He’s presumably thinking of the way gallium eats its way into and through aluminium, which makes having metallic gallium on board of a plane a very, very bad idea.
Gallium nitride’s a very different thing, however. It’s like the difference between table salt (NaCl) and metallic sodium & chlorine gas, both of which are very bad for you.
It’ll be part of a semiconductor, so firstly there’ll be probably only a few milligrams of actual gallium, and it’ll be part of a transistor, so basically in the form of sand.
There is nothing in the linked article to indicate why a laptop charger would become 4x smaller when made with GaN transistors instead of Si transistors. The main determinants of laptop charger size are heat dissipation and transformer size. The transformer can get smaller at higher switching frequencies, dependent upon transistor gate capacitance and other electrical parameters. The heat dissipation requires some surface area to get it out. Perhaps the switching speed is higher? The article didn’t seem to be too interested in the actual facts of the product.
I’ll have to look this up when I get a chance. I work in electronics, but I haven’t heard of GaN transistors yet. Perhaps because I stopped following the trade rags when they went digital.
And UV-LEDs down to about 250 nm.
Only if by “lucky” you mean it will charge your computer.
Basically (from my 30 second skim) it’s a lower RDS(on) per area compared to silicon. To get a low RDS(on), silicon mosfets are big so that they act as essentially hundreds of mosfets in parallel. But this, of course, leads to higher gate capacitance, a longer dead-time between switching, and lower frequencies (=bigger magnetics). So, GaN’s lower RDS(on) leads to smaller devices, smaller gates, lower capacitance, and higher frequencies. I’m sure there are downsides; it seems like they can’t make 600V parts yet, so silicon still rules in many applications.
ChargerLab has a lot of teardowns and in-depth reviews of chargers.
Unfortunately the USB-C PD standards are very poorly implemented, sometimes dangerously so, and any charger, cable or battery bank must be considered suspect until proven innocent. Even Apple took 3 tries to get their USB-C charging cable right, and even a reputable brand like Anker’s 100W PowerPort Atom PD 4 is proving unable to recharge my MacBook or iPad reliably.
One further consideration: the fold-up prong designs are very compact, but if you travel internationally often, a charger with a “figure 8” IEC-C7 socket like Apple’s is more versatile.
My (very layman’s) understanding is that GaN is amenable to high switching speeds, by the standards of power transistor sized things at least, not sure how it compares at teeny-logic scales; and that it also tolerates high temperatures and power densities comparatively well.
Apparently extremely popular in military radar and RF applications; only recently cheap enough to be relevant for making laptop power bricks smaller. I’m having trouble finding the article now; but I ran across a piece a little while back where someone from the military applications side was fretting about the (fairly obvious) implications of GaN showing up in cheap consumer electronics in the field of who can afford, and has the capability to build, fancy high density active electronically scanned array radar systems; something that’s a lot easier if you can get good GaN parts off the shelf at limited cost and with limited hassle.
If you’re interested in why it’s so hard to properly implement charging over USB-C, then I recommend reading the document that describes it. It’s over 500 pages long.
I wouldn’t be too concerned. Maybe if your 480 is a fully-loaded, fastest i7 they offered, punchiest discrete GPU, brightest and highest resolution panel, max RAM and storage; but for more pedestrian configurations it’s definitely not a problem.
I’m not 100% sure why(possibly as a service to customers operating mixed fleets, possibly as some sort of supply chain optimization for themselves); but they default to shipping their 65w charger even for most models where their own 45w charger is explicitly supported. Handy if you do have a mixture of everything from power-sipping X-series ultraportables to T580/590 systems kitted out with all the options, potentially even some of the less demanding workstation models; but overkill for a user of a single one of the less demanding models.
Perhaps @beschizza could tell us whether it was a 45 or 65w Lenovo adapter he replaced; you could also use the strategy of checking with Lenovo whether their 45w adapter is a supported part for your system; and if so knowing that the 61w 3rd party adapter is likely to be fine.
You could also use powercfg (if running Windows) or powertop (if Linux) to pull power consumption data from your system, ideally while you are giving it a worst-case load.
