Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
MacBook Pro Keyboard Failures [Teardown and Explanations] (reddit.com)
27 points by vanilla_nut on May 2, 2019 | hide | past | favorite | 13 comments


It's a great teardown and commentary. I found this presentation easier to follow, as it has inline images.

https://imgur.com/gallery/G1EcegS

I've always been a "Save $$ by buying used Macs" guy, but I'd be very hesitant to purchase a used MacBook with this terrible keyboard . . .


Thanks for linking the imgur album -- I didn't realize there was a version with inline text! Still, I prefer the linked images, since I have a very hard time reading text in between moving gifs (needless to say, I would not be a subscriber to the Daily Prophet). Glad to have both here so visitors can read either version they prefer.

I've also been very much a fan of Macs because they tend to be very well-built and last a long time for the money. Both of my last macbooks (a 2009 and then a 2015) have served me very well for much longer than your average laptop tends to last if you use it every day and take it with you everywhere. Unfortunately this keyboard in the current version is a dealbreaker: the frustration, cost, and potential of having to lose my laptop for a week+ or just lug an external keyboard around is not worth the money. Add in the useless touchbar and my personal preference to not limit myself to usb-c/avoid low travel keyboards and you've got a perfect excuse to avoid upgrading.


++ on the touchbar being worthless. I use a touchbar MBP for work. It's managed to complicate the most frequently used functions (brightness + sound adjustments), and I experience bugs / temporary freezing of both of those sliders until I cmd + tab switch apps.


I absolutely love the writing style here. Very professional, and then this lovely tidbit "I found some household dust and threw it on the mesh.". I cannot help imagining the author walking to the parts bin, ruffling around for a while, and then picking up a small box with a "aha, there's my box of standard my household dust!" comment


I'd not be surprised if there was such a thing as "standard household dust". There's standard peanut butter after all: https://catalog.data.gov/dataset/srm-2387-peanut-butter


Why is nobody concerned about the absurdly low key travel distance? Dust, I'm sure they will find a way to mitigate. Please don't normalize the horrible keyboards, even ignoring dust issues. You'll be typing on touchpads before you know it.


It's hard to say if I actually mind the low travel distance. I find these keyboard very hard to type on because of the increased key size/small gaps between keys, so I constantly make typos and can't seem to orient my hands correctly without looking at the keyboard. Because of this, it's hard to look at the key travel issue in a vacuum.

Also, please read the article. This entire teardown is focused on refuting the "dust intrusion" hypothesis -- the writer actually seems to think that issues are rooted in malfunctioning metal domes under the keys linked to either wear, heat, humidity, or a combination or each. Basically there's no need to mitigate dust because it can't get into the dome mechanism even in the 2015-2017 keyboards before they installed extra dust guards, and part of the reason the 2018 still has issues is because, surprise surprise, the issue isn't just dust.


I do, I hated it since the 2015 Macbook and even more so as time passes. But between Keyboard reliability and low travel, the latter is not as important. I would like the Keyboard failure rate within 5 years of purchase to be less than 1%.

I really wish they just revert to the 2015 design. I have had multiple Post 2015 MBP and sold or returned, I am typing this on the last batch of 2015 MBP and its AppleCare will be running out soon.


Absolutely incredible effort by the guy who did the teardown.

And the extent that Apple went to just to protect tiny delicate mechanisms from dust is astonishing.


I wonder if the dust impermeability of the mechanism is not a part of the problem. While it would prevent most of the dust to get in, it could also prevent the dust from getting out.


Cool teardown, but I'd be interested to see a disassembly of a known-bad key to compare and see what looks different.


This is my personal opinion, after seeing this teardown. If you look at some of the images it looks like the actual contact areas on the PCB that the metal domes contact on are rather small, and both the dome and the contacts look like they're made out of a hard material.

0) The top side sealing on these keyboards (at least some of them; one of mine had a hole somewhere) isn't that good. One method I used to fix these keys is to dab some rubbing alcohol or ethanol into the dome thing with a qtip, and somehow it finds its way under the dome. The purpose of the black tape mentioned in the article is likely to direct the light from the backlight of the keyboard, it's easy to damage with this method of cleaning it and it causes the backlight to look bad. Solution: electrical tape carefully cut to size.

1) I feel like the debouncing (filtering the tendency for hard-on-hard contacts to make intermittent contact) is inherently difficult with a design like this. There's a lot of things to go wrong here, and even on a brand new keyboard I noticed that before the system boots (on EFI password, or FileVault2 Unlock Prompt) the keyboard is MUCH more sensitive to key repeats - Probably because half of the debouncing is done in the driver.

Perhaps if they could find out how to add a thin enough slice of carbon-filled rubber (like the black pads under the rubber remote control buttons in a telly remote) under the domes, combined with larger contact pads, the softness would allow the contact to conform to the mating surfaces and make more reliable contact. Furthermore it will absorb some shock and reduce the tendency for the switch to 'bounce'.

Furthermore, carbon filled rubber doesn't tend to corrode.

2) Perhaps they could move to capacitive sensing and do away with actually making contact. This way, there would be less issues with contact bouncing, and no current actually has to flow.Then all the electrics would be completely sealed off. This has the drawback of needing a complicated keyboard controller. Last I remember the keyboard still had logic gates hardwired to do things like SMC reset so this would be harder.

2b) Or just take the key matrix sheets from the 2012-2015 MacBook Pros, which were pressure activated by way of rubber protrusions under the rubber domes, and put the new butterfly keys on top and use a tiny metal disk a la iPhone 4s power button to press on the actual key matrix

Another oddity I noticed on a machine I recently had the keyboard changed at the Apple Store on is that the right Shift key (which just got stuck again), is very close to the left side of the aluminum, more so than the right, and isn't centered. It happens that the key is most stuck on the left, so perhaps the accidental interference fit is causing it to stick.


And after considering all of these potential fixes and mitigations for keyboard problems, any sane individual is forced to wonder: why on earth would anyone want to create an entirely new keyboard mechanism when so much can apparently go wrong? Really gives you respect for the folks who designed some of the original scissor switches when you realize that one of the biggest, most hardware-talented tech companies out there can't design anything better (or, perhaps to put it less subjectively... more reliable).

Surprisingly akin to the experienced software developer's reflex of trying to avoid total rewrites at all costs: even if something has issues, like scissor switches breaking when you try to remove keys or falling off under heavy usage, it's often easier to try to fix warts on something battle-tested than to improve core functionality of a brand-new form-over-function design.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: