Cleaning Alps: return click woes

User avatar
an_achronism

21 May 2021, 13:10

I have recently acquired a few Alps boards which are in very mixed condition. Some are in excellent nick functionally, with the switches performing very well indeed, but absolutely atrocious condition cosmetically; others are the other way about. The latter are less disgusting but much more frustrating to deal with.

I started with a board full of Alps SKCM Orange which is cosmetically OK but had serious binding on several keys (including 1U ones, so not a stabiliser issue). Disassembling the switches was very easy, as was brushing out dust and dirt accumulated over the years, which I did and then very carefully put them back together. I was extremely careful to ensure I was re-assembling exactly in the same configuration that the switches were originally in, with the tactile leaf going in the same way, the top housing in the original orientation, and the slider going neatly onto the spring top. And yes, I closely examined Chyros' video on the subject to be absolutely certain that I'm putting them back together correctly. But despite this, the switches I have cleaned and reassembled all have the same problem to varying degrees of severity: they all click at least some of the time when the key resets, on the upstroke.

Others keys I haven't cleaned also have this issue inconsistently and I've really tried to find info on this but the only thing I can find is suggestions to do a "paper mod" wherein the tactile leaf is afforded a tighter fit against the housing wall by inserting a small piece of paper in between them. However, I think this assumes that the click is being created by the leaf pulling away from the housing when the slider hits it such that it then snaps back against the housing to create the click when the slider clears the leaf, and I'm not totally convinced that's what's happening here. I tried it anyway, several times with the paper cut differently and inserted slightly differently each time, but all it ever did was interfere with the movement of the mechanism and that obviously isn't progress.

I'm not sure but I'm worried that some part of the mechanism may somehow be getting slightly bent during re-assembly and that's what's creating this return click. I've stopped cleaning in case I create the click on more of the switches but the board is basically unusable in its present condition.

Does anybody have any ideas about what might be going on here and/or how I might fix it then avoid it happening in the first place when cleaning the rest of the switches?

EDIT: Oh and the four wee corner tabs that are meant to hold the tactile leaf steady when the slider passes (which you could remove intentionally if you wanted to turn tactile Alps clicky) haven't been noticeably bent so I don't think it's that. I could just go "f**k it" and pull all the tabs off to make them full on clicky but I don't want that. These were designed to be quiet tactile switches and that's what I want them to be. If I want a clicky board, I'll use a different board.
Last edited by an_achronism on 21 May 2021, 14:18, edited 1 time in total.

User avatar
Muirium
µ

21 May 2021, 13:23

Was I the source of the vexing oranges? (I thought they were salmon but I’ve been wrong on that before!) I feel your Alpine pain.

User avatar
an_achronism

21 May 2021, 13:53

Muirium wrote:
21 May 2021, 13:23
Was I the source of the vexing oranges? (I thought they were salmon but I’ve been wrong on that before!) I feel your Alpine pain.
... Yes. Hah.

And I thought they were maybe salmon from one of the pics but seeing them in person I'm pretty sure they're orange.

Cleaning them did seem to fix the binding but I don't want them to click unevenly either, so now the board is just sitting there taking up space. Would very much like to make it useful but don't know what to do.

User avatar
NeK

21 May 2021, 21:57

Sorry to say this, but by cleaning them you removed the most essential thing of those early ALPS, their factory lube.

Believe it or not that lube helps, not only in how smooth and nonbinding they are, but somehow it also helps in mitigating the horrible upstroke click. Yeap, you heard that right.

However, I do know what is causing the up click and also have a solution for you to try. You are right to think that it doesn't have anything to do with the leaf hitting the stem or the back of the housing, no it's not that. The real cause of the upstroke click is the friction that is happening between the nails of the leaf and the stem, that makes the stem drag up the leaf along on its way up. This moves/rotates the leaf ever so slightly and then when the stem passes the point where the "nails" slide under it, they suddenly push the leaf back down hitting hard the bottom of the housing, making that horrible click sound (and tactile feeling too). It's worth noting that his up click problem happens on both tactile and clicky switches.

So the solution here is two fold, first the nails and the stem should be properly lubricated* and therefore prevent the stem from dragging the leaf. And secondly and most importantly, the leaf should not have clearance to move laterally inside the housing. You can fix that by putting the stem and the leaf in the upper housing, have the stem in the pressed position, so the leaf is folded against it and with a toothpick move around the leaf. It should not be moving. If it does, then the little ears at the bottom of the leaf should be bent outwards a bit, in order to hold the leaf in place.

Try it and let us know.

* There is a whole thread about that lube, and it still is under "investigation" and research to find a replacement. There is plenty of info on that thread about it.

viewtopic.php?f=2&t=23878&start=360
Last edited by NeK on 21 May 2021, 22:28, edited 1 time in total.

