Do BOX switches stretch PBT and break ABS keycaps?

User avatar
hansichen

29 Jul 2018, 21:38

I have strong doubts that putting a metal keycap onto a stem will have any impact on a stem of a switch.

User avatar
Darkshado

29 Jul 2018, 23:50

Same here, if anything I'd be worried about breaking the oversize stem on the Box switch.

A possible (tedious and inconsistent) DYI approach would be to use a very small file or hand-spin a small rotary tool (e.g. Dremel) bit to remove a bit of material from the stems.

In a perfect world: once new stems are made within MX tolerances, Kailh/NovelKeys should have replacement stems available to retrofit.

User avatar
Wodan
ISO Advocate

30 Jul 2018, 09:07

chuckdee wrote: Just because there is a problem, doesn't mean that the switches are bad. Just the manufacturing tolerances are off.
It‘s also a demonstration of very bad QC!
With high precision components that claim to have a 50mil actuation life time, failing to even get easy-to-measure dimensions right just shows how little QC is actually done at Kaihl even with brand new designs.

Makes me wonder hiw many other issues with their BOX design will show up. This is just so common with chinese companies. They have good ideas, put them together somehow, puke that shit into the market somehow and then wait for customer feedback and use that to make a „new improved“ version of their product. That‘s why they are so much quicker than american and european companies with their „innovations“ because you are the ones doing their QC. Gateron has a list of shit like this happen - just never fucked up precious caps like BOX does. Remember the early milky top Gateron with binding issues? Clearly an attempt to improve the Cherry design that couldn‘t have passed any serious QC. Then these extended rim Gaterons that Massdrop built into their KC60 boards with Costar stabs - but the stab wires would hit the extended Gateron rim and pull all stabilized keys 0.5mm down. Then the melting of early Zealios ... which still seem to be in beta stage looking at the number of improvements happening with every new round.

It‘s really a paradox that we feel like it‘s good customer service when we report design issues to a manufacturer and they fix it in a new revision. Good customer service would be fixing these design issues long before the product ever gets into the hands of a customer. Most of these chinese Cherry clones have undergone more design changes in 3 years than original Cherry switches have in 30 years.

What really bothers me is that these cheap, half-baked China switches are continously being marketed to the community as the latest innovation and something prime or exclusive when they are really just the shit you find in 25€ mechanical keyboards on Amazon. And people buy into this shit and put these switches into their treasured customs. You‘re building a keyboard that is supposed to last 10-20 years but your choice of switches is based in most upvoted stem color on Reddit. Can‘t wait for people to travel to China to make sure their Vagina BOX switches have the right shade of Pantone pink.

andrewjoy

30 Jul 2018, 10:32

I think the idea to improve the MX design is a good one. I agree that the volume of testing could indeed improve, as much as i am not a huge fan of many cherry switches ( vintage blacks excluded) the videos of the testing they do are bloody impressive. You may not get the best feeling switch in the world, but you sure as hell know it will work.

I think if they where sold as such (n sold as "early access" switches i think it would be better, but they are priced like a product that has had the amount of testing genuine MX has had and they have not.

That being said i still do want to try box navy in a full board.

User avatar
Wodan
ISO Advocate

30 Jul 2018, 11:05

The problem with "improving" the Cherry MX design is that it has been established as _THE_ standard for the past 30+ years. There are SO many components and boards and tools that are made specifically for the original MX design - part of the reason why I thought the Gateron 1:1 clone approach is a great idea.

Just look at the issues you seem to be having with the QMX silencing clips as soon as you use switches that aren't 1:1 copies. Then there's so many kinds of stabilizers for MX switches, each with their own special needs ... Plate cutouts, PCB footprints, keycap restrictions ...

Then the question what to REALLY improve about the MX design when it's so deeply established. Using better materials for housing and stem, tweaking the tactile bump and spring weight. Reducing friction points and improving stability. These are the improvements in detail that would really help make the MX design better. And it's something Gateron and Cherry have been working on.

Kaihl instead never got their clones hyped as much as Gaterons and went to create a new MX compatible design named BOX which comes with so many question marks and solves so many things that weren't issues to me. The ball dropping this thread is evolving around isn't even what bothers me the most.

Decoupling the click/tactility from the actuation is just a stupid and cheap way of "fixing" the restrictions of the original MX design click and bump! With the manufacturing tolerances of Kaihl, I have been getting non-actuation clicks out of my Hakko True switches easily. Coupled with these gorilla weight springs, this even becomes an issue while typing on it. You sacrifice the actual purpose a click has (signal actuation) for a better perception of the click. That's just design over functionality! Why not add a small solenoid to the PCB and let that click for you - that's a much more honest solution and will actually click on actuation.

And this asymmetrical little plunger going into the "sealed" contact housing of the BOX switches is also solving a non-issue to me. MX switches have really never had a reputation for being vunerable to dirt. Of course you will find grindy, dirts boards but the majority of old and worn boards - trust me I found a LOT - are actually quite smooth and feel great. Even with keycaps you coud do your make up in. So thanks for breaking the established 50mil actuation golden crosspoint contact mechanism, turning it into something 1/3 the size triggered by an asymmetrical plunger through a sealed tube that will magically have the same rating or even better?
That little sealed plunger alone is giving me worries about longevity of the whole design coupled with the large surfaces on the outside of the BOX switch design that look like something that loves to catch dust and grime.

My advice here is stay clear of BOX switches for at least another 3-5 years and let Reddit do all further beta testing for you. Keep an eye on people giving "long term reviews" of their switches and try to get impressions what they feel like with a couple of million actuations. There is absolutely no reason to rush or hurry here because only two things can happen here. Either this design does manage to get back on track and will eventually become another standard design in the mechanical keyboard community and a few popular designs for these switches will evolve to become widely available. The right spring weight and actuation characteristics etc. The other alternative is that these switches just never really make it and remain the choice for 30€ waterproof mechanical keyboards that are made for people who wash their monster energy drink and WOW boss battle ejaculate off the boards on a weekly basis. In this case you should hurry to try them because they will be out of fashion soon and you might miss the opportunity to ruin a nice kit with these stem gapers. I wouldn't be sorry I missed these switches if they turned out to become something that could be missed :)

With all this China stuff its better to wait for other people to do the QC before you get into it.

User avatar
Elrick

30 Jul 2018, 11:14

andrewjoy wrote: I think the idea to improve the MX design is a good one. I agree that the volume of testing could indeed improve, as much as i am not a huge fan of many cherry switches ( vintage blacks excluded) the videos of the testing they do are bloody impressive. You may not get the best feeling switch in the world, but you sure as hell know it will work.

That being said i still do want to try box navy in a full board.
Actually have several WASD V2 keyboards with the traditional Greens and the new MOD 55g switches installed. The brood loves the 55g MODs, they do have a decent tactile feel when using them.

Agree with the Zealios fad because bought two keyboards one with 62g Tactile and another with 78g Tactile and they aren't. They have more a linear feel rather than their so-called "tactile" description going on all over the web. Also their prices are ridiculous for what they are. Weren't impressed at all by them.

The BOX Navy switches scares the kids because the kick back is very pronounced, hence they could only use it for about ten minutes before crying, how hard it is to work the keyboard with these switches ;) .

Out of all the Cherry MX switches the BOX Navy's rule simply by their feel, the kick back is truly astounding compared to every other Cherry switch that was made before it. The standard Cherry switches are comatose in performance, just missing that special jolt, like the Blue or White Alps gives you when you use their switches over a long period of time.

Even though I like Alps I HATE their prices and unavailability. So due to their own scarcity and stratospheric pricing for even a second hand, toilet stained keyboard, I would rather purchase another brand new BOX Navy keyboard instead.

User avatar
Elrick

30 Jul 2018, 11:39

Wodan wrote: I have been getting non-actuation clicks out of my Hakko True switches easily. Coupled with these gorilla weight springs, this even becomes an issue while typing on it. You sacrifice the actual purpose a click has (signal actuation) for a better perception of the click. That's just design over functionality! Why not add a small solenoid to the PCB and let that click for you - that's a much more honest solution and will actually click on actuation.
That is why WE are still waiting for our F77's to arrive, when ever that will be. But that's an unnecessary complication in adding MORE complexity that will easily break down. A solenoid will cark it far quicker than any BOX Navy switch.

