Hardware fix for Dell “AC power adapter could not be determined”
blog.project-insanity.org
blog.project-insanity.org
The content is something like `DELL 01 AC 065 195 033 {23bytes} 01`, where 01 matches the revision number on the label of the PSU, 065 means it is a 65 watts adapter, 195 means the output voltage is 19.5V, 033 means the output current is maxed at 3.3A. The {23bytes} contains the country where it is produced, the Dell part number and date codes which I don't have enough data to understand how that is coded.
There was a github project which provided tools to copy and reprogram the chip, HP follows a similar protocol for its high end laptops, it is possible for a Dell laptop to use a HP PSU, if the BIOS is not implemented correctly.
I agree it should be an open standard, but the implementation is really similar to EDID. As we are moving forward with USB-C power supplies, this is probably going to be irrelevant very soon.
There are multiple charges with the same barrel jack, but different wattage. You plug a low wattage power adapter to a high powered laptop. Without the chip: The laptop can draw to much power and in the best case scenario it damages your power adapter, in worts case damage your laptop or burns down your house. With the chip: the laptop can throttle, so it draws less power.
USB Power Delivery does the same thing, just without the properitary protocol.
The best case scenario is that your charger is not a garbage fire hazard, and so it slowly drops the voltage when the laptop tries to draw too much.
Are they doing that trademark hack thing where they threaten 3rd party manufacturers who put that "DELL" string in their product without a license?
https://en.wikipedia.org/wiki/Sega_v._Accolade
I read it like Sega was on their way to losing the appeal, so they settled it out of court.
[1]: https://en.m.wikipedia.org/wiki/Psystar_Corporation#Legal_is...
Usn-c for charging, despite its warts, are quite the game-changer. Especially in these absurd times where batteries are non-replacable being able yo juice-up from a power bank is fantastic. (yes you could do that before but it was very niche and/or expensive and with less utility).
Looking around, stuff like this isn't uncommon. Apparently a lot of HP's won't work with non-HP USB C chargers at all, for instance.
I had a xps 15 with 130w charger, and used it with the dell usb-c/thunderbolt docks (wd15/tb16). The wd15 had a 130/180 watt power supply option, after subtracting for the dock's power usage they were specified to provide 90/130w to the laptop. I had the 130 watt wd15 and it still worked perfectly with the laptop. The laptop never discharged while connected (maybe if you spend hours compiling a big project on 8 cores while playing games with the nvidia card?). But I assume it would charge slower/not at all if under heavy load. And the bios warning for a lower power adapter can be disabled.
It should be capable of taking 130w over the USB c port; dell explicitly specifies that it should be used with the 180/240w wd15/tb16 docks which provide 130w to the laptop. XPS 15 9560 iirc.
On the MacBook Pro I'm typing on right now, I replaced my the battery myself a few months ago. It took about an hour including a thorough cleaning of the laptop internals and I documented it here https://www.youtube.com/watch?v=CAYFDDxUxHA . That's still 59.9 minutes longer than it would have taken if had a modular battery, but about the same price in dollars. And a fun little weekend project.
I'm not qualified to answer whether the ID pin constitutes DRM, and whether it is legal or not, but Dell's customers definitely were having issues with dodgy chargers before this was implemented. The implementation sucks though, as the design means it's easy to short the ID pin, rendering the charger useless.
Let me see if I could dig it up...
It is definitely functionally DRM since it prevent using other suitable chargers for no reason.
https://hclxing.wordpress.com/2014/02/06/hacking-the-dell-la...
https://hclxing.wordpress.com/2014/02/09/hacking-the-dell-la...
https://hclxing.wordpress.com/2014/02/09/hacking-a-dell-powe...
https://hclxing.wordpress.com/2014/02/26/hacking-a-dell-powe...
An one more for the Apple Magsafe one:
http://www.righto.com/2013/06/teardown-and-exploration-of-ma...
I don't even want to use illegit power supplies. Rather, I have a growing pile of genuine Dell ones (and jacks) that developed intermittent connections on that pin and no longer detect as authorized.
Anyways long story short: give contact spray a try. Worst case the contacts are deformed too badly and it wont work.
The plug wiggles and doesn't make a good connection.
Dell manufactures the whole spectrum of laptop quality, from cheap Walmart garbage to high-end mobile workstations. If you buy their consumer-grade stuff, yeah, it's often junk. But their business- and professional-oriented laptops are usually decent quality. Latitude and Precision, for example. I have both of those and they've been workhorses for me.
Mid-range Thinkpads would also be a fairly safe bet.
Having those chips run at 95+ degrees most of the time (when doing serious work for which the laptops were designed) and depending on Intel's built in thermal throttling (A. Last. Resort. Measure!) is ridiculous.
Dell has even been smartassing it with lower throttle thresholds, god forbid they use better cooling instead.
