Want to read Slashdot from your mobile device? Point it at m.slashdot.org and keep reading!

 


Forgot your password?
Close
typodupeerror
×
Hardware

Demand Grows For Tiny Phone Chargers Using 'New Silicon' (ft.com) 91

Posted by msmash from the shape-of-things-to-come dept.
A tiny phone, tablet and laptop charger, the first to use gallium nitride rather than silicon chips, has seen sales four times greater than predicted [Editor's note: the link may be paywalled], prompting the Chinese company behind it to try to ramp up production. From a report: Anker, a Shenzhen-based company that specialises in computer and mobile phone accessories, unveiled a line of chargers using gallium nitride (GaN), which conducts electrons 1,000 times faster than silicon, in January. The use of GaN allowed Anker to virtually halve the size of its charger, while retaining full-speed charging. Another Chinese-owned company, RAVPower, has also started using GaN in its chargers. "Silicon limits have been pushed almost to the extreme," said Steven Yang, co-founder and chief executive of Anker. "But GaN is at [the next] phase."

The introduction of the new semiconductor into the consumer market came after a series of military and other commercial applications, in everything from electric vehicles to radar systems. Raytheon, the US defence group, said in 2017 that it had spent $300m researching GaN since 1999. Like some of its peers, it uses the material in its active electronically scanned array (AESA) radars, which are able to detect stealth fighters at long range. "Once the power technology is out of the box it will be widely adopted around the world and that means everyone can produce power-switching modules," said Stephen Bryen, a former deputy undersecretary of defence and senior fellow at the American Center for Democracy. "And that is what is used in the radars -- that's the nexus between commercial and military use."
This discussion has been archived. No new comments can be posted.

Demand Grows For Tiny Phone Chargers Using 'New Silicon' More

Demand Grows For Tiny Phone Chargers Using 'New Silicon'

Comments Filter:

  • Those look great!.

    The headline is one of the worst I have ever seen though, no link to a product to look at, and the only actual link is a HUGE PAYWALL, not "may be paywalled" lol

  • Like Li-ion, then what is the point?

    • Re: (Score:2)

      by duk242 ( 1412949 )
      You're the MVP :3

    • That's nice, but it still takes up two outlet spots by the looks of it. It's well past time that designers rethink the wall-wart so that we can plug two in next to each other. It's ridiculous that I need a multi-outlet strip to plug two items in.

  • Could someone summarize this for me in a way that isn't scientifically illiterate?

    • Not here...

    • Re:"conducts electrons 1,000 x faster than silicon (Score:5, Funny)

      by Gamasta ( 557555 ) on Thursday July 04, 2019 @03:54PM (#58874208)

      Sure!

      Electron travels at about a third the speed of light "c" in common conductors such as metals.

      This new super material conducts electrons at about 1000x that speed, so roughly 300c. Since c is the speed limit for pretty much all particles (save tachyons) and c is the barrier for causality, time and light, it means that electrons conducted by GaN travel faster than time itself. Thus your phone would be charged before you plugged it in.

      It's just amazing! No wonder 4x more people want to buy these chargers (did they project a single sale?)

      • Re: (Score:1)

        by Anonymous Coward

        it was originally 4 billion times more people who wanted to buy the chargers (now and in the future) but at 300c the Lorenz contraction brings it down to just 4x

      • Re:"conducts electrons 1,000 x faster than silicon (Score:4, Informative)

        by doug141 ( 863552 ) on Thursday July 04, 2019 @05:18PM (#58874540)

        No, the electromotive _force_ moves that fast, but the electrons only move about an inch a minute.

        • No, the electromotive _force_ moves that fast, but the electrons only move about an inch a minute.

          Is that the only mistake that was in the GP's post?

      • Re: (Score:1)

        by Anonymous Coward

        My laptop is charging on one of these right now, I read this post before it was written.

      • "This new super material conducts electrons at about 1000x that speed, so roughly 300c. Since c is the speed limit for pretty much all particles (save tachyons) and c is the barrier for causality, time and light, it means that electrons conducted by GaN travel faster than time itself. Thus your phone would be charged before you plugged it in."

        It's obviously could not have been made without the Endochronic Properties of Resublimated Thiotimoline and so it has a US patent.
        https://en.wikipedia.org/wiki/... [wikipedia.org]

    • Re: (Score:2)

      by robi5 ( 1261542 )

      Big nothing. The claim is that using GaAs, these power adaptors - which are, these days, all switch mode, ie. based on semiconductors rather than electromagnetic coils - can be made smaller. Linked article is fully and knowingly paywalled. What products others lined seem to raise a lot of questions:

      1. OK nice that theverge shows the things next to a 5W charger, but no idea how eg. the 45W adaptor compares to a MBP brick. It looks like maybe the same size, except flatter, oh and it covers the socket surround

      • Re: (Score:2)

        by robi5 ( 1261542 )

        Btw. I don't understand these self-professed tech writers who have "decades of experience" etc. - they don't even do the obvious thing when the claim is that these things are smaller: no side by side photo anywhere. No money to plunge for one, despite journalists traveling a lot and benefitting from small bricks? Or even these early adapter adopter folks don't trust these? Again the 60W Anchor, on the photo at least, looks as large as the standard adapter.

