Hacker Newsnew | past | comments | ask | show | jobs | submit | gubby's commentslogin

Interesting idea, I suspect it isn't possible with the existing connector. You either need to increase the voltage or the current.

The new 240W spec is 5A and 50V. Assuming 5A is the thermal limit of the connector and that can't change, to get 1500W you'd need 300V, which is probably higher than the connector can support in terms of conductor spacing.


300V DC is also quite dangerous, 48v/50v is already up there.


Maybe you can get there if the device doesn’t actually require a steady stream of 1500W. For example, induction stovetops often work with cycling their coils. Couple the idea of a charge pump, a whole bank of capacitors and Tesla’s trick of running cables over spec by constantly monitoring temperature and making on-the-fly adjustments, and you might get somewhere.


You may be able to run cables 50-100% over spec for a short time. But not 1000%. Thought experiment (don't try at home!): Take a random USB cable and short circuit a car battery. It will start burning within seconds.


If you short circuit a car battery with a wrench it will melt the wrench.


50 V in the USB-C connector ?

What can go wrong ?


In power-over-ethernet setups, you can have up to 57VDC at the consuming device: https://en.wikipedia.org/wiki/Power_over_Ethernet#Standard_i...


50V is pretty much the limit that is still considered safe for humans to get in direct contact with.


At uni we learned that 30V can penetrate skin in the right (wrong?) conditions. Once it's through the skin, your body conducts really well, so even a tiny amount of current can stop your heart. But you need a path for the current to get across your heart, so I'm guessing that even if 30V gets under your skin, you'd have to be pretty unlucky for it to form a circuit through your heart.


plenty of grounded things you might be gripping with your other hand at the same time :(


And by safe it means 'won't die' not 'pleasant feeling' still be careful with 50v!


With dry hands it shouldn't be bad, with wet or sweaty fingers it probably can become quite a "shocking" experience.

If there is some liquid poured over the connector or if the cable is damaged, this could really become nasty.


As a teenager I touched a phone plug with my tongue. While it only delivered 48V it was far from being pleasant


You're just lucky it didn't ring at the time!


Touching a 9V battery to your tongue isn't pleasant.


Having done both I can assure you that the phone plug experience is far sharper. It's the similar "lemony" shock, but far far stronger


Licking a Nintendo game card is also not pleasant.


Which Nintendo console, what game and for what purpose?


Mmm yes 50 V over pins that are half a millimetre apart. I see no problem with this. No problem at all.


> There's not a single smart plug on market that won't connect to cloud to control plug from anywhere

Look at Shelly devices, they are exactly what you are looking for (in no way affiliated)


Interesting point, but of course it’s not possible to create a platform with an average win rate of 80%.


In some video games you may randomly face an AI opponent without the game being explicit about it. That way the average human win rate can be above 50%. I don't know yet if this is good or bad.


Nor any actual group of people. This "optimal" is zero sum: someone else has to have a lower winrate in order for it to work.


And yet when you see a professional boxing match, both boxers' win rate is usually above 90%, not 50%.


A boxing career has approximately 0 rated fights. A cheap career has hundreds to thousands.


Grounding isn't really about dispersing energy over an area, it's about creating a situation where faults can be detected and circuits automatically opened to prevent shocks or fires. It actually gets very involved to reason through the exact purpose of ground connections when you start considering all the different types and combinations of faults (indoors, outdoors, person in contact with true earth, person touching exposed metal pipe, etc etc).


Very often with contract law there is mostly no such thing as 'knowing a claim is not legally valid'. A lot of contract law is established by precedent, meaning in those cases nobody including the judge knew the enforceability of the issue before the judgement was written. Cases are also rarely identical examples of previous cases, meaning there can be an element of precedent setting even in cases that seem routine.


Yes. In those cases it is fine.

I don't think it should be illegal to have unenforceable clauses. Just ones that are known to be unenforceable. For example look at your average ISP out TV provider contract and there are a handful of these designed to prey on the public that doesn't know better.

It would definitely be difficult and messy to enforce, quite likely to messy to be feasible, but I would like to see some rule like this passed.


I suspect (IANAL) that in a sense, it already is, in that anybody testing this in court would deny the existence of any form of contract acceptance, and that would be the end of it.


And that heat ends up at the worst possible place, on the junction itself, meaning the device has to also work to remove it.


For an article about a specific scientific concept (the proportional land use for different types of generator), it’s surprising they actually got that concept wrong in all their examples:

“These can have a small land footprint of just 8m2 per MWh”

This is implies that the land is being consumed by the generator. I assume there is an implied “per year” or “per day” but it’s a little sloppy.


They look identical to me too, but I know that I can perceive RGB hues.


This has been true of OSX, especially since around 10.15. The performance degradation of the upgrade to 10.15 from 10.14 was so immediate and severe that I reverted it.


The interesting thing about this is that early Mac OS X was the opposite, with each release being faster than the previous. The very first consumer release of Mac OS X (10.0, Cheetah, 2001) was said to be rather sluggish. Addressing this, Puma was an improvement over Cheetah, which was then replaced with the even faster Jaguar, which was replaced with the even faster Panther, which was finally replaced with the even faster Tiger. There were some slowdowns when Leopard was introduced (which is why I have Tiger installed on my PowerPC Macs), but this was fixed when Snow Leopard was introduced, which had no new consumer-facing features but had many optimizations and infrastructural improvements.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: