Framework Laptop 16 Deep Dive – 180W Power Adapter
frame.workTwo week old dupe: https://news.ycombinator.com/item?id=36246309
> Beyond reducing environmental impact through efficiency and recycled materials, we’ve enabled longevity by making both the 2m USB-C and 1m AC cables removable, letting you swap a cable if your cat chews through it.
It baffled me when I first actually examined a laptop that used USB-C natively to discover that its power adapter didn’t use a removable USB-C cable. I’d just assumed implicitly that they would, because obviously you would do it that way since the cables are so frail and always fail long before the brick, and lots of people would like to be able to use a different-length cable.
So far I haven’t had an A–C or C–C cable last for even six months, with not particularly stressful usage. Nor have I had an A–Micro-B cable last much beyond a year in the longest case (normally much less). In all cases, they become unreliable. I’ve tried quite a few different brands by this point, mostly on the cheaper side of things but not all, and have observed no significant difference between cheap and expensive. They all fail.
Does anyone make cables that don’t fall apart ridiculously quickly, failing at the cable/plug junction or within the Micro-B plug?
DC barrel jacks, they fail eventually too, but they’ve at least tended to last two to four years, which is slightly less absurd. My current ASUS laptop’s barrel jack became unreliable after a year and a half in a way novel to me: the metal of the jack has been etched by the latching mechanism in the laptop, to the point that it no longer works if sitting forwards or backwards, only up or down (the less common, and thus less etched, orientations for me), or if you unseat its latching by pull it out a fraction of a millimetre.
On the contrary I’ve never had a usb cable fail short of physical damage (from a cat).
What are you doing with your cables?
On the topic of connectors, I still think Apple’s lightning connector is the most robust connection the industry has enjoyed in recent times, and most robust for its size ever. I’m sad that iPhones are going the way of USB-C.
The lightning cable being sealed is amazing. The fact that there’s no cavity to trap dust on the cable is amazing.
There’s still the dust catching port on the iPhone itself but avoiding the cable having a dust cavity is a huge win.
The only cables that failed for me were lightning that I had in my car for emergency charging.
> What are you doing with your cables?
Plugging and unplugging things up to a few times per day. Occasionally using a phone while plugged in. Using a laptop in a lap. These last two will occasionally produce mild lateral stress. The cable will occasionally (like, every few weeks at most) experience mild tension due to moving a device while the cable is mildly snagged, and on rare occasions they have experienced more extreme jerking tension (and this was likely a factor in my most recent cable’s failure, but most of the cables that have failed have never experienced anything like this). As far as twisting: most of the cables have been used just on a desk or similar (other than when the connected device is being held), only rarely being moved (once every few months, at most), at which time it’s coiled with a radius of at least 15cm.
Seriously, these cables are not often being abused, and a couple of them have even absolutely always been treated very gently (and accordingly they’ve failed in the Micro-B connector), and I’ve heard similar stories from others. No one’s laptop power cables last more than five years, that I’ve heard. The slight flexing that you get from completely normal use is enough to do them all in.
As for Micro-B connectors, they’re supposed to be rated for 10,000 insert/remove cycles, but I’ve never had one last even 2,000 cycles before starting to become unreliable in the connector (… if it hasn’t already become unreliable in the cable/connector junction), even for a cable that was always used in ideal circumstances (gently plugged in, resting on a table). Accordingly, when I read cycle rating numbers for hardware, I start by dividing by ten, then maybe it’s reasonable.
I suppose I should also mention that I do have two very short A–C and A–Micro-B charging cables that are generally used less than once a week which have lasted for 2–2½ years so far and are still fine. But that doesn’t count since it’s not anything like daily usage.
There's just huge variation in how people use and treat their stuff. The only cables I have ever had fail are: 1) Cat 5 cables that I made myself when I first started making them and was bad at it, 2) a cable on headphones that I tried to yank while the cable was stuck in a chair caster, and 3) an old Apple Lightning cable that lived in my backpack and failed after something like 300 miles of backcountry trekking.
These are all reasonable failure cases in my opinion: 1) incompetently homemade cables, 2) an individual traumatic incident, and 3) extended use in harsh outdoor conditions. Apart from that, I've never had any cables fail, even the notoriously frail Apple cables.
> Does anyone make cables that don’t fall apart ridiculously quickly, failing at the cable/plug junction?