      • Re: (Score:2)

        by Guspaz ( 556486 )

        The 30W Anker charger (the Atom PD1) is roughly the same size as the 5W iPhone charger. That's quite impressive.

        I agree that the 60W charger seems extremely large in comparison to the 30W charger. I don't understand why, you've got 2x the power and ports, and yet it's like 10x the physical volume.

    • Re:"conducts electrons 1,000 x faster than silicon (Score:5, Informative)

      by AmiMoJo ( 196126 ) on Thursday July 04, 2019 @05:18PM (#58874550) Homepage Journal

      I think they mean that using gallium they can create transistors and diodes with 1000x lower resistance, and therefore lower waste heat.

      1000x is a bit of an exaggeration, but the tech does work. These are basically little switching power supplies, which as the name suggests uses a silicon switch in the form of a MOSFET. The switch gets warm because it's not a perfect conductor of electricity, some of that energy gets turned into heat. Making it more efficient means you can have a smaller power supply due to less need to dissipate heat.

      • Re: "conducts electrons 1,000 x faster than silico (Score:2, Informative)

        by Anonymous Coward

        To add to that, the real savings in size comes from the smaller inductors and capacitors that can be used due to higher switching speeds. The higher switching speeds are possible because of the lower resistance and lower heat dissipation, partly, but also because the larger band-gap and of the GaN devices and the higher electron mobility just plain allows them to switch faster, in absolute terms, on-and-off. Also, it fewer stages are needed because higher voltage can be switched per chip. Less stages means

      • Does this not have more applications in Graphics cards and CPUs than chargers?

        • Re: (Score:2)

          by AmiMoJo ( 196126 )

          Maybe... I think the issue may be the cost. One for a single charger isn't too bad, but if you look at the power regulation for a high end CPU or GPU there are a lot of FETs in there. Silicon ones are really cheap but maybe for really high end stuff or heat constrained stuff like servers it makes sense.

    • Re: (Score:2)

      by mark-t ( 151149 )
      I would take that to mean that propagation delay through a circuit is 1/1000th of the time it takes in an otherwise equivalent circuit built on Si. That is, however, just a guess.
    • Not from the Article but information about GaN FETs.

      They are faster but not in the actual speed of electricity but the speed at which they can switch on and off efficiently. A Transistor (GaN or any type) when it is on has very little power loss, when it is off has no or little power loss but when it is switching (transitioning from on to off or off to on) has power loss. The faster this can transition the lower power loss (this is called switching losses.) Does this mean less heat, yes all else being equ

      • So "conducts electrons 1000x faster" really means there is much lower intrinsic capacitance (which is the source of the "switching loss").

    • Transistors with lower 'on' resistance made with gallium instead of silicon mean less heat dissipated in the charger, so it can be smaller.

      Also the charger is a proper USB-C PD charge so it can negotiate higher voltage with the device and so send more power at lower current, reducing the losses in the cable.

      So a combination of new semiconductor tech and proper product design.

    • Essentially electrons in crystalline materials (which GaN and Si are) behave "quantumly". This can lead to weird effects such as an electron appearing to be heavier or lighter than you would expect depending on the material. Another property is the speed at which the average electron appears to be moving (drift velocity) when a given electric field (E) is applied. Mobility is simply dv/dE which ideally is linear (isn't the case at high E). The crystallinity of a material is important to this value. Gra

  • I don't get it (Score:1)

    by Anonymous Coward

    the bulgy part of a PSU is the transformer, who cares about the silicon? did they put a PSU in silicon?

    • Re: (Score:2, Informative)

      by serviscope_minor ( 664417 )

      the bulgy part of a PSU is the transformer, who cares about the silicon? did they put a PSU in silicon?

      The higher electron mobility of GaN means you can switch faster without dissipating too much heat so you can have smaller magnetics. They're also good to 400 degrees so you can run them small and hot.

    • Obviously, you don't live in the UK. For a couple of decades, the largest part of any transformer has been the 3-pronged mains plug.

      • Obviously, you don't live in the UK. For a couple of decades, the largest part of any transformer has been the 3-pronged mains plug.

        This is an interesting history behind the UK plugs. They were originally made in 1940 to be used as obstacles to prevent the Nazis from landing on the beaches of Dover and advancing inland. They were designed to inflict maximum debilitating pain to the foot of any German soldier stepping on one. But the invasion never came, so after the war they were repurposed as electrical plugs.