Hakko's are garbage had tried to use them but they fail, maybe the designers concentrated wrongly on it trying to 'feel' like a Topre (they claimed that) or like a Clear switch (also claimed that).

Perception is everything and if it works then it works regardless of how it does that. Don't forget the mid production versions I think during 2010 of Cherry Blues, Browns and Reds, that felt like rubbing sand paper along their stems constantly. YES it was Cherry produced, if even the masters of their own designs failed to produce decent functioning switches for a time, then Kailh have done one better. New design off the blocks but it works (to a degree) and the price is indeed very affordable.

Also the Kailh designs is open to anyone wanting to refine the design further which quite often fails as the Hakko atrocity proves.

Have no idea how long my BOX Navy's will last (no interference by Input Club) but it's certainly a nice ride whilst using them.

User avatar
Wodan
ISO Advocate

30 Jul 2018, 12:59

My criticism of some Chinese clones shouldn't have implied Cherrys production has been flawless but we're talking about different levels of failed QC here.

The issues I've been mentioning aren't just about the smoothness of a switch or the type feeling and click ... it's stuff that ruins a build or makes switches unusable in combination with certain other components.

Type feeling and switch choice is such a personal perception, more choice is great and obviously the healthy Chinese competition has awoken Cherry and helps reach more people with cheaper mechanical keyboards. But the sloppy QC and the hype marketing for these fancy colored limited edition China clones is just ridiculous and makes me stay FAR away from these products.

User avatar
MehAdviceGuy

30 Jul 2018, 13:21

Hello Guys!
I have compiled a list of stem thicknesses of different switches that people has provided to me.
You can take a look at it here if anyone is interested: https://docs.google.com/spreadsheets/d/ ... sp=sharing

If you have some measurements of stems then send them my way and i will update the list with what you give to me.

EDIT: I also trimmed the BOX part of a BOX Jade keyswitch to see if the stem size widened at the end, I can say that on my sample size of one that it does not widen or narrow at the end.
Image

User avatar
chuckdee

30 Jul 2018, 14:17

Wodan wrote: My criticism of some Chinese clones shouldn't have implied Cherry's production has been flawless but we're talking about different levels of failed QC here.

The issues I've been mentioning aren't just about the smoothness of a switch or the type feeling and click ... it's stuff that ruins a build or makes switches unusable in combination with certain other components.

Type feeling and switch choice is such a personal perception, more choice is great and obviously, the healthy Chinese competition has awoken Cherry and helps reach more people with cheaper mechanical keyboards. But the sloppy QC and the hype marketing for these fancy colored limited edition China clones is just ridiculous and makes me stay FAR away from these products.
In my opinion, this one failure shouldn't stop innovation, and these Chinese clones have done what Cherry hasn't been willing to do (well, up until they started to cut into their margins)... innovate. Innovation is never without its risks and failures, but that shouldn't stop forward progress. This isn't to excuse their QC- it should have picked up on it, which makes me think that they use minimal caps in their QC process when they should be using caps from the majority of large keycap producers. But none of this should impede innovation.

User avatar
Wodan
ISO Advocate

30 Jul 2018, 14:36

chuckdee wrote: In my opinion, this one failure shouldn't stop innovation, and these Chinese clones have done what Cherry hasn't been willing to do (well, up until they started to cut into their margins)... innovate. Innovation is never without its risks and failures, but that shouldn't stop forward progress. This isn't to excuse their QC- it should have picked up on it, which makes me think that they use minimal caps in their QC process when they should be using caps from the majority of large keycap producers. But none of this should impede innovation.
Unfortunately, the Chinese approach to innovation often impies this terrible banana products and hillarious QC and I've mentioned more cases from Gateron too. You can see it in other products as well. I've gotten a mic arm on Amazon from a Chinese brand and they made a list of shortcomings from previous revisions they have overcome. And there was this really funny story with the "Koosh" Nerf darts from China. The first batch they sold had the tips glued on really badly and they would come off so easily so the second batch had so much glue on the tip that they would get stuck in the blasters ... both batches have obviously not really been tested because once you started using these darts the issues became obvious instantly.

So I absolutely embrace innovation and we've already seen a very positive impact on the market. But it's not the completely new designs (low profile, BOX, Flaretech, ACE Pad) that drop my panties because but the small details companies like Gateron and Cherry and maybe others improve in the classic MX design that actually help me. I'm in the customs game and want a switch that is really worth being put into a nice keyboard kit. For that, I don't consider the price I paid for the switch or the color of the stem but the faith I have in the manufacturer of said switch.

JBert

30 Jul 2018, 15:06

Sup wrote:
scottc wrote: Image

just saying
dude the stem broke of the switch not the stem of the keycap, I don't see the connection. Box switches indeed do strecht key caps/break. Rip TMX Bro caps artisan got gaped to oblivion to the point it just flies off on non box switches. One of the reasons i don't use box switches anymore.
Scottc must likely have been making a potshot on Kailh's QC practices in the past, and how they haven't learned.

Pictured is a Kailh PG1511 Yellow switch on a Rapoo V7 where the stem cleanly snapped off. Picture and source are in that last link from back in 2011. Sadly the original report is in Chinese, so I can't give you more context.

User avatar
hansichen

30 Jul 2018, 15:21

chuckdee wrote:
Wodan wrote: My criticism of some Chinese clones shouldn't have implied Cherry's production has been flawless but we're talking about different levels of failed QC here.

The issues I've been mentioning aren't just about the smoothness of a switch or the type feeling and click ... it's stuff that ruins a build or makes switches unusable in combination with certain other components.

Type feeling and switch choice is such a personal perception, more choice is great and obviously, the healthy Chinese competition has awoken Cherry and helps reach more people with cheaper mechanical keyboards. But the sloppy QC and the hype marketing for these fancy colored limited edition China clones is just ridiculous and makes me stay FAR away from these products.
In my opinion, this one failure shouldn't stop innovation, and these Chinese clones have done what Cherry hasn't been willing to do (well, up until they started to cut into their margins)... innovate. Innovation is never without its risks and failures, but that shouldn't stop forward progress. This isn't to excuse their QC- it should have picked up on it, which makes me think that they use minimal caps in their QC process when they should be using caps from the majority of large keycap producers. But none of this should impede innovation.
Cherry was the company who introduced silent switches and also with mx silver some shorter travel "gaming" switches. When the mechanical keyboard hype started they developed stuff together with Corsair.

For ages Cherry built boards which were mainly sold for industrial reliable standards, just look at the masses of iso de boards on ebay. Every post office, every medical place etc used Cherry switches for ages. And clicky switches were always like an unwanted child by cherry, hardly any vintage board came with mx blues and only few 1000/3000 boards were sold with mx blues compared to mx blacks. Also today the market for clicky switches is rather small, tactile and linear switches won. I can fully understand why cherry isn't investing in that regard, the development of good silent switches was imo the better decision.

Though we all know that cherry should introduce a new tactile switch, mx clears are really nice but they are a bit heavy for everyday mainstream usage. But the market for this is probably rather low cause most mainstream people are fine with mx browns. I personally would expect a silent tactile switch in future but not too much innovation in regard of normal tactile switches.

User avatar
hansichen

30 Jul 2018, 15:22

JBert wrote:
Sup wrote:
scottc wrote: Image

just saying
dude the stem broke of the switch not the stem of the keycap, I don't see the connection. Box switches indeed do strecht key caps/break. Rip TMX Bro caps artisan got gaped to oblivion to the point it just flies off on non box switches. One of the reasons i don't use box switches anymore.
Scottc must likely have been making a potshot on Kailh's QC practices in the past, and how they haven't learned.