User avatar
an_achronism

21 May 2021, 22:18

NeK wrote:
21 May 2021, 21:57
Sorry to say this, but by cleaning them you removed the most essential thing of those early ALPS, their factory lube.

Believe it or not that lube helps, not only in how smooth and nonbinding they are, but somehow it also helps in mitigating the horrible upstroke click. Yeap, you heard that right.

However, I do have a solution for you to try. The cause of the upstroke click is the dragging of the tactile leaf by the stem on the way up. This moves the leaf ever so slightly and when the two "nails" slide under the stem, they suddenly push the leaf and it hits the housing making that click sound.

So the solution here is two fold, first the nails and the stem should be properly lubricated* and therefore prevent the stem from dragging the leaf. And secondly and most importantly, the leaf should not have clearance to move laterally inside the housing. You can fix that by putting the stem and the leaf in the upper housing, have the stem in the pressed position, so the leaf is folded against it and with a toothpick move around the leaf. It should not be moving. If it does, then the little ears at the bottom of the leaf should be bent outwards a bit, in order to hold the leaf in place.

Try it and let us know.

* There is a whole thread about that lube, and it still is under "investigation" and research to find a replacement. There is plenty of info on that thread about it.

viewtopic.php?f=2&t=23878&start=360
You know, this was something I wondered about when I opened the very first switch, and I was specifically looking for lube. But I couldn't see any, and when I say "cleaned", I just mean "brushed the dust off". I didn't thoroughly clean in alcohol or anything abrasive, so I would be surprised if I took any lube off if it was there in the first place. I guess it was there and I'm wrong though. Still, they were screwed up really badly so leaving them alone wasn't an option anyway tbh. Still, when I noticed that I was creating this upstroke click I immediately stopped opening any more switches. So I've probably only opened about 5 of them, or so. That's mostly the number row and a couple of alphas that were especially awful for binding. The rest is still as it was when I got it.

I have been reading the lube thread already, as it happens! I wasn't sure which particular Alps switches it might apply to, or if they all had factory lube on them.

I did at least check for movement and they seem to be quite solid and secure, with all 4 tabs being at approx 90 degrees. This seems to be what makes folk turn to the "paper mod", effectively blocking the gap that might be formed by the leaf bending slightly away from the housing as the slider passes by so that it doesn't snap back and create that click. But yeah, I tried both bending the tabs out even more than they already are and inserting a small piece of paper, and neither helped. It seems perhaps the missing ingredient could be the lube. Which is unfortunate, because I really, really can't be arsed individually lubing every switch on a keyboard I don't really intend to use regularly. It was always one I wanted more as a curiosity: "how do these switches feel and sound?" rather than "this is going to be my daily driver now". Unfortunately, how they feel is extremely bind-ey, to the point of being totally unuseable.

Honestly, if I'd been paying more attention to this specific board I wouldn't have bought it; I was too busy thinking about the IBM Model F XT and various other boards I'd been trying. I had typed on it a wee bit but not really had a thorough test, and as such didn't notice the binding. As soon as I got it indoors and properly sat down to type, though, it was immediately apparent and my heart kinda sank.

User avatar
NeK

21 May 2021, 22:34

I updated my post to be more specific on how the click is happening. It is the leaf hitting down, not to the sides or front/back.

You should try my method exactly as I described it. You should certainly find a tiny weeny bit of movement, and that's what you need to fix. Also make sure that when you bent the ears outwards the leaf is perfectly flat and not bent inwards, becuase that will not fix the issue in this case.

User avatar
an_achronism

22 May 2021, 00:15

NeK wrote:
21 May 2021, 22:34
I updated my post to be more specific on how the click is happening. It is the leaf hitting down, not to the sides or front/back.

You should try my method exactly as I described it. You should certainly find a tiny weeny bit of movement, and that's what you need to fix. Also make sure that when you bent the ears outwards the leaf is perfectly flat and not bent inwards, becuase that will not fix the issue in this case.
Ta, I'll have another poke at them later. I have nothing to lube them with, though, so the best I can hope for is probably reducing the click to begin with. I'm honestly not that enamoured with the layout (which I expected) or the switches overall so it might not be worth it from my point of view to properly re-lube them (even without being opened, several switches are binding badly and sound/feel dry as a bone, and several have the upstroke click). But at least if I can stop them clicking and reduce binding, it'll be useable. If I find myself wanting to get more use out of it then I'll go down the lube rabbit hole...

User avatar
Bjerrk

22 May 2021, 09:20

an_achronism wrote:
22 May 2021, 00:15
But at least if I can stop them clicking and reduce binding, it'll be useable. If I find myself wanting to get more use out of it then I'll go down the lube rabbit hole...
The crappier the starting material, the bigger the satisfaction when you make something great out of it. :)
And lube can work wonders.