  • I'm calling bullshit on this (Score:3)

    by ukoda ( 537183 ) on Thursday July 04, 2019 @05:32PM (#58874616) Homepage
    Ok so GaN can switch the same current in less space but if you open up a charger it not the silicon taking up most of the space, it is the passives, particularly inductors and transformers. This is going to be an incremental improvement at best, not the huge break thru the article claims.

    • Re: (Score:3, Interesting)

      by nonBORG ( 5254161 )
      It is the ability to switch faster (more times per second) with the same or less heat that allows other parts to shrink.

      • Re: (Score:2)

        by ukoda ( 537183 )
        Sounds like an invite to EMC compliance issues to. Seriously, I think most SMPSU run in the 100kHz to low MHz but silicon can quite happily switch 100s of watts at 100s of MHz so I don't think switching frequency is the key to smaller PSUs. My experience with SMPSUs is getting inductors right is the key.

        Sure I could see this kind of tech replacing a TO220 and heatsink with SO8 package and taking 10% to 20% volume out of PSU but that does not justify a "Tiny Phone Chargers" headline. I guess "Reduced S

        • Re: (Score:3)

          by tlhIngan ( 30335 )

          Sure I could see this kind of tech replacing a TO220 and heatsink with SO8 package and taking 10% to 20% volume out of PSU but that does not justify a "Tiny Phone Chargers" headline. I guess "Reduced Size Phone Chargers" does not get as many clicks, hence my old man grumbling, I don't like marketing hype. BTW the best way to get a tiny charger today is to put them a metal case so you can jam everything together and still get the heat out. Also helps with EMC, but plastic is cheaper so bigger it is.

          The main

          • Re: (Score:2)

            by AmiMoJo ( 196126 )

            The limit on size will be the separation needed on the AC side, which itself depends on the AC voltage. In 100-120V countries you can get away with less separation. Most of these chargers are universal voltage though.

            The AC needs to be separated from itself (live and neutral) and from the low voltage side. It's also a good idea to make the prongs that go into the AC socket pretty robust, and legally the charger has to surround them by a certain amount (i.e. they can't be on the edges where they could get po

    • Yes, the things limiting the size are typically a transformer and a heatsink. You can have smaller transformers if you increase the switching frequency, but that tends to decease efficiency (and need bigger heatsinks) unless you use higher spec ferrites and have faster rise and fall times. GaN could help with the lower rise and fall times though I was under the impression that the speed of silicon mosfets wasn't really the limiting factor (more likely the ferrite). It would be interesting to know what frequ

      • Re: (Score:2)

        by ukoda ( 537183 )
        My experience with SMPSUs is they run from 100kHz up to about 2MHz and for transmitters I have seen MOSFETs used for over 100W and over 100MHz so I don't think MOSFETs are the limiting factor. GaN does sound like a step forward and it would also be interesting to see how they could be used for RF power applications. I still take exception with the claims of "Tiny Phone Chargers" when the photo only show a modest, if any, size reduction.

    • If you have fast enough transistors, you can generate switching frequencies that need less magnetic core material. Go fast enough, and you don't even need any core material at all, using an air core transformer.

      • Re: (Score:2)

        by ukoda ( 537183 )
        Yes, I understand this but MOSFETs can already switch much faster at the required power levels required, admittedly in a bigger package, but PSU designers are not doing that for whatever reason. I'm not arguing that GaN is not a size improvement, more call bullshit on the 'Tiny' claim in the headline when 'Reduced' is what we are talking about. Take a look at the photos of the amazing products featured in the article and be prepared to NOT be amazed at how tiny they are.
  • The article mentioned GaN as a connection between the controversial death of Shane Todd, PhD 31-year-old engineer. Todd was leaving Singapore after being concerned about Huawei, China, US security laws, work related threats, and being asked to commit export violations of US high tech.
    https://www.justice4shanetodd.... [justice4shanetodd.com]
  • Why are people so fascinated with smaller? I thought this article was talking about a power pack, because you need to fit that in your pocket. But a wall wart? It pisses me off when they can't be plugged in side by side but beyond that who cares?

  • I don't care much for a smaller, cooler or faster-switching PSU. But if these transistors are 1000x faster, can they be used in CPU:s or other chips to make them faster?

    • Re: (Score:2, Informative)

      by Anonymous Coward

      Both Si and GaN can achieve comparable switching speeds. Both have comparable high-electric field saturated velocity and comparable low-electric field mobility. The most significant difference between them is their energy bandgaps. Higher bandgap means GaN can sustain an electric field that is dramatically higher than Si before avalanche breakdown occurs. This means that GaN and Si devices with identical architectures (as identical as practically possible) will find the GaN avalanche breakdown voltage t

  • we've been using GaN devices for years for Ka radios, sometimes with devices so fresh from the factory that the part number produces zero google hits!

Slashdot Top Deals

To stay youthful, stay useful.

Close