Topre slider variants and their differences

[the_marsbar]

I’ve read a couple of reports about variations in the slider length of the HHKB Type-S. Some people claim (e.g., here (reddit), here (geekhack), and here (this forum) that the HHKB Type-S sliders are longer than the ones in the Pro 2 (to compensate for the silencing material I guess), however, on PFUs own website, the key travel is listed as 4.0 mm for the Pro 2, and 3.8 for the Type-S.

In the two first references above, it sounds like a change might have been made to the Type-S at some point, introducing the longer sliders. My question is then; Is it possible to tell the slider length (or which keyboard version) one has, e.g., by looking at the production date? If this has already been answered somewhere else, forgive me, and I would like to know about it (post a link to the information). I guess it could also be the case that PFU made a mistake on their website (I'm leaning towards this; here (imgur) is a gallery showing differences in slider heights (Novatouch vs. Pro 2 vs Type-S).

[Nuum]

The length of the slider influences the keycap height, not the travel of the switch. The height of the bottom of the slider above the PCB corresponds to the switch travel, which get's reduced by the silencing rings.
So both the reduced travel and the longer slider is correcr for the Type S.

[the_marsbar]

Would that mean that trimming the feet of the sliders in the Type-S would result in something similar to a purple slider from a silenced Realforce?

[the_marsbar]

Sorry for bumping this - I'm still curious if trimming the feet of the sliders will result in increased travel, e.g., to compensate for installation of silencing material.

[Laser]

I also wonder if we aren't missing some other elements - is it possible that the type-S slider housing is also taller?

Adding the ring lowers the slider a bit and, since it's said that Topre tactility happens "at the top", it's possible that the slider pushes on the dome exactly enough to remove some of that precious "top tactility". Trimming the legs can't solve this, raising the housing (somehow ...) would do it. Squared-shape housing rings, anyone?

Or maybe thinning the dome-contact surface of the slider? -> This goes well with this statement: "On the right is a silent Realforce slider, which has approximately 1mm of relief molded into the bottom. We can observe a similar molding detail in the HHKB Type-S slider" This also gives the answer, the housing is not taller, but the inner bottom of the silenced slider stands higher, so to speak, to compensate for the ring.

Therefore, to achieve a type-s slider, one should 1) shorten its legs and 2) create a 1mm of relief into the bottom of the slider (i.e. make the dome-touching base thinner from below, if possible ...)

EDIT: the "spacer" solution (rising the housing): https://imgur.com/a/DLuOw

[the_marsbar]

Thanks so much for that. That looks like a neat solution. I guess the gap between the base and the house will increase by the same amount, but it might be possible to fix that.

The good thing about a solution like this is that it can be undone later. I like that.

[Laser]

I guess the gap between the base and the house will increase by the same amount, but it might be possible to fix that.

Ah yes, that was what I was describing in the PM I've sent you

[the_marsbar]

I tried the paper spacing trick, which was suggested in the imgur post that Laser referred to.

Tactility has definately been restored, but the dampening effect has lessened, which I don't like. I also don't like that the keys sit slightly higher.

I think my next step will be to get a silenced Realforce 104 and harvest the sliders. There should be almost enough for two HHKBs, if the top row is left stock on one of the boards.

Another option would be to remove some material from the bottom of the slider, but that would be very time-consuming, and I don't think I'd be able to produce consistent results with the tools I currently have.

[the_marsbar]

After I made the modification I was a bit surprised at the increase noise level. My hypothesis is that since the circuit board was raised a bit, maybe it was raised so much that some of the keys hit the plastic. I just tried removing parts of the straight pieces of paper (leaving only one layer, and keeping the paper where the screws are), and I think it's much better now.

This makes me this that the amount of spacing needed has to be adjusted very carefully, which is why using washers might not be a good idea. It's much easier to control the amount of spacing using layers of paper.

[mog_genius88]

Ive owned 5 brand new type S'. 4 of them were models made in 2017 and one was from 2015 or 14 (i forget since i dont have that one anymore).

The stems on the 2014/15 model are shorter than the stems in the 2017 models. The type S made before 2017 likely have 3.8mm distance due to the stems not being adjusted for the o rings. The 2017 models i have had the stems adjusted in length so they (edit) feel like they have the same 4mm travel as the pro 2.

[the_marsbar]

Thanks for that. Any chance you have some evidence to back that up? Photos? My Type-S is from 2015-12.

[mog_genius88]

Thanks for that. Any chance you have some evidence to back that up? Photos? My Type-S is from 2015-12.

Sorry no photos. I no longer have the older type S and i will not be reopening the 2 i still have as ive already opened and closed then several times.

Hopefully someone else with a post 2017 type S can post a picture

[the_marsbar]

I hope so. It would be really awesome to see the difference. At least I'd really like to see it

[mog_genius88]

I hope so. It would be really awesome to see the difference. At least I'd really like to see it

There was a side by side comparison shot someone already. Im currently looking for it

[the_marsbar]

What I have seen is a side-by-side comparison of a slider from a Type-S (not stated if 3.8 mm travel slider or 4 mm travel slider) and a regular HHKB Pro 2 slider here (same link that Laser posted above).

Is that what you're thinking about? I've spent some time looking, and I haven't been able to find any real proof of different versions of the sliders in the HHKB Type-S (including any pictures showing the differences).

[Laser]

I know this is a quest for perfection, so I'm also a bit curious, but (to chime in again) I think that the most important problem to be solved was that tactility loss at travel start, and that 3.8mm vs 4mm (5%) doesn't affect the typing feeling *that* much, relatively speaking.

[mog_genius88]

What I have seen is a side-by-side comparison of a slider from a Type-S (not stated if 3.8 mm travel slider or 4 mm travel slider) and a regular HHKB Pro 2 slider here (same link that Laser posted above).

Is that what you're thinking about? I've spent some time looking, and I haven't been able to find any real proof of different versions of the sliders in the HHKB Type-S (including any pictures showing the differences).

I think that was it but yes... the type s i purchased after 2017 all looked like the purple slider where the slider + o ring ended up being the same height as the stock pro 2 slider.

If you have a regular pro 2 you can compare the slider on you 2015 type s to a slider in the pro 2.

[the_marsbar]

Yesterday I took apart the two keyboards. I took out a slider from each keyboard, took a few photos and took some measurements with a digital vernier caliper (brand unknown, probably not the most accurate). I also took some measurements with a regular (analog) vernier caliper. I took photos of the readings on both of them. A couple of photos (notice that the Type-S slider has more "rounded" edges when seen from above:



Here are the measurements of the sliders:
- Pro 2 slider dimensions: 11.40 mm (total height), 8.90 mm (height of the cylindrical part)
- Type-S slider dimensions: 11.59 mm (total height), 9.20 mm (height of the cylindrical part), 0.32 mm (depth of the notch under the slider)

I also measured the thickness of the silencing rings, but those numbers should naturally be taken with a grain of salt (I tried not to compress the rings when measuring their thickness):
- Hypersphere silencing ring height: 0.40 mm
- Type-S silencing ring height: 0.41 mm

Conclusions: The slider from my Type-S keyboard is ~0.2 mm longer than the one from my Pro 2. The difference in the height of the cylindrical parts of the sliders is ~0.3 mm, which is also the depth of the notch under the cylinder of the Type-S slider. My guess is that this is needed to avoid the keycaps hitting the slider housing when fully pressed. Another thing I noticed; the silencing rings have similar thickness, however, the rings installed in the Type-S are much softer, and I guess that's what causes the sound to be more thocky on the Type-S (I did not apply lube in any of the keyboards).
I'm still not sure whether I have sliders with 4 or 3.8 mm travel in my Type-S (which measurement determines the travel?). It would be really cool if other people with newer and/or older Type-S keyboards could do the same measurements.

If I at some point manage to get hold of a Realforce with purple sliders, I will also make sure to take out one of the sliders and do the same measurements.

Update: I am now using two sheets of paper for spacing the PCB and the top part of the case. The sound is as nice as without the sliders, but the key presses feel more crisp. So far I’m happy with the modification.

[the_marsbar]

A small update in case anyone would be interested; I have a Realforce 104 on the way, and I also ordered tester packs with all the different BKE Redux domes, so I guess I will be doing more comparisons than I first thought. First thing I'm going to do is measure the purple sliders and compare them to the two other ones. I'll probably remove one from the Novatouch that I have too.

[Laser]

Cool, ordering popcorn

[the_marsbar]

I did some measurements yesterday, which I will have to write a bit about at a later time.

Regarding the BKE Redux domes, I tried installing lights on the spacebar and Enter on a HHKB, ultra light on Fn, and heavy on Esc. After using the keyboard for 15 minutes, my initials thoughts are: yikes, those things feel weird. I only like the ultra light to far. The other domes are really different in terms of the weight compared to the stock HHKB domes. I'll probably try them in the realforce as well, to see if it's better there.

[Laser]

I also modded my Novatouch (no pics, alas) by cutting scotch tape alongside in half-wide long strips, and putting (sticking) each strip on the dome sheet, *between* the dome rows, so that the novatouch housings would sit a bit higher when placed on the dome sheet/PCB. I had high hopes on this mod, but the result is *just* a little better than before - I'll probably try again with some thicker scotch tape. Or maybe the key is to ensure the extra distance on the holes around the screws, as in the original mod posted above (imgur).

[the_marsbar]

In theory that should work, and have the same effect as using sliders with indentations. However, it is not trivial to determine how much the slider housings should be raised relative to the PCB. It depends on the thickness and the hardness of the material used. E.g, the silencing rings used in the silent HHKB and Realforce keyboards is the same thickness as Hypersphere's silencing rings, but they are much softer.

An issue that may arise is that if you raise the housings too much, the keycaps may start to hit the housings, which will be a plastic-against-plastic sound.

[Laser]

The funny thing is that I returned from work and, typing again on the modded Novatouch, I find it much better than yesterday right after modding it

[the_marsbar]

I'm sorry for the late update. I was thinking about how to present my findings (in particular, making a drawing that would explain what I measured). I took measurements of sliders from a HHKB Pro 2 (black), HHKB Type-S, CM Novatouch, and a Realforce 104 (purple/silent sliders).

Here's a figure showing the slider and where I measured (E points to where there's a notch in the silenced sliders):


Here are the actual measurements (I measured the thickness of the silencing rings on the HHKB Type-S and Realforce 104 to be 0.4 mm):


The measurements should of course be taken with a grain of salt (the numbers in the table represent single measurements of one slider from each of the keyboard). I might do some more measurements at some point, if I manage to get hold of a better set of vernier calipers (I borrowed my dad's electronic one, probably a cheap unrealiable one). If anyone has suggestions on anything that could be improved, I'm happy to hear them.

[Laser]

That's pretty interesting! Especially the differences between the 4 types of sliders. It could be interesting to add a 5th type, the slider of Plum/Niz Topre-clones. Also now I start to wonder (I forgot if this was discussed/settled), are the slider housings the same? I'm looking at the differences of the slider leg heights (C in the figure).

Also, comparing the leg height (C) between Novatouch and HHKB Type-S (which are pretty similar!) it seems the Novatouch slider has a nice chance of being modified (-> preserving tactility when silenced) by creating an artificial notch (E). I'm thinking at two ways (if I'm repeating myself, sorry):

- heat based: using a thumbtack stuck in a slim wood stick, the thumbtack sharp end heated from below, and the Topre slider pressed upon its heated head to create the notch (the wood stick would be so slim so that the slider legs would go around it - riding the stick, so to speak )

- drill based: having an electric drilling machine placed vertically, fixed position, with the drill head sticking out less than 1mm (ideally, 0.4mm - E in the table) from a (also fixed) horizontal, holed wooden stick. Press the slider on the wooden stick, the drill head carves the notch at the right depth.

The risk probably revolves around the cruciform plastic interior of the Novatouch slider - depending on how it's stuck to the slider interior walls, it may (or may not) come off, or weaken, when creating the notch.

[the_marsbar]

Also now I start to wonder (I forgot if this was discussed/settled), are the slider housings the same? I'm looking at the differences of the slider leg heights (C in the figure).

I was thinking exactly this, when I first looked at the measurements. I'll have to open the boards up again at some point. It's a while since I used my Type-S keyboard, but as far as I recall, it doesn't feel/sound like the Pro 2 with purple sliders...

[Laser]

What seems to be missing, at the point we're at, is somebody like Matt3o (or other members, from those active in the Workshop area) to come up with a custom, silent (i.e. adequate leg length, and with the 0.4mm notch present) Topre-to-MX slider solution. Something that goes well with Topre keyboard slider housing, and goes well with the Plum keyboard slider housing - if it's possible.

[Muirium]

Is there something wrong with these? Besides the hideous colour!


https://kbdfans.cn/collections/electros ... -adapter-x

Dawdling over my Topre mount keycaps order there, a few other things of theirs caught my eye. These come with damping rings, for a price. I wonder what happens when they are fitted. The caps run low? Are there collisions when you bottom out?

The other thing is their Bluetooth 60% PCB but that’s a whole other topic.

[Laser]

Those are better than the JTK ones, but worse than the Novatouch ones (feeling-wise). Faults: (1) those intended for the larger keys have to be filed in the corners to fit, (2) at least my batch has a very small angle (when mounting keycaps, they are very slightly rotated to the right, if I remember correctly), aaand (3) the stems are a little too thin, i.e. the keycaps are a little bit loose - again, that happened on my batch, I don't know if all are like this. Also they are not fit for silencing rings (no internal notch), and probably the legs should be shortened too in that case. That doesn't mean one can't use the KBDFans very-thin-silencing-rings, but *some* tactility is lost, in my experience. There are some who say that they don't feel any tactility loss, however.