What's that I'm hearing from the bottom of the rabbit hole? There's cake, you say?

User avatar
an_achronism

24 May 2021, 07:00

So no amount of tweaking the tabs seems to entirely stop the minute dragging that's causing the click, ergo I'd need lube for that to work. However, the "paper mod" does appear to make a positive difference now that I've fine-tuned it more precisely. Cutting mat with 10 mm grid, each square of paper that size, last 1mm folded up to cup around the bottom of the tactile leaf. Seems to hold things in place and prevent the click.

Obviously this does very little for key feel, just limits the annoying return click. But it's something.

User avatar
hasu

24 May 2021, 10:05

I had annoying click noise similar to yours on my orange swithces, I think at least.
Also found "paper mods" and its idea totally made sense to me but I didn't give this try in fact because I didn't like to use paper in my Alps for some reason. Instead I used this bending method and worked for me. Idea behind this method is same as "paper mods".

https://www.reddit.com/r/MechanicalKeyb ... _noise_on/

User avatar
NeK

24 May 2021, 18:49

I still am not certain of what is the exact cause that results to this tiny dragging of the leaf, but I'm pretty sure the click is caused of that tiny displacement.

I think that it is a combination of the following:

1. before a switch is opened for the first time, the housing is more strongly shut, and thus it holds the leaf tighter so it doesn't hit hard making the up click, once you open it, and you then close it, it may not close with the same tightness, leaving more "room".
2. the missing lube and the friction makes the dragging far more probable.
3. the leaf has bent (with time or with use? who knows) in such a minute way that it is a tiny bit shorter in height, so there is more room left to move.

I guess, IF they are true, #1 and #3 are more important factors than #2. I am still "researching" the bending of the leaf in a correct way and what is actually going wrong with it.

I think that #2 doesn't matter that much, it just helps a little bit.

Once we understand exactly what is the mechanism, then we can say for sure.

User avatar
an_achronism

24 May 2021, 20:55

hasu wrote:
24 May 2021, 10:05
I had annoying click noise similar to yours on my orange swithces, I think at least.
Also found "paper mods" and its idea totally made sense to me but I didn't give this try in fact because I didn't like to use paper in my Alps for some reason. Instead I used this bending method and worked for me. Idea behind this method is same as "paper mods".

https://www.reddit.com/r/MechanicalKeyb ... _noise_on/
Thank you for that! I actually saw that post when I was trying to figure it out but decided not to bend the leaves in case I messed them up permanently. I already had one mishap whereby a tactile leaf fell out while I was reassembling a switch (on a different board, luckily not one that was as valuable/nice) and the leaf got trapped underneath the housing and folded in half, which I've corrected about as well as I can but still somewhat alters the sound of that switch (only really on the upstroke, it's just a little less crisp and sounds vaguely like a piece of foil stuck in the switch or something like that). In the end, I decided the paper method was less potentially destructive and gave that a bash. I expect that the paper method possibly alters the sound more, and maybe even the key feel every so slightly, but I'd rather that than mess up all my tactile leaves and/or create inconsistencies from switch to switch. At least with the paper method I can cut every piece to be pretty much precisely the same size, and fold them the same way.

On a somewhat unrelated note, incidentally, thank you for your work on TMK, and thank you for making it free software (or open source, if you prefer). I'm looking forward to farting about with various converters in the near future. I've got an M0110A, AEK, AEK II, Sun Type 5c, and some weirdo Apricot XEN-i board with an as-yet unidentified protocol to get going, for starters.

NeK wrote:
24 May 2021, 18:49
I still am not certain of what is the exact cause that results to this tiny dragging of the leaf, but I'm pretty sure the click is caused of that tiny displacement.

I think that it is a combination of the following:

1. before a switch is opened for the first time, the housing is more strongly shut, and thus it holds the leaf tighter so it doesn't hit hard making the up click, once you open it, and you then close it, it may not close with the same tightness, leaving more "room".
2. the missing lube and the friction makes the dragging far more probable.
3. the leaf has bent (with time or with use? who knows) in such a minute way that it is a tiny bit shorter in height, so there is more room left to move.

I guess, IF they are true, #1 and #3 are more important factors than #2. I am still "researching" the bending of the leaf in a correct way and what is actually going wrong with it.

I think that #2 doesn't matter that much, it just helps a little bit.

Once we understand exactly what is the mechanism, then we can say for sure.
If believe you are correct, personally. At the very least, #1 seems to be the case, #3 seems probable, and if indeed all Alps SKCM Orange are factory lubed (I say "if" because I have seen no trace of lube of any kind on any of the switches on this board) then #2 seems likely as well.

Post Reply

Return to “Workshop”