The secret Thinkpad powerbutton code to bring dead laptops back to life (2007) | Hacker News Reader

The secret Thinkpad powerbutton code to bring dead laptops back to life (2007)

masnick.com

210 pointsWayToDoor76 comments

I just tried this on a laptop that died 5 years ago. Same symptoms - one day it got really hot and then was dead dead dead.

Spent hours disassembling it and probing various test points, all to no luck. It seems the issue was somewhere before powering up the CPU, and therefore kinda out of my expertise.

But this method seems to have brought it back to life.

Shame I've now cannibalized the screen and trackpad and CPU heatsink :-(

Any static charge would have dissipated after a couple months at most. Laptops don't have that much capacitance, let alone shielded so perfectly to hold charges for long periods of time.

Even without the trick it would come back to life after a few hours at most. There is no way the static charges would just stay.

> But this method seems to have brought it back to life.

Did you try just turning it on before trying this method?

I wonder this too. If so, and just turning it on didn't work after the 5 years, then the static discharge explanation for this button sequence sounds bogus.

I was under the impression that pressing the power button while unplugged was to discharge any electricity left in capacitors, to make sure everything was completely depowered

That's precisely what was described, wasn't it?

"I asked the guy what the power button pushing incantation did and he said “static discharge” so apparently there was some sort of static that caused a short or something."

Not really, no

So the tech (or the person writing this up) mixed up "static" and "flea power"[1].

I believe the idea of "draining charge that shouldn't be there" has been properly conveyed nevertheless.

[1] https://en.everybodywiki.com/Flea_Power

I have, in several decades of operating and servicing computers and other electronic equipment, never heard of the term "flea power". According to that article it seems to have originated in a Dell service manual, which erroneously calls it "static electricity".

Is it incorrect to call it static electricity, though.

I mean... it's charge at rest... so electric and static, but in that same sense, there's static electricity in batteries, and static electricity in DRAM (in-between refresh cycles), and any oscillator at the moment when the first time-derivative of charge reaches zero.

I submit that a definition of static electricity as "charge at rest at non-ground potential" is so broad as to exist in any non-trivial circuit and many trivial circuits. Such a definition is therefore less useful than more common notion as electricity generated as a result of physical surface interactions of dissimilar materials.

Calling it "static electricity" is likely to cause reasonably intelligent and well-educated people to conclude it's related to wearing wool socks in Winter while being insufficiently grounded, as opposed to charge being there as a result of normal circuit operation.

Is it really, though -- isn't what happens inside capacitors basically the same thing as we call "static electricity" when it happens between other surfaces? Or, IOW, isn't pretty much anything potentially (Heh!) a capacitor, only pretty much all things except the ones purpose-built as such are really bad capacitors?

So what do you call it?

Looks like there's no standard term that everyone uses; I'd say it's entirely forgivable that the tech called it "static" for lack of a better word.

I've always referred to it as "residual" power/charge.

It's difficult to imagine how the underlying fault here is anything other than a latch-up[1] event, likely manifesting on a upstream controller/supervisory chip in the power distribution chain.

The description of this "flea power" reads like some cargo cult narrative that summarily conflates a handful of disjoint electrical characteristics/phenomena. It's like the modus operandi of interwebs consumerism.

[1] https://en.wikipedia.org/wiki/Latch-up

The article makes it sound like it's a secret code like a Konami code that the laptop will recognize, but it's not the case.

Yes, this is not an uncommon technique for doing a "true hard reset" on electronics that are normally "soft-off". The idea is that pressing the button will cause it to try to turn on if there's still some charge left, which then drains the charge away quickly.

My really old Lenovo Thinkpad started insisting to reboot after shutdown a while ago, and after a hard drop that left the battery dislodged, it froze after being powered on.

I had to force it to power off, but that apparently broke it, it won't light up the screen and respond to inputs despite blowing the fan. I disassembled the screen and replugged the data cable, pressed power button in various ways to no avail.

I was about to give up before reading a forum post describing a similar laptop, and emptyed the RAM slots and found it beeped when powered up, so the CPU is OK.

I then unplugged the battery and CMOS battery and pressed power button for a while, put RAMs and batteries back on, and it just revived.

Removing RAMs made the difference, it's done if RAMs were soldered on.

It happened again a few days ago after it was left with no power, and I had to do those thing again, otherwise it has no issues. I think the CMOS battery is probably dead.

This sort of trick will fix a lot of annoying little issues on most PCs, like if you have a USB device that you know works but your PC won't recognize it. Shutdown, unplug the power, then hit the power button on your PC a few times. Everything should be good. Works every single time I need it.

Once in a while after a resolution switch (e.g. exiting a game), one of my displays will go into its sleep mode and won't come back. Unplugging it and hitting the power button a few times fixes this, no PC restart needed. My friend had a TV that'd do something similar once in a while and this fix worked for him as well.

I'd be willing to bet the specific sequence in this article isn't really important, but rather that you're draining every capacitor in the PC by hitting the power button without a power source available (my understanding of what's happening with my fix).

Modern equipment usually has a power controller chip that sits between the power button and the actual "power on" signal that's sent to the power supply. There's no direct physical connection between the button and the PSU anymore.

The one I usually work with (LTC2954) does have an RC circuit attached to it to determine how long to hold the button to force a shutdown. And I'd wager a guess that this is the capacitor you're discharging by kicking the controller a few dozen times. But there's no possible way I can see that you're discharging every cap in the system.

This reminds me of an old ViewSonic LCD monitor I have. It occasionally corrupts its EDID memory for some reason, which causes the computer to not recognize the monitor anymore.

The procedure to restore it sounds similarly weird. You plug and unplug video connectors, the monitor power and boot/shutdown the computer in a certain sequence. I didn't believe that worked when I first read about it. I've done it several times over the years however and it did in fact restore the EDID every single time.

https://www.tablix.org/~avian/blog/archives/2010/06/the_curi...

I had something similar happen, also with an old ViewSonic!

Unplug it, press the power button, watch as the LED lights up briefly (before the monitor discharges), then plug it all back in and it would work.

I had a similar issue with my Apple Watch.

One day the screen went black and stayed that way. Hard reset did nothing. The screen didn't turn on when charging, but the watch got warm. Apple tech support said it was a hardware problem and I would need them to service the watch.

This is a 1st generation watch though, and their nearly $200 service fee seemed excessive. So I just let the watch sit around for a week or so. Then I tried charging it again and it worked!

What must have happened was that it was either a really bad software issue of some kind or an intermittent hardware issue that prevented the watch from working or rebooting, but when the power gave completely out and it was recharged, everything was set back into a functioning state.

The watch has worked perfectly (or as well as it ever did) for weeks now.

I had the same thing happen with an old iPod touch or iPhone a couple of years ago that I had bought second hand and which had worked as it should until one day the screen went black. I let it sit for a day or two but when it still wouldn’t respond I got angry at it and threw it in the freezer in order to make it turn off sooner. This might not be a great idea, but the next day I took it out and let it sit for a little while and then put it on charging and the screen came back on when it turned on and it worked as it should after that.

Soooo good!

I copy the "trick" here, to have it available, in case that page disappears:

"push the power button 10 times in a row at one second intervals. Next, you push and hold the power button for 30 seconds. Then you put the battery back in and push the power button… and she lives. The computer came back, good as ever"

Your mind is gonna be blown when you discover the Internet Archive's Wayback Machine!

https://web.archive.org/web/20080310221514/http://www.masnic...

Domain parking often vandalises the Archive record for a domain with their robots.txt. It's far better that it's recorded in multiple places plus archive.org

does robots.txt apply retroactively though? That would be bad..

At one point, it did. I'm not sure about now.

It was archive.org's attempt at automating "please remove X from the archive" but I don't think it was entirely successful and I think they may have stopped doing it.

The idea was if you decided you didn't want /cats.html archived you'd robot.txt it, archive.org would notice, and delete the archive; but it also meant if the domain changes and robots.txt just blocked everything, everything would be lost.

Amazing that in the comments (last one dated 2015) that this was still the case up to (and perhaps exceeding?) the t410s. That's a fairly recent device and wouldn't surprise me if this secret discharge incantation was still on the books in later models as well.

Is this specific order a thinkpad related order, or is this just a generalized discharge reset? I realize that its fairly common to need to discharge, but is the order (10 presses here, hold for X amount of seconds there) important?

I doubt its anything thinkpad specific, unless someone can find a datasheet for the parts involved.

This sounds like a sequence designed to get people to wait long enough until the computer is fully drained of power, causing any latch-ups to clear. Sometimes a tech (or their troubleshooting tree) will have you unplug/debattery the computer, then read out serial numbers or something to accomplish this task without you noticing. I bet if you just removed the power sources, then waited a minute timed by a stopwatch it would work just as well.

There is a art to making troubleshooting steps resolve a problem without the customer needing to tell the technician they made a bone headed mistake. Really well written steps like these ones make it so the customer is completely unaware that the actual problem was that they did not wait long enough.

(Rewritten from memory)

The apprentice noticed his machine was malfunctioning. He consulted his sensei who suggested to simply turn the machine off and back on.

The apprentice tried this many times yet the machine still did not work correctly.

Then the sensei turned the machine off and back on. It worked.

> A novice was trying to fix a broken Lisp machine by turning the power off and on.
> Knight, seeing what the student was doing, spoke sternly: “You cannot fix a machine by just power-cycling it with no understanding of what is going wrong.”
> Knight turned the machine off and on.

> The machine worked.

http://www.catb.org/~esr/jargon/html/koans.html

This is also the reasoning for blowing on a cartridge/connector

On newer Thinkpads with a built-in battery that you can't take out, you also normally have a hidden reset button behind a small hole on the bottom, which you can push with a paper clip. This "reset" button will soft-disconnect the battery (which might then require you to plug in the AC-adapter to bring it back to life).

It's unfortunate this isn't mentioned in the repair manual -- that simply mentions 'go into the bios and use that to soft-disconnect the battery'. Which is awfully difficult if you're disassembling the computer to repair the screen :-(

But at least I now know this trick for the next time I need to do something like that. Thanks!

From the electronics designer perspective: this sounds very much like resetting the KBC by simply forcing any and all capacitors around it to discharge. KBC is a microcontroller, with its own firmware, and it 100% can lock up.

Keyboard controller?

Kentucky Baked Chicken, duh...

Sometimes these magical sequences are really just a set of hoops you send your users through so they will actually do what you want them to do (unplug machine for 60 seconds, plug back in). Otherwise they'll half-ass it and miss the simple fix.

Exactly - I suspect that often "unplug, wait X seconds, hit power, plug in" would do most of it but the "wait X seconds" part is what people have trouble actually doing, so you add the busy work.

The ones where weird sequences do something are with power still plugged in (the chipset can react differently when it notices the power button is held down as it turns on, etc).

Having worked phone support for a dial-up ISP in high school, I can confirm that tons of people will assume and skip steps for any number of reasons - faulty mental models, misunderstanding the reasons we wanted to check things, who knows.

If there was a "no dial tone" call the first thing I needed to check was obviously if the phone cord was plugged in. And I had a few of these calls that ended the same way: me several pages deep in the troubleshooting tree before the customer sheepishly admitting that the phone cord wasn't plugged in.

I asked my coworkers and they all had their own tricks. One liked to have the caller unplug/plug the wire back in at the modem while they looked for a serial number printed by the jack. Another knew the Bell system pinout order and would have customers unplug and "check the wires" on both ends to be sure they "had the right wire". No one cared what the results of those questions were – of course the cord you bought from Radio Shack was wired up correctly. But you checking means you have to unplug and plug back in and that makes you discover that "oops, when I moved the filing cabinet I must've ripped the plug out of the wall". Or "oops, the pin on the back of the modem broke off and I was too cheap to get a new one and it wasn't making contact". Or whatever.

Indeed I've been doing this to "dead" laptops for years. It discharges any charge left anywhere in the laptop, and resets the charging/power distribution stuff. Works very nicely, but usually is easier: Disconnect power, take out the battery, hold the power button for 30 seconds. Voila. Fixed.

I read your comment and in my head I said "This." but I hate when comments start with a solitary "This." so I have managed to avoid that through meta-commentary.

About a week ago my fairly new Clevo OEM/ODM (the actual badge on it doesn't really matter) laptop wouldn't come back to life from sleep – screen was black. The AC cable/power light and the battery light, both those lights were on meaning the machine wasn't dead dead. After waggling the mouse a bit, running my finger over the trackpad, pressing the power button a bunch of times, and striking random keys on the keyboard … I still had a dead laptop. So I did what any one of us would normally do in this situation. I held down the power button for a good ten seconds expecting the machine to power off, and from there I expected to boot the machine back up and have it complain to me that it had been unsummarily shut-off and could I please not do that.

No dice.

Tried this reliable technique a few times and still no dice. I could not believe that my newish laptop (about three months old) was dead as a door-nail. So unresponsive that not even holding down the power button until it switched itself off would work. Absolutely black-pilled and pretty determined to ask the manufacturer for a refund I rang up tech support fully expecting to be on hold for about half-an-hour but thankfully got through fairly quickly. I explained the situation to the techie and was surprised when they said, “Disconnect power, take out the battery, hold the power button for 30 seconds.” I've been around computers a long time and I'd never hear this one. Extremely dubious I did what I was told and, as you say, “Voila. Fixed.” Machine sprung back to life. (After reinserting battery, reconnecting power, pressing power button – obviously.)

