ITT Courier review (Honeywell Hall effect D1B3S)

[Chyros]

I was up until 5:30 AM yesterday making all the pictures for this review, but I'm pleased with the result ^^ . Hoper you enjoy the video!

[Crazy Canadian XXIV]

YEAH, BITCH, MAGNETS!

That's gotta be quote of the year.

[derzemel]

I want a keyboard with Hall Effect switches now

[Madhias]

Great video, and the explanation graphics were it worth staying up so long!

[Scottex]

See this here?, this is the smoothest switch in the world

Reservoir dogs reference maybe?
Great review as always

[scottc]

Chyros, thank you for the fantastic explanation of how the switches work! I'm no scientist, but I find this sort of stuff incredibly interesting and you explained it with such clarity. I'd love more of that kind of thing in the future!

[Chyros]

Thanks guys, glad it was worth the effort .

See this here?, this is the smoothest switch in the world

Reservoir dogs reference maybe?

Haha, no, I didn't even think of that. Would've been a good one though xD .

Chyros, thank you for the fantastic explanation of how the switches work! I'm no scientist, but I find this sort of stuff incredibly interesting and you explained it with such clarity. I'd love more of that kind of thing in the future!

I've got something like this for another very interesting switch planned for you guys in the far future, after the Alps trilogy .

[seebart]

Finally, your first "older" vintage keyboard. Very nice. Now you know how smooth Honeywell Hall Effect is and believe the fact that myself and many others here aren't mental when we continue to rave about these and proclaim that these dwarf any younger mechanical switch in every way. Your keycaps look very much like the sphericals on my Honeywell Hall Effect keyboard. Nicest keycaps I own. I do have to break it to you...the predecessor Micro Switch Hall Effect feels even nicer and I'm not exaggerating! The only clicky counterparts of that era that I own (IBM and Fujitsu) are a whole other story.

[Engicoder]

Another great review.Beautiful key caps! A historical note: ITT Courier was originally just Courier Terminal Systems and was the one of the largest manufacturer of IBM compatible terminals that offered a lower cost option. Courier was purchased by ITT in March of 1978.

My guess is this is from one of their 3278 compatible terminals.

[courtesi]

I'm confused - that dust would cause issues for the switches yet these are completely sealed? So how would dust get in there?

[Redmaus]

Oh yeah I know Hall Effect when I tried out XMIT's Texas Instruments board and it was not just smooth but powerful.

That thock knows no bounds.

[Chyros]

I'm confused - that dust would cause issues for the switches yet these are completely sealed? So how would dust get in there?

Good question. They're sealed as in they can't be opened. But dust can definitely still get in.

[seebart]

As I understood dork vader there are some that are completely sealed vs. the normal kind but I don't know that really since I only own one Honeywell Hall Effect keyboard. But Chyros since they cannot be opened how do you know that there's dust inside?

Here's dork vaders' shots of the internals:

https://www.flickr.com/photos/dork_vade ... 2093794783

I added that link to our Honeywell Hall Effect wiki page with some new pictures:

wiki/Honeywell_Hall_Effect

[bocahgundul]

are there someone that can use this board in usb? or actually using this board?

[Chyros]

As I understood dork vader there are some that are completely sealed vs. the normal kind but I don't know that really since I only own one Honeywell Hall Effect keyboard. But Chyros since they cannot be opened how do you know that there's dust inside?

I actually have another Honeywell Hall effect keyboard. It is not ready to be shown yet. It was quite dusty when I got it however, and the switches feel much scratchier than those on the ITT keyboard, hence why I got it. I'm experimenting with getting the dust out of them at the moment.

[Chyros]

are there someone that can use this board in usb? or actually using this board?

No, this will have to be reverse-engineered first.

[bocahgundul]

are there someone that can use this board in usb? or actually using this board?

No, this will have to be reverse-engineered first.

Ah yeah. Someone genius needs to reverse-engineer this board

I really love the sound of this board

[seebart]

I actually have another Honeywell Hall effect keyboard. It is not ready to be shown yet. It was quite dusty when I got it however, and the switches feel much scratchier than those on the ITT keyboard, hence why I got it. I'm experimenting with getting the dust out of them at the moment.

Great can't wait to see your second one also.

[Halvar]

Superb video once again, thanks!

As this keyboard has the standard IBM 3270 terminal layout, I wonder if it maybe also uses the same protocol as the IBM beamspring boards. Didn't someone somewhere even make a converter for that? I just remember that the protocol was pretty poor, like you couldn't get key-up events for most keys except modifiers, which was one of the reasons for xwhatsit to make his own controller for beam spring boards.

[XMIT]

You certainly can't get key-up events for most Hall switches, the ones that "pulse low". I discuss this in detail on a separate thread:

Honeywell Micro Switch Reverse Engineering and Conversion thread
workshop-f7/converting-my-wang-t12379.html

[Halvar]

Oh wow, thanks XMIT for linking me to that thread, don't know how I could miss what you were up to in that thread the whole time. I have one Micro Switch and one obscure Hall effect keyboard at home, so I just read the whole thread with great interest. Bummer about the missing key-up signal!