But hey, they save £1 on every heatsink, so F U, buy a new laptop when this one breaks!
Frankly I am sick of this. Just make proper heatsinks, I can personally come and shove a £2 coin up yours, thanks.
I don't know about you, but I prefer my laptop to stay cool enough to stay on my lap and be able to type on. Plastic is a good enough insulator to keep the heat away from me, and it's pretty easy to find laptops with adequate heatsinks.
Specced the heavens, max memory, ECC, Xeon.
I also had issues with the dock, but it was exclusively the dock, the TB16 is a flaming hunk of garbage that has no business being on my desk.
The issues mostly stem from the fact that the USB controller (which powers nearly everything in the dock) is shockingly terrible.
But, regardless, there are more issues because the 5520 (and probably 5530) have only 2PCIe lanes to the thunderbolt port. I was surprised to learn that "thunderbolt3" != "thunderbolt3" in all cases.
I have a 5k laptop that is a buggy pile of trash because someone tried to save a dollar on a USB controller.
Honestly if I could lug this thing at Michael Dells' head I would do so I'm so frustrated.
But anyway, the laptop by itself is fantastic, it does get toasty under full load but I tend to disable the GPU so it's bearable. I considered getting the new WD19 thunderbolt dock before I realised about the thunderbolt PCIe lanes.
Since then the one I have has been rock solid. I just gifted it to my kiddo b/c it's overkill for my current role. I will miss all of the jokes about my space heater though. Nothing will unify a group of 30 of my colleagues more than jeering me as I draw that beat up 17" coffee table from my backpack like some kind of nerd excalibur.
(Also more to the point I've experienced OP's issue a few times with it over the years.)
I have personally purchased three generations of the Thinkpad X1 Carbon over the years. I know the Thinkpad T4?? are popular also.
I would rather buy a new HP Elitebook or Dell Latitude than anything from Lenovo at this point.
But yes, they support Windows well enough.
There's also the whole Superfish incident, injecting malware to the OS from the BIOS is scary- and while it did not affect the Thinkpad line specifically I am increasingly wary of the company because of it.
And since the hardware is so hard to service these days, and (soldered components, plastic clips) and it isn't very special either there's not enough "appeal" to suade my opinion in their favour.
iFixit regularly rate those devices 10: https://www.ifixit.com/laptop-repairability
1170 Fan assembly
For access, remove these FRUs in order:
* “1010 Battery pack” on page 67
* “1020 ExpressCard blank bezel” on page 68
* “1050 DIMM slot cover” on page 73
* “1070 PCI Express Mini Card for wireless WAN” on page 75
* “1080 Keyboard” on page 77
* “1110 PCI Express Mini Card for wireless LAN” on page 85
* “1120 Keyboard bezel assembly, FPC cable, and Bluethooth daughter card” on page 87
* “1150 Speaker assembly” on page 95
* “1160 LCD unit” on page 97
And by the way, the "good thinkpad" series (ie. letter + two numbers) were both made and designed by Acer/Wistron and significantly mechanically simpler than the previous IBM designed thinkpads (eg. if you disassemble "low-cost" 600E, you end up with ~10 random-ish machined aluminium brackets and how to disassemble the thing is highly non-obvious).
https://www.dell.com/community/Latitude/Major-Latitude-7390-...
https://www.reddit.com/r/sysadmin/comments/8wbhhv/high_failu...
https://www.reddit.com/r/Dell/comments/ar1mas/the_latitude_7...
Also they are the only manufacturer so far IIRC which include a 'Pre-boot DMA Protection' option in the BIOS which helps mitigate the Thunderbolt 3 DMA attack
I only use it for travel and watching movies. There as a short "brain freeze" when plugging it in, but it continues to work normally.
The hardest part was finding a USB charger that:
- has decent wattage
- is not expensive
- also charges non-c USB
- does not explode (and bring down the whole trains outlets...)
- throttles/splits current (instead of running hot, shutting down and leaving your equipment uncharged for the night...)
I settled on this one: https://www.amazon.de/gp/product/B07PR8LDGL/ref=ppx_yo_dt_b_...
Only downside: no qualcom quickcharge 3.0 for the non-c USBs
On windows you can use CPU-Z, on linux look up for some command line solution.
Didn't do heavy load testing, but see the same ~2800mhz on AC/battery/usb-c (it's an i5-6300HQ)
I don't buy dell laptops for this exact hardware flaw.
I am willing to accept that cables can get damaged, but I don't think a PSU outright failing is something we should accept as "normal". Unlike cables, their longevity shouldn't be related to physical handling.
There was an engineer who showed how shoddy many phone chargers are and how dangerous they could be.
edit: Yes, of course some of the vendors will be gone or judgment proof. So don't buy from disreputable vendors. Which brings us back to the original problem -- if the user is willing to buy from them, they could just as easily be buying an entire laptop or a toaster for their kitchen. So solve it the way you solve those instead of stamping out competition, because you need to solve those anyway.