If the T480 is one of the models that supports the recent ‘fast charge to 80%’ behavior(the 490 is, can’t remember if the 480 got that, offhand) a lower wattage adapter may well increase charge time a bit; the highest stress use case is typically trying to dump charge into a battery while the system is also being used normally; but that becomes less of a concern if you hate your AC adapter less.
The laptop is a T470 and the bricks’s 90W, but it also charges just fine from LG UltraFine 21.5" monitor’s USB type-C port, which I think is 60W.
I’ve seen people with big laptops using their wee little 18W apple chargers. I presume they’re not getting an increased charge, but are slowing the drain.
I know some laptops with USB-C charging are unreasonably picky, but also that many USB-C chargers are garbage.
I have a USB-C dock that outputs 60 watts, even though my 65W brick is plugged into it. I get a BIOS message at power up claiming that an underpowered charger is connected. Still boots though but the charging rate is poor.
I say this with the usual caveats about situations where the vendor hasn’t specifically assured me that it’s a supported configuration and I’ve had the chance to personally test for lies and unexpected quirks; but your mention of the dock/BIOS message reminds me of some behavior I’ve encountered with recent Lenovo docks:
We mostly use the 40AH dock(which is USB-C based; but uses the slightly oddball side connector rather than a single USB-C cable like the 40AS or a single Thunderbolt connection like the 40AN); and it comes with a 135w adapter because it can support some of the punchier models along with a bunch of peripherals pulling power.
I was prepping a system (T490) which ships with a 65w USB-C adapter and supports the 45w one; using a 40AH I had available just for convenience. Because I didn’t feel like mucking around with the power strip I plugged the 40AH into a 90w adapter rather than shuffling things around to get its adapter plugged in and got similar BIOS complaints about an underpowered situation on every boot, despite the fact that there was nothing plugged in to the dock except the one fairly low power computer(with a fully charged battery, so no extra load from that). My understanding is that the dock is capable of generating complaints on its own initiative and (at least when it’s a Lenovo dock and a Lenovo system with firmware that understands and heeds them) getting them delivered on every boot; even if there is no problem.
Obviously, especially with the wonderful world of USB-C’s myriad variations, the only way to know for sure is to test; and you can take the word of a fungus on the internet for what you will; but I’d suspect that a direct connection of a T480 to a 61w USB charger(assuming all the ducks are in a row with the cable being used) will work fine; but I know for a fact that some docks will complain bitterly unless they are given an adapter of the power they expect; and depending on how closely they collaborate with the computer can sometimes make their complaints known via boot time messages. I wouldn’t generally try to replace a dock brick with an underspecced one; both because of that and because dock bricks are typically the ones that you don’t pick up and carry with you all that often; while direct power/charge bricks are size and weight critical.
It’s annoying; but arguably fair: a laptop has a battery and various sophisticated power management options at its disposal, so it is well equipped to degrade gracefully if underpowered, all the way down to ‘still draining the battery, just more slowly’ and (unless the vendor is a dick) should still be able to at least charge the battery while turned off from even a fairly feeble adapter, just not fast.
A dock, by contrast, has nothing to fall back on; and if it browns out you get frankly unacceptable things like connected monitors going on the blink, peripherals connected through the dock falling off the bus; potentially docked computers reverting to battery mode power settings rather than AC mode ones; so one can at least understand why they would want a brick large enough that you won’t encounter unexpected brownout behavior even if you plug in a bunch of high drain devices at the same time your computer tries to top up its battery and do some 3d rendering.
I speak as one who hasn’t had much opportunity to verify the behavior of 3rd party adapters(I don’t yet have a USB-C era laptop myself; and work provides me with ample test subjects but is one of those ‘we order equipment from the vendor according to the guidance of the Accessories and Options Compatibility Matrix(warning: the fact that this documentation is available is part of what, along with the keyboards, trackpoints, and PSREFs, makes me all warm and fuzzy about Thinkpads; but it’s a gigantic XLSX file); not piece it together from 3rd party bits unless there’s no alternative; then we validate it personally’ type environments.)
This topic was automatically closed after 5 days. New replies are no longer allowed.