Pictured is a Kailh PG1511 Yellow switch on a Rapoo V7 where the stem cleanly snapped off. Picture and source are in that last link from back in 2011. Sadly the original report is in Chinese, so I can't give you more context.
I don't know the context of the picture but to me it looks like that the board fall down on the floor. In that case kalih is not to blame as the same happens to Cherry too. Low profile cases without bezel are not made to be thrown onto the floor

User avatar
chuckdee

30 Jul 2018, 16:06

hansichen wrote:
chuckdee wrote: In my opinion, this one failure shouldn't stop innovation, and these Chinese clones have done what Cherry hasn't been willing to do (well, up until they started to cut into their margins)... innovate. Innovation is never without its risks and failures, but that shouldn't stop forward progress. This isn't to excuse their QC- it should have picked up on it, which makes me think that they use minimal caps in their QC process when they should be using caps from the majority of large keycap producers. But none of this should impede innovation.
Cherry was the company who introduced silent switches and also with mx silver some shorter travel "gaming" switches. When the mechanical keyboard hype started they developed stuff together with Corsair.

For ages Cherry built boards which were mainly sold for industrial reliable standards, just look at the masses of iso de boards on ebay. Every post office, every medical place etc used Cherry switches for ages. And clicky switches were always like an unwanted child by cherry, hardly any vintage board came with mx blues and only few 1000/3000 boards were sold with mx blues compared to mx blacks. Also today the market for clicky switches is rather small, tactile and linear switches won. I can fully understand why cherry isn't investing in that regard, the development of good silent switches was imo the better decision.

Though we all know that cherry should introduce a new tactile switch, mx clears are really nice but they are a bit heavy for everyday mainstream usage. But the market for this is probably rather low cause most mainstream people are fine with mx browns. I personally would expect a silent tactile switch in future but not too much innovation in regard of normal tactile switches.
None of that obviates what I said above. Just as competition helps the consumer in the way of pricing, it also fends off stagnation.

rich1051414

30 Jul 2018, 19:19

I pulled the SA Maxkeys keycaps off my box royal switches, and they aren't loose on cherry blues, so I am thinking cap thickness has a lot to do with this.

User avatar
Wodan
ISO Advocate

30 Jul 2018, 19:31

rich1051414 wrote: I pulled the SA Maxkeys keycaps off my box royal switches, and they aren't loose on cherry blues, so I am thinking cap thickness has a lot to do with this.
They probably also have great pelvic floor muscles

User avatar
sealclubber

30 Jul 2018, 19:32

FWIW Mike from Novelkeys just released a statement concerning this https://www.reddit.com/r/MechanicalKeyb ... novelkeys/

User avatar
SwissArmyTin

30 Jul 2018, 19:43

My current main board uses Box Jades and Tai Hao ABS doubleshots, and none of my caps seem to have any indication of stress, warping, or cracking. Fit fine on MX Blues and BTC dome with sliders (the only other MX-mounts I have currently), so I guess I'm lucky :P

User avatar
ThePewster

30 Jul 2018, 19:56

Wodan wrote:
rich1051414 wrote: I pulled the SA Maxkeys keycaps off my box royal switches, and they aren't loose on cherry blues, so I am thinking cap thickness has a lot to do with this.
They probably also have great pelvic floor muscles
:lol: :lol: :lol:

User avatar
chuckdee

30 Jul 2018, 20:11

sealclubber wrote: FWIW Mike from Novelkeys just released a statement concerning this https://www.reddit.com/r/MechanicalKeyb ... novelkeys/
Just coming to post this. A very good response, at least going forward. But no mention of back...

User avatar
ThePewster

30 Jul 2018, 20:49

chuckdee wrote:
sealclubber wrote: FWIW Mike from Novelkeys just released a statement concerning this https://www.reddit.com/r/MechanicalKeyb ... novelkeys/
Just coming to post this. A very good response, at least going forward. But no mention of back...
Exactly. I've got 60 Hako Royal Trues lying around, and hence, I tried to check those unused switches with fresh keycaps.

Tested SA Carbon and XDA Blanks (aliexpress) in this order:
- Cherry MX Blue - Required some force to fit both type of keycaps
- Kailh BOX Jade - More or less similar force as for the above switch
- Novelkeys X Hako Royal True - More than usual force to make the keycaps fit

Removing keycaps from Hako Royal True was a bitch since the switches weren't soldered or plugged and I wasn't trying to rip things apart. When I tried plugging the same keycaps back into Cherry MX Blue and Kailh BOX Jade switches, and both keycaps slipped in easily without much pressure applied as earlier.

Hako Royal Trues immediately affected the grip of SP's SA Carbon key. Plugging YMDK or KBDfans version of XDA keycaps on Hako Royal Trues had a very minimal effect.

User avatar
chuckdee

31 Jul 2018, 02:57

hansichen wrote: I have strong doubts that putting a metal keycap onto a stem will have any impact on a stem of a switch.
https://geekhack.org/index.php?topic=96 ... msg2644362

In case you don't want to go to geekhack, there's an image of what he used and the following quote
I have got a workaround for the problem, if one has some loved box switches.
Just put a metal keycap a few times on the box switch, and the stem will be a little tighter.

I used a cheap metal keycap which I got from ebay.

User avatar
Elrick

31 Jul 2018, 07:19

ThePewster wrote: Removing keycaps from Hako Royal True was a bitch since the switches weren't soldered or plugged and I wasn't trying to rip things apart.
I've taken to applying some PTFE to the Key-cap inner part, so when it pushes down on my BOX Navy's it's easier to later remove them.

Remembered some years ago when Signature Plastics first released their SA profiles, I read a lot of negative feedback on these new key-caps having a super tight fit on ALL Cherry Switches. In fact some Cherry Switch stems were removed out of their own switch casings when the key-cap was removed, that was how serious it had been.

Had to individually apply some PTFE on every single new SP design and then it was far easily to insert and/or remove them.

Always have a spare aerosol can or bottle of PTFE around, when using any keyboards ;) .

User avatar
Elrick

31 Jul 2018, 07:27

sealclubber wrote: FWIW Mike from Novelkeys just released a statement concerning this https://www.reddit.com/r/MechanicalKeyb ... novelkeys/
At least the manufacturer has gone back to their ORIGINAL design specs.

Really want to conduct a good ole Country Boy Witch Hunt for DAREU executives, they need to be swinging from some trees until the crows pick out their eyes on a lazy hot afternoon.

User avatar
MehAdviceGuy

31 Jul 2018, 11:45

Response from kailh if someone is interested: https://www.reddit.com/r/MechanicalKeyb ... nd_stress/

User avatar
MehAdviceGuy

01 Aug 2018, 08:50

We now have people reporting in that XDA Keycaps have started to stretch and crash from BOX switches: https://www.reddit.com/r/MechanicalKeyb ... _switches/

I have also heard that cheap keycaps should be having issues. (I have destroyed some from a old AJAZZ keyboard by putting the keycaps on BOX Jade switches.)

User avatar
chuckdee

01 Aug 2018, 17:29

Elrick wrote:
sealclubber wrote: FWIW Mike from Novelkeys just released a statement concerning this https://www.reddit.com/r/MechanicalKeyb ... novelkeys/
At least the manufacturer has gone back to their ORIGINAL design specs.

Really want to conduct a good ole Country Boy Witch Hunt for DAREU executives, they need to be swinging from some trees until the crows pick out their eyes on a lazy hot afternoon.
And this is the reason it seems shady to me. They changed for one manufacturer and then blamed them for the error when Kailh should have only used that spec for that one manufacturer. It's not their fault that Kailh used the same specs on all of their switches, IMO.

IronSunsetscales

04 Aug 2018, 17:42

First time I heard about this issue, was in a video Thomas had put out this morning. To my horror I realized that I do have a board with box navies and a SA carbon on them. And yes each stem shows a while line of distress, probably even cracked. No snap sound either when you put them on, they don't even struggle to go onto the switch either, yet I have a fantastic set now ruined to these sh*t :/


EDIT:

Slapped on some PBT and ABS Tai-Hao caps on them, each of those snapped and cracked while they're being pushed on.

User avatar
abrahamstechnology

05 Aug 2018, 00:24

hansichen wrote: Cherry was the company who introduced silent switches.
What about Alps Damped?

Post Reply

Return to “Keyboards”