Or put it another way: If OEMs really wanted to solve this, they would stop overcharging so much for chargers and battery replacements so that people wouldn't have to play the Chinese equipment lottery to avoid paying a 1000% markup.
You may not get the chance. Some damage is irreversible and a lawsuit will probably only bring you some legal expenses. Preventing disasters is smarter and cheaper than inviting them and then clogging up a courtroom to complain that the perfectly preventable disaster actually happened after you invited it.
I recently bought a pair of Taotronics headphones from Amazon. They are a Chinese knockoff of more premium ANC noise cancelling headphones, at a quarter of the price.
The performance is incredible, and if you slapped a Bose brand on it, I could absolutely believe it's the real thing.
I would like to buy more "Taotronics-like" products from "reckless" marketplaces. For example, I'm looking at NASes. The top brand names like Synology are selling ~$40 worth of components BoM for $400. That's ridiculous!
Jeff Bezos is often quoted as 'Your margin is my opportunity'. As a consumer, I LOVE it.
Private sales are different than commercial sales.
I'd argue that any marketplace that allows other commercial sellers on its platform should be held accountable for anything safety and tax related their sellers do.
They should take necessary precautions to verify the sellers and routinely check the products. In Germany VAT dodging by Chinese companies selling on Amazon (let Amazon deliver with FBA and just don't pay taxes afterwards) got recently regulated so that Amazon is liable and the amount of Chinese crap dropped significantly.
[0] http://www.righto.com/2014/05/a-look-inside-ipad-chargers-pr...
Fake chargers and cables are very convincing.
How could it possibly be “your choice”?
That's also true of a million other categories of dangerous equipment, and yet almost none of them lock out competitors by using proprietary interfaces. A shoddy saw blade could kill me, too, but I still can go buy any blade from any manufacturer and put it in any table saw, and it'll work.
OTOH, there are gratuitous incompatibilities galore as soon as you get into software. Network protocols, file formats, executable formats, and so on.
What could explain the situations where we see compatibility versus incompatibility? If it's obvious and visible, and Average Joe can make it fit (like with a hammer), they go ahead and make it compatible. If it's secret and internal and they can hide the differences in ways that Average Joe can't work around (like software), they feel free to make it as incompatible as possible.
The main reason for this is so that different wattage power supplies can be used with the same connector. The laptop determines the power supply type to see if it can safely draw its full load.
USB-C takes this one step further by having a full negotiation process before supplying any of a number of voltage / wattage levels.
The power supply is "constant current, constant voltage" - ie. it is 19 volts for all currents under 3 amps, but if you try to exceed 3 amps, it will supply exactly 3 amps at whatever voltage that works out to.
The benefit of this scheme is it's super easy to implement on both sides, and fully universal - the device just keeps increasing the current draw till the voltage starts to drop, and it then knows it's hit the limit. The device doesn't need any kind of logic at all - a dumb light bulb will either work if there is enough power, or not work if there isn't, but nothing will get damaged.
It's also compatible with Y splitters - so you can plug in two laptops to the supply, with no extra electronics, and both can share the power. (there is no guarantee of even power splitting tho!)
A big benefit is the power supply can have fully variable capabilities - for example, most power supplies are thermal limited, and on a cold day it could offer slightly more current to charge your battery slightly faster - all without any digital logic, protocols, or anything manufacturer specific.
Anything more complex with sense resistors, ID chips, extra wires, etc. just costs more, and provides less utility, and almost looks like a deliberate attempt to reduce compatibility and increase sales of chargers.
I suppose the key question is: Does it identify the device and then decide not to work with non-Dell devices? It's fine to read required information, but unnecessarily constraining what happens based on that information brings it back to DRM.
The resistor setup is both simpler, and unlike the memory chip, can't easily be repurposed for DRM.
With USB type-c chargers it became more or less redundant although the USB type-c power spec implementation is all over the place with many cables and chargers on the market capable of ruining your devices completely.
This isn’t as much of a DRM as a reliability measure if you are going to use cheap power adapters you are quite likely to be able to ruin your laptop.
There are plenty of 3rd party adapters that can send the correct signal to a Dell laptop, those who can’t usually will power the system but not charge the battery simultaneously.
With USB type-C many adapters would charge the laptop at 60W instead of the 100-130W that original Dell adapters can, some of the certified USB type-c docking stations/adapters would be able to charge it at full rate if they are actually capable of delivering the current required.
Not everything is done because vendors want to lock you in, laptops now are smaller and power hungrier than ever and people want to charge their laptops at full rate with cables that could barely charge a phone.
With USB-C things will mostly work but will work slower I can charge my MBP 15” with my Nintendo Switch power adapter it will just take longer and likely will cause the Switch adapter to die the few times I had to do it I’ve noticed the adapter cosplaying as a heating plate while charging the Mac.