Hacker News Reader

Comment by e63f67dd-065b | Hacker News Reader

Not an electrical engineer, but 240w through typical USB wire gauges doesn't seem safe at all.

Does somebody have a good resource for the state of the USB-C ecosystem? How to determine which cables support what features, which icons mean what, etc.

View on HN

It's the same as right now.

We have 100W (so 20V at 5A), we will have 240W (48V at 5A).

Gauge doesn't change, since current stays the same

Well, except none of the cables or connectors of them was designed for 48V

That's not true, a lot of the connectors just-so-happen to be rated for mains levels. This is because often USB chargers aren't ground referenced, so they hover at a little above 1/2 mains voltage. This is why if you touch some chargers or devices that are charging you get this weird electrical "sensation" sometimes.

They are galvanically isolated. If you're getting any sensation at all, something is clearly wrong.

No, they're not. Two-prong power supplies need to have EMI suppression capacitors which connect the secondary side of the power supply's transformer to mains [1]. The mains voltage is present on the output, but the maximum current is low enough that it is safe. It is considered not an issue by manufacturers [2] and it is a fact of many practical power supply designs. Please have any idea what you're talking about before you throw around technical terms like "galvanically isolated". You're clearly not an EE.

[1] https://electronics.stackexchange.com/questions/478607/feeli...

[2] https://www.dell.com/en-us/blog/notebook-tingle-sensation-wh...

Being connected via 2 caps is considered "galvanically isolated"

If you're feeling "tingling sensation" your protection caps are fucked Mr. EE

Galvanic isolation means "no current is allowed from primary to secondary side". This is true of capacitors... but only in theory. Real life capacitors have leakage resistance and therefore do not, in practice, form full galvanic isolation.

Capacitive coupling still counts as galvanically isolated. Dielectric resistance of typical Y caps is in GOhm range, so that's basically galvanically isolated, unless you're some kind of purist. You certainly will not be able to feel that.

Reactance of a typical Y capcaitor at 50Hz is in the range of single unit MOhm so even that does not cross perception thresholds of most individuals even at 230 V. But it may for some, I guess.

Also what the heck with the first link. Apliances with metal enclosure must be grounded. That guy has no business having 50V against ground on the metal enclosure.

48V at 5A over 3ft seems fine with 21 AWG copper wires: https://www.omnicalculator.com/physics/dc-wire-size

It really depends on the voltage. Ohm's law says that current is proportional to voltage divided by resistance.

I = V/R

This means that as voltage increases with a given fixed amount of power, current decreases. Indeed, a watt is a measurement of power calculated by multiplying current in amps by potential in volts.

Joule's equation of heating says that the power (P) lost to heat in a circuit is equal to current (I) squared, multiplied by resistance (R).

P = I^2R

Thus, increasing voltage reduces heat generated over a given conductor.

Which is why major power lines run at thousands of volts as I understand it.

240W @ 48V (voltage limit for "low voltage" stuff) is "only" 5A. On the high side for a small cable but feasible nevertheless.

and for >3A special cables that signal that they can handle 5A are required.

I have one such cable. It's fairly thick, but it doesn't say anything on it.

The signals are electrical, not optical.

I have a Thunderbolt 3 USB-C cable capable of 100W power delivery. The gauge on that cable is thick, not your everyday USB-C cable.

> 240w through typical USB wire gauges doesn't seem safe at all

That's why you need an e-marked cable to be able to do that.

The safety depends entirely on the voltage, no?

Not really no - the term for if a wire will melt/catch fire is ampacity, which depends on amps.

It’s very possible to vaporize a wire with just a few volts with enough amps!

The safety issue with voltage is due to the need for thicker insulation to avoid shorts, but the difference for 5V vs 48v isn’t that much.

It’s possible to carry a lot more power for the same amps (aka same wire ampacity) by upping the voltage, as Watts=amps*volts

That is why interstate power transmission goes up to millions of volts.

Most electrical codes also have carve outs for ‘low voltage’ as it’s tedious to try to regulate all the various ‘not dangerous’ stuff, which is considered 48 volts and below, which is why it’s limited to 48v. It’s considered ‘not dangerous’ due to the resistance/insulation of human skin making it a bit harder to kill your self with it, but with some work, still possible. It doesn’t take much amperage, if you’re trying - USB 2.0 has more than enough.

AC needs thicker insulation vs DC, but that is due to the peak of the AC waveform being much higher than the nominal voltage. 170V for ‘120V’ AC, for instance.

Cable fires are caused by the current flowing through a wire exceeding its current rating

Voltage is largely irrelevant

Have a look at the gauge of mains wires (~14-22 AWG) compared to what you'd put on a serious 12V-48V battery (2-8 AWG)

Reply on news.ycombinator.com