EEVblog
ฝัง
- เผยแพร่เมื่อ 28 ก.ย. 2024
- By popular demand, Dave tears down the dodgy tactile switch from his aircon, and compared with a couple of others.
Forum: www.eevblog.co...
Subscribe on Library: lbry.tv/@eevbl...
EEVblog Web Site: www.eevblog.com
The 2nd EEVblog Channel: / eevblog2
EEVdiscover: / eevdiscover
Support the EEVblog through Patreon! / eevblog
AliExpress Affiliate: s.click.aliexpr...
Buy anything through that link and Dave gets a commission at no cost to you.
Donate With Bitcoin & Other Crypto Currencies!
www.eevblog.co...
T-Shirts: teespring.com/s...
#Teardown #Microscope #Switch
Haha this is epic level trolling.
"Mutter mutter 15 minutes fixing a dodgy switch mutter mutter"
"Really, here's 15 more minutes of me taking that switch apart... enjoy!"
@@Okurka. i forgot to add one to my Xmas list 😂
dave gives what the people ask for
MUTTER!
Mutter is a german word which means mother or mum or nut!
@@DavidKrautscheid It also means to grumble. At least it does in swedish.
Any of you switch afficionados out there? Weeeeell don't mind if I do! 😉
When it comes to these tactile microswitches the difference between the cheapest of the cheap and the good quality "expensive" brands like ALPS is actually surprisingly low. Yes, they look a bit rougher around the edges, but if things like bouncing isn't a big issue for you they will perform just about the same. Even the cheapies last for a ridiculously long time, as you know from your look at snap domes in the past. I definitely think the failure of your switch was due to mechanical wear on the plating, which when worn through opened up for tarnishing with a less conductive oxide.
I would contact an ALPS rep in Australia. They love sending out sample packs. They have these beautiful "catalog" boxes which open up with a beautiful display of many of their switches which you easily could just spend an entire wonderful afternoon flicking and clicking. They have a broad selection of snappy ones like these all the way to really nice soft touch long travel ones too. Truly a sight to behold!
Bet the reps will send one in to the next mailbag, not only from ALPS but every other switch manufacturer. Dave will be swimming in switch parts.
@@SeanBZA No doubt
This was way more interesting than I may have thought. Thanks
Taking everything to bits is always interesting, especially if you have a nice microscope like the TaGaRno!
@@rkan2 Tagano
@@EEVblog fiXed
@@EEVblog Tagarno
@@sylvainlazare1114 LOL, yeah Tagarno
Now this is the random EEVBlog stuff I subscribed for 😅
Soo good to see this video after reading the comments on the last video! Love your responsiveness to comments dave
1:15 - 1:33 Oh, really? So THAT'S why some of my older equipment thinks i'm pressing buttons other than the one I'm actually pressing! I've wondered this for years, and never really figured out what could possibly be wrong!
Huh...!
Ha.. I watched the entire video and it took until your comment to click with me..
I have a volume button on the steering wheel of a car that sometimes goes down when it is supposed to go up. I bet it's the same thing.
Thank you - I just run over that again and got it, genius, resistors for different voltages instead of a multiplexer, and if you won't go to the expense of a multiplexer then the switches will be cheap and the resistor network will go faulty too! Wow, I learnt today! :D
i took apart one of these switches once when it failed on an old scooter charger
the dome was totally etched through, the switch put 5v across a pulldown in to a 2.2u cap ... the pulldown was 4.7 ohms and passed an amp
i assume someone at some point forgot to put in a 'k' in the schematic and it simply worked and no one caught it, but that poor 1/4w resistor and 7805 survived
but for this i assume repeated pressing etched away a criminally thin plating of silver and let it oxidize
Silver oxide is reasonably conductive, the black is silver sulphide, which is an insulator, or at least a poor resistor with 5V applied. Silver plated with a thin coat on the one side of brass sheet before being punched out and into the die to shape it. The more expensive uses stiffer and harder nickel, which is more resistant to corroding, and can get by in regular atmosphere with no coating on it, but yes for the best you want heavy gold plating on the disk and the body contacts. A flash coat will work for a while, but will wear away, making the contact problem worse.