I always hear that Apple cables are prone to falling apart but I’ve had USB-C chargers for my laptop since 2017, carried them around everywhere, didn’t baby them, and they’re mostly fine. They’re kinda yellow and brown now instead of white but that’s the worst of it. Genuinely not sure if I’m doing something different or if I’ve been lucky or what.
It’s largely based on usage and manufacturer. I have 5 year old iPhone/lightning cables, but I also had a cheap cable die in the first week.
USB-C has similar ranges of build quality, but you also want to avoid picking your phone up by the cable etc.
At least some USB-C cables aren't very good. The USB-C charging cable I got with my Dell laptop failed due to inadequate strain relief (rubber sheath failed right where it met the plug.)
The Magsafe cables are the ones I've had problems with every time. I read that a decision to remove PVC from the plastic was what made them so fragile.
> with not particularly stressful usage
You are doing something to your cables. Even my cheapest of cheap micro b cables have lasted 6ish years. I don't think I've ever had a premium cable fail on me.
Are you overly bending them or coiling them too tightly?
Braided USB cables are robust in my experience, you might try those. With respect to DC barrel jacks, it's a mixed bag. My old ASUS Eeepc had the barrel jack get loose and make poor connections after about a year or two and I've heard others say they had the same sort of problem. On the other hand I've never seen nor heard of a thinkpad's DC barrel jack failing in any way. Those have a tight snap fit even after 15+ years of rough use.
I hate to blame the user, but if all cables are failing then maybe you're being too rough with them? I hate to suggest that because I've had cables fail despite receiving no abuse; Apple's old magsafe cables were notorious for this and the fanboys even more notorious for always blaming the user. It's possible you've had bad luck with cables and weren't mistreating any of them. It has been my experience that some cables deteriorate despite not being abused, but other cables put up with a lot of abuse.
A common failure mode is charging the device on the bed. The bed pushes up against the connector and stresses it over time.
Absolutely. I believe I've lost an audio port to this effect before; the plug for my old headphones was large and long, giving it a lot of leverage when I left my laptop on a cushioned ottoman. The plug/cable itself was fine but that port became unreliable after a while and eventually failed completely. Lesson learned, I always put a hardcover coffee table book between laptops and cushions now.
I stand by my opinion that the rectangular ThinkPad power connector is the finest power connector ever made. I have never had one get loose or wear out in any way, nor do they flatten out if you accidentally step on them as has happened with some of my type c connectors.
I have several Anker 100 watt USB C cables scattered throughout my house and they've all lasted a over a year now with no signs of any issues.
Apple's cables work fine too but they of course eventually get worn down at the ends and eventually need to be replaced. I also tried Aukey, but their cables stopped working for me after a month or two. I had them replaced under warranty twice and all three times it stopped working within a month or two.
So at least for cables I use with higher wattage adapters I exclusively use Anker & Apple USB-C cables now.
IME Anker cables didn't work that well with frame.work throwing it into a power negotiation(?) loop. Basically udev was eating 100% of a cpu switching to and from battery if it were to discharge below 80%.
Apple macbook air cables have been rock solid though.
I just bought a high end Anker multi-port USB/PD brick and a collection of Nomad USB cables (C-C, A-C, and one that's C-C but includes C-A adaptors for both ends). To be fair, the only places I've consistently had issues with cables are the [shared] kitchen charging station the whole family uses every day (laptop, ipad, phone, watch, kindle) and in the car for Android Auto & Car Play. I've never had a cable fail on my desk. The new stuff I bought is to upgrade the brick in the kitchen with higher power delivery on multiple ports, and to replace a failing cable in my car. As a guy using a 16" MacBook Pro, having a multi-port adaptor that can adequately charge the laptop while also charging a phone is helpful.
> It baffled me when I first actually examined a laptop that used USB-C natively to discover that its power adapter didn’t use a removable USB-C cable. I’d just assumed implicitly that they would, because obviously you would do it that way since the cables are so frail and always fail long before the brick, and lots of people would like to be able to use a different-length cable.
They don't want to have to deal with customer support cases from people who tried to charge their laptop over a $5 cable from amazon.
