the cable from becoming crispy.

You're not thinking this through.

Yes resistance goes up the warmer the cable gets, but you know neither the resistance of the cable at 20°C nor the temperature of the cable. Simple example, say the user is charging their phone, you detect the cable getting plugged in and measure the resistance at say 1 Ohm to make the numbers nice. Cool, now at what resistance do you determine the cable is too hot and reduce the current? Copper's temperature coefficient is about 0.4% per °C so the resistance should be 1.12 Ohm at a safe 30°C increase right?

Wrong! The cable resistance you measured could already have been at 50° and you're now pushing the cable into fire hazard territory. This isn't theoretical either, people plug in different devices after another all the time (this also eliminates any sort of memorizing if you were thinking about that). So what are you just going to wait 5 minutes at safe charging current and see if the temperature goes down? That's not going to fly in a market where some devices charge 50% in like 20 minutes.

And all of this is ignoring another important design goal of usb-c: dumb charge-only devices exist and should continue to function with a minimal of extra added components. USB-C allows this with the addition of two simple resistors that set the current you want. Measuring resistance on the other hand either requires a accurate voltage source on the device side (expensive) or somehow connecting the power line to some other line (how that would work without additional control electronics I have no idea).

a safe 30°C increase

That's far higher than what the implementations assume.

>Smartphones already do this. They essentially measure the resistance of the cable by measuring the voltage and correlating it with the current drawn, and regulate the current accordingly.

No! Where did you hear this?

Smartphones usually either rely on type-c pd nowadays or do far less standard Samsung/Apple resistor on usb line check.

Some power banks also do terrible thing where they increase current pull until they see voltage starting to collapse but that’s to check for OC limits on load switch. However that’s really starting to go away because type c just works

> Some power banks also do terrible thing where they increase current pull until they see voltage starting to collapse

If I’m not mistaken, that was explicitly allowed under USB Battery Charging. But chargers have to opt in to that via signaling on D+/D-.

It's been common for many years. Look for a document titled "Mediatek Pump Express Introduction". Another useful phrase to search for more information: "cable impedance detection". Apparently several patents on it too.

>Mediatek Pump Express Introduction

Yep read it. It does even more terrible things (modulating current to change voltage), but no monitoring for voltage drops.

It’s also rarely supported, and looks like they just switched to pd3 (the sane protocol)

but no monitoring for voltage drops

Did you not see:

"Do cable imp. Calculation. Protect cable from burnout" on page 12

"Adjust Vbus according to cable impedance/quality" on page 14

and "Cable Impedance Measurement" on page 15?

Pump Express is their fast charging protocol, but even before that, in the days of feature phones, they were already doing the measurements and adjusting charging current.

I've had a phone like this - it'd increase current until voltage start to sag. It was horrible. Never again.

The difference between high-current and low-current cable was very small, so once cable got a little bit worn out, the detection was basically random. You plug cable, and get 1A charge. Unplug-replug, and it's 0.5A now. Unplug-replug and it gets 2A.

I am very glad that the system has died out and we have cables that simply report their capacity directly.

(that said yes, it could have been simpler, like having a symmetric pinout. But this would be a marginal improvement at best, as USB-IF seems to be incapable of producing simple designs. It'd still be thousands of pages of docs)

You just had a bad implementation.

cables that simply report their capacity directly.

...which won't true when worn or the manufacturer is lying.

I don't think it is possible to have "good implementation" of cable impedance sensing, given the inherent properties of the connectors. Wetting current is a thing, after all, and it can be quite big for work connectors.

Cable-reported capacity does not really change over cable lifetime - the currents used to sense resistance are miniscule, and would not be affected by cable being worn. For example, extra 0.5 ohm of cable resistance completely ruins the current sensing, but make absolutely no difference when measuring 5K or 1K resistance.

And if manufacturer is lying, that's annoying but it's 100% deterministic - you know the cable is bad, you can throw it away. With impedance sensing, the same cable can be good or bad, depending on the mood, and that's why it should not be used ever.

Ethernet doesn't support 240 W.

I am very happy that I can charge my phone and laptop from the same charger, and don't ever have to worry about whether the cable I'm using will be a fire hazard.

PoE has been around for longer, and while it doesn't go that high (90W --- but at much longer distances), it uses passive cables just fine.

POE is an quite expensive thing to implement on a board. Flyback transformers are essentially required to support the standard.

At the cost of much higher complexity on each end. USB-C still supports completely passive old chargers and devices.

The cable can still be a fire hazard if it’s a cheap cable that lies about its rated wattage, no?

That goes for anything that lies about its rating though

Right but consumers are much more likely to buy a cheap USB charge cable that lies about its rating (via internal chip) than an Ethernet cable that lies about its rating on the package.

It doesn’t seem to be a huge problem with extension cords.

Do ethernet cables even specify a wattage?

The reason for detecting the orientation of the connector is for higher speed communication. USB-C 20gbps uses both sets of pins on the connector to shotgun two usb3.2 10gbps to get 20gbps. That is why the technical spec name for 20gbps is "USB 3.2 gen 2x2". That is what the "x2" means.

Knowing that USB has this feature is follows that USB-C needs to be self orienting in case both ends of the connector plugged in different orientations.

You say Ethernet got this part right, well it got this part right by not having a reversible connector. Ethernet has 4 tx/rx pair and USB-C has 2 rx/tx pairs per usb 3 connection with 4 in total for 20gbps. The difference is reversibility. Is it worth the tradeoff?

Most if not all Ethernet NICs have auto crossover and polarity detection. Some tolerate arbitrary lane swaps too.

That might work for Ethernet, but how would you do that for any unidirectional USB-C alternate mode without protocol-level feedback such as analog audio or DisplayPort video?

If you want to allow all of

- Reversible connectors

- Passive (except for marking), and as such relatively cheap, adapters and cables

- Not wasting 50% of all pins on a completely symmetric design connected together in the cable or socket

there's no way around having an asymmetrical cable design that lets the host know which signal to output on which pins.

You do it like Lightning did with a chip that controls the pinout as well as reporting the use case.

That’s basically how USB-C does it too (except that the chip isn’t strictly necessary; an identifying resistor does the job for legacy adapters and cables).

USB-C was heavily inspired by Lightning.

> Cables should be dumb pipes, not devices with their own active circuitry.

You only need the active circuitry for high-power (240W) or high-speed (Thunderbolt) cables. Ordinary cables just need a single, cheap resistor.

C-to-C cables never have a resistor for identification; those are for devices or legacy cables/adapters to USB-A/B.

“Active” is also somewhat ambiguous: E-marked cables can be both passive or active in terms of what they do with the signal on the high speed data lines (such as amplifying it, converting it to optical etc.)

> More evidence that USB-C is an insanely overengineered spec. Cables should be dumb pipes, not devices with their own active circuitry.

Why, to what end? They don't add any noticeable amount of cost to the cables and it's a whole lot better of a solution for consumers then requiring all cables to carry 5 amps and thus making them thicker.

> Did someone imagine a use-case where they wanted to detect the orientation of the cable?

They didn't imagine anything, USB3.0 5Gpbs with only 2 differential pairs, like you'd have in a usb-c to usb-a cable, requires this. And you can't just connect them together like you'd do for USB2, the resulting stubs degrade the signal too much.

Reply on news.ycombinator.com