Body on the cheaper ones is only going to be tin plated, because silver is expensive, and tin plate is good enough to pass initial testing. Likely stamped out of a long roll of tin plated steel, and then used in a long strip to injection mould a load of switch bodies, then punched out and formed after assembly.
What about the domes over PCB mounting. Would a gold plated PCB pad under the dome make an equally reliable contact?
@@pahom2 Yes the gold plate on the PCB is good, but needs to be thicker than ENIG for a long lasting contact, needs to be the thickness used for edge connectors, as the ENIG coat is more there for keeping the exposed copper surface from corroding than as a wearing layer. Thin coat will wear off after a few thousand cycles exposing the bare nickel plate under it, and that in turn will wear away exposing the copper which will oxidise fast.
If you want the most reliable contact cheap you will want a capacitive sensing pad under the PCB surface to sense distance from the dome, which will last till the PCB itself wears away. There are plenty of more expensive keyboards that did exactly this, or where they used hall effect sensors per key to detect a moving magnet. Not a cheap option though.
@mister kluge Generally well known brand names, and the higher end switch assemblies they supply. ALPS and Klixon come to mind, but there are a number of Aerospace suppliers of them that make them with gold alloy contacts that are rated for hundreds of millions of switch cycles. You pay for them though, as each comes with it's own serial number and test certificate.
Tin oxide is a tenacious insulator! As Apple learned with the Apple /// -- the memory board plugged into the motherboard with standard pins at each end. Engineers specified rhodium and gold for the pins and connectors. Purchasing went with tin on tin, which was cheaper. Oops. Thermal cycling, tin oxide, failing computers...
I like how you switched it up with this video.. 😏
Thanks for the dive... I was going to suggest that you take the actuator from the bad switch and move it to a new one... melting the top down should be not too hard. Now I want to see the original switch fully repaired! Get some silver polish and a Q-tip and then treat it with Caig Labs magic liquid! Rotate the dome 90deg. I've done switch repair with vintage musical equipment, you can drill out rivets and use 2-56 hardware to bolt them back together.
Ah, that is why my old Denon DVD player believes I'm pushing the hdmi type button when I'm actually pushing the eject button. I always thought that it had to do with a bad seated flex cable or even a broken flex cable. Thanks for this video!
really enjoyed this
Yeah, especially nice to see with yhe Takano, really nice quality shots.
This video was definitely very interesting. Who knew what was inside. As always, thanks for your fantastic and informative videos. I've watched so many I'm developing an Aussie accent.
It's a plastic dust compacted and stuck to the bottom of the dome. Heat it up - it should show itself right away.
I have repaired and replaced various micro switches with and without micro controllers, unfortunately as a rule they are not made to be replaced. Good to see the effort though. Cheers from Canada.
Silver does oxcidise but silver oxcide is still very conductive.
Great video - I must get around to fixing a mate's synth with dead keys!
I just finish fixing one 2 days ago and now this video pops up!
Mine failed in the exact same way.
As a temp fix while waiting for a replacement, I scratched off the oxide layer and the switch is working again (for now)
In my Saitek X55-Rhino stick I had to replace some of them already within fife years of use. But in my case the mechanical tactile dome failed and lost it's springiness resulting in low tension to the four way switches in the upper part of the stick. Good luck I had several of theses switches in spare salvaged from an old VCR front panel. :)
Is it just me or does the package for the "CNK" say "CKN"
That camera is absolutely fantastic holy shit.
"It's probably easier just to go down the shaft really and then squirt the whole thing" - Dave Jones, 2020
I just took apart a Schurter Inc. 1301.9302 that I got from Digikey and it looks almost identical to the CNK switch (the one on the left at 13:38), except the split/slot in the contact is at a jaunty angle. Puzzling. Patent dodging? Who knows.
Maybe some of that CAIG Laboratories DeoxIT would help it. It was always around in the repair depot around here.
that era of tact switches were very prone to failure. Contact cleaner would never really help; replacing the switch always did. Glad to finally see the failure inside. One would assume an issue with the material used for plating?
Basics! This is basics, if you reach the people who want to learn
Learning these new interesting things is never boring!
"Surgery? What surgery? I did understand autopsy!?" The doctor said
My Roland D-50 needs several of those small suckers exchanged.