I'm someone who's pretty hard on cables (I don't think an Apple cable has ever lasted more than 3-6 months of daily use in my hands, even magsafe, which was expensive back when they were soldered into the power bricks) but I've had good luck with C-C power cables, just using somewhat generic brands like Nimaso.
The cables that keep dying for me are the short, expensive, full speed thunderbolt cables for my dock. Those start glitching out once a year. But the 100W charging cables are fine.
Almost 7 years later, I am still using an Anker Powerline+ braided USB-C (C to C) 6' cable that I purchased for use with the original Pixel (1st gen) phone in 2016. It's the cable I use every night to charge my phone, and it comes with me when I travel. Never had an issue. Granted, I'm not using this in a challenging environment like i.e. in my car, but that's still pretty impressive durability with somewhere on the order of 2,400+ inserts.
I’m pretty hard on my laptops and have never had an apple charger break despite very heavy use for 1 year+.
I have had apple chargers fail in the second year of operation.
I rarely have cables fail, and don't really understand how others have this problem, beyond the kneejerk thought of user error.
I have a few Anker cables that seem to be fairly durable. The C2Cs and C2lightnings have lasted for years at this point. Whatever come with the Pixel phones are also okay. My A2Cs are all fairly old, too, mostly brand-less, but still working.
Buy name brand cables from places like Cablematters, Belkin, Monoprice, Anker, Club3D, etc.
really? I have had 0 USB-C cables die on me. And I have some cables for over 3 years now.
My current client-issued contracting laptop has a 230W power adapter. Runs cool and silent'ish during Laravel/TypeScript webdev. JetBrains IDE barely manages to spin the fans a bit noticeable before it goes back to smooth sailing. So a 180W adapter no longer causes me to react much.
But I bet it would run pretty hot and loud if I toyed around with stable diffusion and a bunch more containers.
And then there are aboslute units like the Razor Blade 18 with 330W power adapter: https://www.razer.com/gaming-laptops/razer-blade-18 (13th gen i9 with RTX 4090)
An M2 Pro draws around 20W with IntelliJ development, usually it hovers at around 12W. Gaming with GeForce Now keeps it at 12W as well. I am baffled by the need for over ten fold power adapters.
I don't think 13gen intel laptops draw much more than 40w for mundane development like ours. But I'd have to measure.
My laptop sports a RTX 3080. That's probably the main driver for such max power needs.
Obviously Geforce Now doesn't incur in more power draw as it delegates running games to the cloud.
How about a "Deep dive" into pricing already. My sinking suspicion is they're intentionally holding back pricing as long as possible, as revealing how expensive this thing is going to be would cause the wheels to fall off the hype-train somewhat.
That's not a deep dive. It doesn't even show the PCB.
It's not a "deep dive" of the power adapter - it's part of a "deep dive" series of the laptop model, writing an article each about individual components.
Yeah, I wouldn't call any of these "Deep dives". Perhaps they're catered to the audience that might be reading their blog, but I doubt that audience differs much from the HN audience.
Yeah, I'm all for the idea behind framework but there's very little actual technical details being shared other than controller ICs
LTT had a video recently where he went to their production facilities and show how they’re actually making the laptops. Was pretty informative, more informative than the information we’re getting from framework themselves.
Virtually all of the electrical energy put into a computer is turned into heat, right? Some of it comes out as visible light from the screen and some tiny portion of it is.. converted into information I guess? But generally a 180W computer is putting out very nearly 180W of thermal energy, as I understand it.
My point is that laptops with 90W adapters can get too hot to comfortably touch, I think at 180W you're well past the point where actually using it on your lap is advisable.
180W isn't as bad as it seems. I have an ASUS Zephyrus 14, which ships with a 180W charger (and has a full Ryzen 9 5900HS + GeForce RTX 3060, 40GB of RAM) -- under normal use, it doesn't get particularly hot at all. (it never draws more than ~60ish watts)
If you stress-test everything to the max (Turbo Mode CPU+GPU, max power draw, while playing a game at 120fps) then yes, you can draw 150+ watts and it's too hot to safely have on your lap. But you can set it on a table and still use it comfortably. (The keyboard/trackpad never gets hot)
Notably, being able to choose which mode you want anytime is super useful. (I can turn this into a respectable gaming rig at a moment's notice, and it weighs less than a 16' MacBook Pro. I can also flip a switch and it's basically an ultrabook with 7 hours of battery life and silent fans -- or I can run it all day off a tiny 65watt USB-C charger in the office in that mode)
Having a high-watt charger means you have the option to use that electricity, not that you are required to do so.
> My point is that laptops with 90W adapters can get too hot to comfortably touch, I think at 180W you're well past the point where actually using it on your lap is advisable.
You're mixing up thermal energy and temperature. Consider that you can easily burn yourself on a 60W soldering iron, but it'd be quite hard to do so on a 1500W oil-filled space heater.
You could keep temperature down with fans, but my 90W thinkpad has fans so strong it sounds (and feels) like a hair drier and still runs very hot. I'm having a hard time imagining a 180W laptop that isn't hotter than hell.
You have a gap in your understanding of the physics here. Even a 400W laptop can be cool to the touch.
A laptop is hot to the touch because it has hot components (eg a processor running at full throttle) and fails to dissipate that heat correctly.
That is mostly divorced from it's power consumption.
If you ran a large robotic arm off your laptop, you'd be using a lot of watts but making only a bit of heat.
There's no gap in understanding, the only way a 400W laptop can be cool to the touch is if it rapidly dissipates all that thermal energy. That would require one hell of a fan, which is what I said with my hairdrier remark. Even 90W already requires a very noisy fan. A fan moving enough air to keep a 400W laptop cool to the touch isn't realistic to put into the form factor of a laptop, and a fan sufficient to keep even a 180W laptop cool seems extremely dubious to me. It would sound like a vacuum cleaner.
What I was missing is the part where the laptop doesn't consume 180W and a substantial fraction of that power is just passing through the laptop to peripherals.
Case in point, I’ve driven my desktop computer up to 600 watts and mostly it just puts out a warm stream of air silently.
You’re missing another variable here. In your case it may just be the thermal interface material (TIM) isn’t doing a good job of transferring the heat from the processor onto the heat sinks that the fan blows over.
I’ve had temperature drops of over 15 degrees c by opting for Liquid Metal, everything else being same.
Additionally, loudness is not necessarily correlated with actually moving air. It can often correlate with obstruction to moving air.
You can equip this thing with 4 Thunderbolt ports. Each such port is required to provide 15W to connected accessories, and they have to plan for the worst-case current, which inclusive of the greediest CPU and GPU you can configure this diving board with adds up to a huge (worst-case) power draw.
Also, if the user has a dead battery,they want to be able to plug in and charge, ideally while also doing some reasonably heavy computing. So you again need more power than the laptop itself is using.
This is the one case (AFAIK) where input power isn't being converted to some kind of heat: it's being stored in the battery as energy for latter.
My understanding is that most Framework 16 laptops won't draw anything close to 180W, and only the people who put beefy discrete graphics cards in the expansion bay will have that kind of draw, and those graphics modules will have their own fans to reject heat.
This is true, it would take a fantastic cooling system to make burning 180W on your lap comfortable. However that doesn't make a 180W laptop useless. Firstly if you are charging it when not on your lap you may prefer the faster charging and the heat doesn't matter much if it is just sitting on a table. (Plus more than half of the energy is going into the batter rather than being expelled as heat). Secondly if you are at a desk then it is nice to have the ability to do demanding tasks, even if a lot of your use is well under 180W having the ability to set it down and crunch some data (like letting a video render for a few hours unattended or playing a game with keyboard and mouse) is very useful. Thirdly short bursts of 180W isn't uncomfortable, so it can be useful to have this capacity even if your average is well under 180W.
I would hope that the power management settings have options to limit the power usage/heat output that can be enabled when using it on your lap.
TL;DR yes, burning a constant 180W on your lap isn't good, but having this capability adds many valuable use cases.
Total peak power envelope does not mean it is all being dumped into core components.
If your processor only maxes up to 40w, it doesn't matter if the laptop supplies 200w or 1000w. It won't run any hotter.
So is apple just gonna buy this company or what?
Doubtful, the philosophy of the company goes against Apple's way of doing things.
Apple likes sealed products that obsolesce in place. Framework likes the computer of Theseus approach.
Apple will have to start re-considering their approach though.. the EU has now mandated that phones need to start introducing replaceable batteries, etc.
There has been far too much e-waste. Why should one have to purchase an entirely new phone just because the battery wore out?
why would apple buy them?