[XMIT]

Bummer about the missing key-up signal!

Yeah. The board can fake NKRO with a fast poll rate but there is a real chance of collisions during the 50µs sense window.

I did experiment with a power on strobe that HaaTa mentioned before. That's the best chance we have of getting reasonable behavior out of these switches. It would also take a new PCB and some work, quite a retrofit for folks who are looking for just a USB converter.

Still, the experience my converter will provide will be no worse than the original experience on these boards when connected to their terminals.

[XMIT]

Mine now.

We negotiates a "starving grad student" discount: I sent Chyros some very nice boards for future reviews and he sent me this. You all should have some very nice reviews to look forward to. I on the other hand have intentions to convert this one to USB as well. Hopefully these switches are all the "hold low" variant - that makes life so much easier.

[XMIT]

are there someone that can use this board in usb? or actually using this board?

No, this will have to be reverse-engineered first.

Ah yeah. Someone genius needs to reverse-engineer this board

I'm no genius but I'll see what I can do.

[XMIT]

From the video: "A Schmitt trigger is basically an analog to digital converter".

Not quite.

The key property of the Schmitt trigger is not its activation voltage or switching property, but hysteresis. It provides some "wiggle room" between low to high and high to low transitions. In effect, it helps to debounce the key switch, and/or makes it so that the actuation and deactivation point of the switch are not at exactly the same spot.

Also, the disgrams around 3:45 only demonstrate "hold low" and not "pulse low" switches.

[XMIT]

I also need to disagree with the "unlimited rollover" comment around 6:56 in your video, sorry. Not because it's not possible (it is), but because most boards are not built this way.

The typical Hall key switch has four pins: power, ground, and two identical sense lines. Regardless of whether they are "hold low" or "pulse low" the two sense lines are typically tied to a row and a column each.

I explain this in more detail here:
workshop-f7/converting-my-wang-t12379-60.html#p284261

In short: as built these boards are typically 1KRO because, when any two keys are pressed, you have no way of knowing if two or four keys are pressed at the same time, since you're sensing rows and columns.

The key idea with a switch matrix is that there are strobe (output) and sense (input) lines.

The way around this is to strobe power to switches, using that as the input, as HaaTa proposed. I've not seen a board that was wired this way.

[XMIT]

Around 8:03 you say that these switches are impossible to open.

This is true. But, there is enough space for dirt to get in and out. I've had good results cleaning them in an ultrasonic cleaner with some agitation. When I did this, I pulled out the Hall sensors for good measure, not knowing what the ultrasonic cleaner might do to fairly delicate sensing hardware. It is well shielded for sure, I didn't want to take any chances though.

[Chyros]

Around 8:03 you say that these switches are impossible to open.

This is true. But, there is enough space for dirt to get in and out.

The key property of the Schmitt trigger is not its activation voltage or switching property, but hysteresis. It provides some "wiggle room" between low to high and high to low transitions. In effect, it helps to debounce the key switch, and/or makes it so that the actuation and deactivation point of the switch are not at exactly the same spot.

I actually mention both of these things in the video .

I also need to disagree with the "unlimited rollover" comment around 6:56 in your video, sorry. Not because it's not possible (it is), but because most boards are not built this way.

The typical Hall key switch has four pins: power, ground, and two identical sense lines. Regardless of whether they are "hold low" or "pulse low" the two sense lines are typically tied to a row and a column each.

I explain this in more detail here:
workshop-f7/converting-my-wang-t12379-60.html#p284261

In short: as built these boards are typically 1KRO because, when any two keys are pressed, you have no way of knowing if two or four keys are pressed at the same time, since you're sensing rows and columns.

The key idea with a switch matrix is that there are strobe (output) and sense (input) lines.

The way around this is to strobe power to switches, using that as the input, as HaaTa proposed. I've not seen a board that was wired this way.

Hmmm, yeah, I understood you guys are getting 1KRO out of it at the moment, but surely it would've had NKRO in its native environment, right? I mean I don't see why it wouldn't Oo .

[Muirium]

I was about to blame it on the matrix — rollover is a chain consisting of the matrix >> controller >> protocol, and the result is the value of the weakest link — but then searching DT for previous discussions brought out Xwhatsit (the beamspring saviour!) who says the matrix is not a source of rollover limits in Honeywell:

Know if they still need diodes for an NKRO matrix? I hear my Honeywell has poor rollover, but I don't know if this is the matrix or controller's fault.

Being open-collector outputs, no you won't need diodes—you only need that on standard `ohmic switching' (IBM patent language ) matrices, as you end up with a conductive loop which can short-circuit some keys and make them appear live.

Those Honeywell switches (gosh they must have been high-end at the time!) therefore support proper NKRO, the same as a capacitive matrix, so if your controller supported it you could hold down all the keys and have them all register individually, just the same as the USB beamsprings.

Very cool! Must have cost a fortune at the time.

So I guess it's possible. But maybe only by sawing off the original controller.

[XMIT]

So I guess [NKRO is] possible. But maybe only by sawing off the original controller.

Yes. With some effort the switches support NKRO. I have not seen a PCB and/or controller for Honeywell switches that does.