So the common failure mode is galvanic corrosion on the tactile dome, because the silver plating is soft and thin wore through?
is it just an illusion or does the center contact of the first one have tiny spikes to penetrate slight corrosion layers?
Yes, it is knurled for self cleaning. That's probably an Omron switch judging by the brown shaft.
@@mrnmrn1 I found the switch afficionado!
I'm actually impressed by the ebay switch. The dimple ensures a large contact surface.
@@clemensruis Yes, but I'm not sure if it also ensures sufficient self cleaning effect. Also, the large contact surface will not make contact all at once, which can cause increased bouncing.
I replaced the same tact switches shown here in my 28-year-old digital setting circle computer on my telescope 3 months ago. I replaced these switches twice before. I did took two apart while my own Digi-Key order was being filled out and shipped as I had an observing session that night. Same thing. The gold tint is silver tarnish. I manage to clean using the wire brush pick tool. I got the computer working plus the lead-acid battery charger that was getting dodgy in its own menu switches. I got the order in and I replaced all of the tact switches in both units. A lot better operation in both units. Then i found 2 switches from the last order 12 years ago and found that they also had resistance values higher than the new switches. Even stored in a bag they will still tarnish.
Interesting to see the corrosion, thanks for sharing. You can clearly see the ebay Dodgy Brothers version is made down to a price. The failed switch looks pretty much the same quality as the good one. If it averages one on and one off a day, say 14 actuations a week, 728 a year, you'd expect at least 10 years minimum life. Octopart datasheets show 100,000 actuations upwards (137 years!). How old was this one that failed?
Quality oem aircon manufacturers may care to take the controller back for the same failure analysis. Then send the suspect switch back to the switch mfg for their failure analysis and quality improvements. All of this costs
Those two side contacts seem as important as the center one, because the disc moves freely. On the faulty one, the plating is worst on the sides.
Also, in C&K one, the center pin is cut in half to work kind of as dual contact? Could potentially double the cycles if there's some galvanic degradation on the contact point?
5:49 ITS CRACKED!!!! The dome is cracked!!!
Looks like a scratch to me.
@@CristiNeagu Perpendicular to the scratch. Its a brittle fracture crack!! I know that telltale sign from a mile away!! Its fatigue from work hardening, which can happen to phosphor bronzes.
@@StreuB1 I don't see it. Not saying it's not there, just that i don't see it. All i see are two scratches, on at 9 o'clock, the other at 8 o'clock.
Scratch. Besides, a crack will not prevent it from making contact.
It's 100% definitely a crack, at 4:04 you can just see it on the top side of the dome. It's also cracked exactly where you would expect it to crack, mid way between the outer contacts where the disc is unsupported and would experience the most flexing. A crack doesn't necessarily stop them working , but it does make the switch action very mushy rather than a nice tactile click.
Likely source of the problem (if indeed it is silver plated): air pollution. Silver is no longer good for electrical contacts. Gold is the way to go. Silver sulphide is actually a semiconductor. See also: www.iaras.org/iaras/filedownloads/ijch/2018/017-0001(2018).pdf
Using a multiple resistor voltage divider can be made almost as reliable as a key matrix by using the switches to select where the ADC is connected to the resistor divider instead of having a fixed resistor and connecting another one in the circuit with the switch. This way the internal resistance of the switch would be negligible in relation to the ADC's internal resistance.
это все хорошо, но эти кнопки еще бывают с большими нестабильными утечками
Measure the resistance between the corroded spot?
IIRC, pure stainless steel is non-ferrous. You might be able to test with a magnet, bearing in mind the plating can skew the results. Maybe a weak magnet to reduce false positives? 😁
Thank you!
The Best of the three switches is the "cheap eBay" one. The other two have silver plating, the oxidation of which, is the cause of your "dodgy" switch. The two ways to get long life from silver low current contacts are hermetic sealing or oxygen barrier, i.e. silicone, or other, grease. Silver plated switches that actually switch some current, typically 500 mA or more, self clean with a small "arc" which cleans of any oxidation until eventually the silver wears away and the base metal gets corroded.
