Chinese USB Hall Effect Keyboard - Review and Impressions

[andrewjoy]

Wait ! I know that voice !!!

[XMIT]

Chyros wrote that article.

Uhm, the springs in my boards are cylindrical not conical. Please fix this.

[XMIT]

I just watched the rest of the video. Everything in there is correct including the mention of the Lorentz force. Well done!

[JarateKing]

It was a pretty nice little rundown, props to Chyros

I don't know if it's already been asked but is there any rated lifetime numbers for your switches? I don't think it'd ever wear down beyond function, but what's the number looking like (for science purposes)?

[scottc]

https://m.reddit.com/r/MechanicalKeyboa ... _keyboard/

The reviewer is annoying, but he airs some legitimate grievances about the switch. Your keyboards will likely have better QC, but the scratchiness is there even on my beta board without dampers.

The reviewer is such a tool We get it, the F key didn't work, it would've been abundantly clear to say a sentence without using "F"s, but the rest of it is totally ridiculous. I couldn't watch more than a minute before I had to close it, the guy was just too obnoxious.

Props to you, XMIT, for taking that "review" as constructive feedback and addressing his concerns.

[XMIT]

The manufacturer claims "100 million cycles". This seems wishy-washy. HaaTa and I guessed that they simply took Cherry's 50 million cycles number and doubled it.

When asked, the manufacturer believes that the magnets will demagnetize first. This seems extremely unlikely for rare earth magnets. From K&L Magnetics:

[Neodymium magnets will] begin to lose strength if they are heated above their maximum operating temperature, which is 176°F (80°C) for standard N grades. They will completely lose their magnetization if heated above their Curie temperature, which is 590°F (310°C) for standard N grades.

(source: https://www.kjmagnetics.com/neomaginfo.asp)

I think the sliders will give out first.

[XMIT]

Props to you, XMIT, for taking that "review" as constructive feedback and addressing his concerns.

Note that he declined my offer to be a beta tester, and has not responded to my offers to provide updated sliders and to try to fix the F key for him. I didn't have to do any of these things - he reviewed a board that was not built to my specification and did not go through my quality control process - but I felt like it was the right thing to do so I did.

The ball is in his court if he wants further support.

[ohaimark]

Props to you.

The failure mode for marathon testing would probably be slider wear. Alternately, if the spring spins or vibrates it could gnaw through the PCB. Especially if the switch tester operates at high speed.

If we're guessing failure modes for regular operation I think the electronics would die first. A capacitor, perhaps?

I don't think marathon testing Hall Effect switches provides useful information in terms of actual keyboard life.

Here's an interesting thought... What happens when the springs get magnetized? Will the actuation point change in a noticeable way?

[XMIT]

These don't have electrolytic capacitors, just little solid state tantalum ones. Those won't fail any time soon.

For endurance testing - I was thinking of putting some of my spare prototype boards in a wood or metal shop for a year or so just to see what happens. I bet that metal dust would make the switches really scratchy, but that's about it.

I think that the magnetic field of a magnetized spring will be orders of magnitude less than the strength of a rare earth magnet. Now, I haven't actually done the math to show that this is so, but, rare earth magnets are really quite strong and you're talking about a really small amount of magnetic material in those springs. It could be a fun exercise to anyone who wants to try to work out the math but I've got better things to do at the moment.

[vivalarevolución]

Got an update from Massdrop. Looks like it will be shipped soon. I'm excited!

[XMIT]

Got an update from Massdrop. Looks like it will be shipped soon. I'm excited!

Oh really? I didn't see that update yet! But the manufacturer told me these should be getting on a plane on the 27th. I'm excited too!

[edit] Heh, that's because I didn't actually order any boards through Massdrop. I get mine direct from the manufacturer. I order some weird stuff and don't want my order confused with a Massdrop order.

In case you're curious, here is what a clear/acrylic 87-key board looks like with Round 6. Works great!

XMIT169 top view.

XMIT_0826.jpg (155.55 KiB) Viewed 5869 times

XMIT169 side view.

XMIT_0828.jpg (71.83 KiB) Viewed 5869 times

[Chyros]

Damn. Might have to get myself a set of round 6. Somehow xD .

[XMIT]

Damn. Might have to get myself a set of round 6. Somehow xD .

Chyros, this one *does* have an XMIT key.

[fohat]

Might have to get myself a set of

Yes, these are begging for really good caps, but group buys have always seemed too convoluted and forbidding for me to figure out and/or participate in, and the "aftermarket premium" is pretty distasteful too.

I would be a lot happier just to purchase something outright.