What. Sorcery. Is. This?

Sounds more like a PMU reset than anything else...it would be interesting to see if the 10 1-second button presses are really necessary.

My DELL has a press and hold the power button an arbitrary period of time to do something similar when it won't reboot.

I think those presses are to make sure the electricity is completely drained out of the laptop.

Jesus H Christ, finally there's an Easter Egg that's actually about resurrection instead of chocolate bunny rabbits and jelly beans and colored eggs.

The seven years worth of "omg, thanks!" comments is pretty cool. Too bad the "small internet" is dying. Had it been a facebook comment or old tweet, it likely would have melted into obscurity.

It's only perceived to be dying because of the Google hegemony directing all organic search to corporate blogspam.

https://millionshort.com/ is still a fairly decent tool for scouring the non-SEO'd web.

very good, Kirse!

And there I found this one https://forums.lenovo.com/t5/ThinkPad-T400-T500-and-newer-T/... which says, literally:

Remove the battery

Press power button 10 times

Press power button down for 30 seconds

Press power button twice

Attach ac adapter

Press power button once. PC should come up. Perform normal start up. Everything should be fine.

Shut down and reinstall the battery

From 2011 though, 4 years later...

This is pretty cool!

After looking at Millionshort, it's pretty sad to go back to Google and realize that it is unable (or unwilling) to move beyond simple ranking and group results into categories.

https://openwebring.com/

I'm guessing the tech's "static discharge" explanation would probably be more correctly described as discharging capacitors? I had a similar issue with an Asus u36sd laptop. I first realized it might the problem when I originally gave up on it(it already wasn't my primary daily driver, just a work bench laptop) and put in away for a bit and then it worked a few months later after being totally discharged and when the problem repeated, I was able to bring it back with a similar incantation of pressing and holding the power button with not power or battery attached. Luckily the capacitor causing the issue was a decent sized electrolytic of the through hole variety(even though it was surface mounted laying on its side hanging off the board) so it was relatively easy fix once I figured that out.

Reminds me of this masterpiece of IoT reset dystopia from a GE smart bulb: https://youtu.be/1BB6wj6RyKo

At least IKEA Trådfri is just "Power cycle 6 times".

This is super annoying with a small child who occasionally likes to flip the switch a bunch of times in a row. I have had to pair my lights way more times than reasonable for the benefit they provide.

https://www.amazon.com/Child-Proof-Light-Switch-Guard/dp/B00... or similar can help; but what I've taken to doing is turning off the power to the circuit, bypassing the switch in the box, and using either the phone or the "remote switches" mounted high.

That video does feel like a parody, but it makes sense that the reset process would have to be convoluted enough as to not trigger it accidentally.

I have a T420 that stopped powering on one day and it worked fine when I swapped the whole keyboard assembly for a new power button... maybe that old keyboard is actually still working

Shoutout from another T420 cult compound

Did a stint at Tech Guys (Part of DSG) and this was taught as what's known as an ATX-Reset, basically clears out any left over electrical charge iirc.

Had the same issue with my previous laptop, an HP Probook 450. A lot of vague problems were resolved by removing the battery and holding the power button for a few seconds.

I'm sure the underlying problem was some kind of hardware damage because I dropped that laptop so often that it's a miracle the thing still boots in the first place.

Tried this on a 7 month old AMD IdeaPad, which died suddenly during Win10 update. No luck.

Sent it back to Lenovo for warranty repair. Waiting on parts... estimate is 4 months!

Not sure if right-to-repair could have helped me get it fixed faster locally.

In normal times, local repair shops would be faster. But right now, there simply are no parts. There are no replacement computer parts, there are no parts to build bikes and even tents have shortages

This also works on Dell rack servers to reset cmos/drain the caps and reset the drac etc. well you only hold power with no ac plugged in for 30 but same thing.

Have done this a few times.

Does the incantation itself do anything or is it just a ritualization of repeatedly pressing the power button to remove all the charge from the system

Sounds the same as what Dell call a flea power drain.

Yeah I've had desktops with this problem. It's not a magical sequence, you just connect the power circuit and let everything drain.

That Thinkpad is probably still working today. Typing this from my trusty old T420.

I’ve had to do this multiple times on my thinkpads!

Also works on ideapad, in case anyone is wondering

i could have seen this before i tossed my t410. (yea, i'm kicking myself)

Reply on news.ycombinator.com