Isn't that called 'wetting current'? I had an old (much heavier duty) pushbutton which wasn't giving good resistance readings on the multimeter. I used it to switch around an amp and then it worked. Same idea, I guess.
Interesting topic. Thanx
Now I know why my middle mouse button isn't working...
It barely worked and stopped again after using cleaner :D
The center contact of the CNK has been Bi-Bifurcated. (always good) Ron W4BIN
Nice Video Dave
Instead of changing the whole console for its malfunctioning we could change the button and save the world but we need to know some information about the design
The saga continues.
Noooooo what a sad bunch 😢
Really interesting video, thanks! I've been looking at mechanical keyboards and switches lately. I'm playing with the idea of building my own DIY gaming keypad, or split curved keyboard.
But now I'm looking at things like the wooting two HE which uses really awesome stuff like all analog keys using hall effect sensors haha
the black stains on the underside of the dome look like deposed carbon from arcing
Samsung TVs from around 2010 have the same problem with the buttons on the TV. Volume goes up or down, channels change, etc. They fail in open position! I once opened one up and only found a tiny trace of some material that bridged the central contact with the outer contact. No signs of corrosion or wear on any part since those buttons never get used. Very strange. It's good that Aliexpress has large kits of buttons of all shapes :) They are all crap, anyway.
9:40 yep, that looks like silver plating .. working with old RF stuff teaches you how tarnished silver looks like.
I wonder if they didn’t do the matrix cause they ran out of memory on the micro?
What's the brand of your camera ....... thx
pass some 20A through to celebrate end of 2020
There is not a outer ring! at 8:42 it's clear! It is an unique center contact.
You're right, the center ring is one solid piece of metal. But it's not a contact. It's not part of the circuit. It's not connected to anything.
I promote Eevblog to my friends who have no interest in electronics because they get to know who teaches me the slang I come out with nowadays. "Come a gutsa" is my favourite as I do that a lot. This video is a good illustration of why so much of what we rely on is effectively straight-to-landfill. ps to Dave; the word "Muriel" has other meanings - lookup: hilda ogden muriel on TH-cam
Man, I have taken these apart before but that's the dirtiest one I've ever seen.
Even expensive high quality switches fail.
I had many Grayhill matrix keypads fail sending multiple digits when pressed.
Can you please be so kind a tell me what brand is the camera you are using for as a microscope... thx a million
If it had any chance of being sentient ultra rare quantum-gap junction and trying to communicate....
I hate these stupid things. Every mouse I've bought with these button switches never lasts more than a few months, if that. All my stuff from the 80's with snap-action switches still works perfectly.
I would of used clippers.
I was developing the user interface for a product which used 5 of these tact switches, they had 22K pull-ups to 5V, and after a month or two of pressing every day they’d start getting unreliable and have to be replaced.
I imagine those two black stains on the back of the tactile dome are the contact points for the edge contacts, and they look awful. Get yourself a pencil eraser, clean the two stains off, as well as the smaller, less severe one in the center, then put that dome back in there and push it down while watching your ohm meter.
You can probably duplicate the result just by touching the sides of your probes to one of those black stains and the black center point. If you don't use the point you won't punch through the black. That almost looks like arcing, not corrosion.
I'm wondering if you could design a higher reliability tactile switch, by putting some sort of ratchet teeth very fine like a file, along the rim of the contact dome and the base of the housing. The idea being that every time you push and release the button, the dome hops and rotates a small amount. That way any black contact arcing is going to be spread around the entire dome, and not concentrate a single location. It's unfortunate you don't get any wiping action with one of these, since that's really the way to have long-term contact reliability in a switch.
Interesting video Dave :-)
Small signal relays fail the same way even with gold plating if they are not passing current ...
The hole in the cheap one is a dust collector. They thought of everything...
02:53 this moovement was even in 0.25 speed extremely fast!!!!! that makes me wonder how fast your body parts can moove!
By any chance would you know whether your Targano Microscope uses a 14-bit micro? The zoom level went upto 4000xH. If so why would they do that? Using a 14-bit micro instead of 16?
Edit: Corrected grammar
Dont know, it's never made sense to me.
Maybe it has a 14-bit ADC.
@@eDoc2020 Yup. Later that's what I thought might be case
Was this the building that flooded? Was it (the switch) damaged by the water?
No
@@Irilia_neko ta
"Nothing to write home about".
I encountered many bad tactile switches.
Many consumer products use such scheme, from vacuum cleaners! To tv controls.
It looks like it's made of chocolate, no wonder it failed. I've replaced loads of tact switches in synthesisers. The clicky discs must oxidise. On a Kawai K5 module they were serviceable, the plungers pulled out and you could clean the internals.
Edit: Looks like the two metals reacted?
I had an issue that these buttons triggered just getting near them! errffffff
stray charges/static in the air and in your body triggering the input on the micro
Very interesting
almost looks like silver mica disease in an audio RF transformer
How are you even meant to reassemble those buttons if theyre just molten shut :/
hot glueeeee
I think it corroded cause this switch always had a voltage applied.
spark
@@rtyhgfplmkoi4784 Not with 5V xD But more like different metals touching and the 5V always applied 24/7.
I assume you used your usual 12" knife to slice off that plastic "rivots". ;)
Heyy thanks a lot!
Can you do the 6*6*7.3mm Silent Switch Micro Mouse Button?
I wanna know how come it's silent?! 😂
Look to be silver oxide causing the issue. I wonder how many mA is this switch handles in the special voltage divider arrangement. It does not seem to be especially hi impedance. A similar switch specification is max. 50mA/12V tech.alpsalpine.com/prod/e/html/tact/snapin/skhh/skhhaja010.html
Nice video, thanks.
Next experiment: Would contact cleaner have saved this one?
обычный спирт спасет. но только кнопки эти не разбираются. А вот разборные кнопки можно мыть. я так и делаю, что бы лишний раз текстолит не греть.
Last time it took 25 minutes to replace a switch, not it takes 15 minutes to open it. LOL
Well it's easy to see that you caused the problem yourself by using this switch to start your car before you put it in your microcontroler project.
Why don't you use tweezers? I've had my two (Torcade) stainless steel for 35 years, I couldn't live without them. A greeting
i hate it when people call them tactical switches
imo, tactical switches are dead silent and have a VERY positive feedback when used
9:08 "It's easier to go down the shaft and squirt the whole thing" - Dave from EEVblog
Ah, I'd like to own a Tagarno... but 2500$+ ....
you really should have just thrown it out then dumpster dove for your own ac unit :P
"No tactile buttons were harmed in the making of this video"
......this is a lie, i just watched 3 of them get disected
One word: Tweezers
Almost looks like there's been s tiny bit of arcing.
I don't Even know wy but that was interesting
Don't turn it on..take it apart..
at least I now know what is wrong with my a/c controller .... cant fixit,but I know what is wrong.
Why tf do I get calm and my anxiety goes away when I watch EEVblog, Ben Eater or Louis Rossman videos? What's wrong with me?
You're a nerd. Welcome to the club.
Hey that's not a knife....
If you think switches are boring try being an electrician, it's all we do lol
What, no switch repair video? Rotate or clean the corroded disk, or replate it. Or do a collab with someone who has a chemical vapour deposition rig and an electron microscope to check their work. You damn kids today, ruin three switches and throw them away. We didn't do that in my day, no sir-ee bob's your mom's dog. We'd walk 18 miles in the harshest Australian blizzard sword fighting rabid wombats and sticky drop bears to the neighbor's garage to see if he had a disk to spare.
What ... seriously? Dave, perhaps we should alert the Guinness World Records, for the cheapest tear-down ever! I mean ... Ali Express sells these 'literally' for less than a penny apiece. Just F^&%^%#$)^)ing replace it already!
What's even more embarrassing is that I actually watched the whole video. Usually, you are brilliant, but really ... what the hell were you thinking?
These switches are a cancer in modern electronics. They fail all the bloody time and are often used in completely inappropriate ways. Anywhere that you need durability and reliability, forget about them. If there's a manufacturer out there that actually makes decent ones, I haven't found them yet.
ALPS hands down. Can't recommend them enough. Any switch will break down over time with enough actuations though, that's unavoidable.
4/10 on iFixit.
great microscope-please don't ever use it to examine a nuns nasty.