[XMIT]

Might have to get myself a set of

Yes, these are begging for really good caps, but group buys have always seemed too convoluted and forbidding for me to figure out and/or participate in, and the "aftermarket premium" is pretty distasteful too.

I would be a lot happier just to purchase something outright.

So, that's part of the long term plan.

I'm exploring two options:

1. Making stock sets of Signature Plastics SA caps. This would be a 104-key ANSI set with additional media, ISO, and modifier keys for a few different layouts, white on black, row profile likely 3-1-2-3-4-3, based on Modern Selectric. Nothing super fancy - just a very nice stock spherical key cap option. These would probably be about $150 a set, haven't explored pricing yet.

2. Whatever matt3o is up to. I'd need to get my hands on a sample before ordering more.

Lots to think about with the next offering!

[zslane]

That's a very cool idea, XMIT. Personally, I'd be more excited about 1-1-2-3-4-3 (don't know why anyone wants to reach for those F keys more than they have to...R1 is really the way to go, I feel) and maybe a very dark gray instead of black, like GB or GX. I think that would give the set a more classic look, rather than just following the same old boring WoB aesthetic that afflicts every cheapo keyboard in the world today.

[need]

Might have to get myself a set of

Yes, these are begging for really good caps, but group buys have always seemed too convoluted and forbidding for me to figure out and/or participate in, and the "aftermarket premium" is pretty distasteful too.

I would be a lot happier just to purchase something outright.

So, that's part of the long term plan.

I'm exploring two options:

1. Making stock sets of Signature Plastics SA caps. This would be a 104-key ANSI set with additional media, ISO, and modifier keys for a few different layouts, white on black, row profile likely 3-1-2-3-4-3, based on Modern Selectric. Nothing super fancy - just a very nice stock spherical key cap option. These would probably be about $150 a set, haven't explored pricing yet.

2. Whatever matt3o is up to. I'd need to get my hands on a sample before ordering more.

Lots to think about with the next offering!

I think people would prefer another round for space cadet instead of white on black

[Laser]

Signature Plastics started to make PBT SA keycaps; the problem is, they are row 3 only - the whole set. I wonder, wouldn't it be possible to design the keyboard already curved (i.e. in the manner of buckling springs ones, Models F or M), so that one could use row 3 keys everywhere but have, in effect, the right keycap angles on each row? I guess the question is - could a (new version of) XMIT's Hall Effect keyboard (and PCB?) be *it*?

[zslane]

SP is working on dye-sub assembly boards for fully sculptured PBT SA keysets. Just gotta be patient.

[XMIT]

ABS is not my favorite material and SA is not my favorite profile. With no constraints at all I'd re-make Alps spherical PBT dyesubs. Given my constraints I'm going to wait and see what matt3o has to offer.

[Hypersphere]

Looking forward to a Round 2.

[vivalarevolución]

Got an update from Massdrop. Looks like it will be shipped soon. I'm excited!

Oh really? I didn't see that update yet! But the manufacturer told me these should be getting on a plane on the 27th. I'm excited too!

[edit] Heh, that's because I didn't actually order any boards through Massdrop. I get mine direct from the manufacturer. I order some weird stuff and don't want my order confused with a Massdrop order.

In case you're curious, here is what a clear/acrylic 87-key board looks like with Round 6. Works great!

XMIT_0826.jpg

XMIT_0828.jpg

i guess shipped from the manufacturer soon would be more accurate. Anyways, any update is a good update, in my view.

[XMIT]

Yeah. My expectation is that these will land on the 29th, take a week to get through customs, another week to be broken down and shipped, and another week in transit. Sorry if that's not as awesome as you had hoped! I think we'll get faster turnaround next time now that we know what we're doing.

[bloo]

Any possibility of seeing clicky/tactile variant for round 2?

[XMIT]

Any possibility of seeing clicky/tactile variant for round 2?

We're still working on finding a manufacturer that can crank out consistent click and tactile leaves, but this is something we're striving for, yes.

[andrewjoy]

Show us some wood !

[bloo]

Any possibility of seeing clicky/tactile variant for round 2?

We're still working on finding a manufacturer that can crank out consistent click and tactile leaves, but this is something we're striving for, yes.

Good to hear.

[Elrick]

Show us some wood !

YES, like desperate perverts we want to see it in all it's glory .

[XMIT]

I'll get an updated prototype with my personal order which should be here in about two weeks.

[andrewjoy]

I'll get an updated prototype with my personal order which should be here in about two weeks.

No excuse , i want to see XMIT's wood!