Author Topic: Wyse keyboard protocol (4P4C)  (Read 13194 times)

0 Members and 1 Guest are viewing this topic.

Offline Soarer

  • * Moderator
  • Thread Starter
  • Posts: 1918
  • Location: UK
Wyse keyboard protocol (4P4C)
« on: Sun, 17 November 2013, 17:00:25 »
Introduction

Many Wyse keyboards share the same 4P4C modular connector, and appear to have more similarities than differences in their protocols.

We should be able to work them out, and convert them :D

Background Info

Much of this background information is a summary of what has been posted in this thread.

This is the most common connector:



But then this unusual 10-pin DIN is used on the WY50 and WY75:



The WY50/75 is clearly a different protocol, and we'll have to examine that separately.

There's a cross reference for keyboard compatibility here. From that list it appears that there are four main groups...
  • Wyse 55, 55ES, 60, 65, 99GT, 120/150/160/185, 285, 325, 355, 370, 520
  • Wyse 30, 30+, 35
  • Wyse 50, 50+, 75, 350
  • Wyse 85
Somehow the 'ANSI' only supports Wyse 60, 99GT (a subset of the big group), the 'Wyse 285-520 Keyboard' only the 285 and 520 (again a subset of the big group), and the 'ANSI Gate Array' doesn't support Wyse 60 (but supports the rest of the big group) - so those differences must be something other than protocol, surely, but what?

Well, WY50 etc. use the 10-pin DIN and use completely different protocol. WY30 keyboard looks from its circuit diagram to be basically compatible with the big group, but maybe lacks some required keys. It's possible that the WY85 keyboard (as opposed to the 85 Gate Array keyboard) didn't use the scanner chip, but still has a protocol that's compatible with the first 128 clocks of the scanner chip's protocol. I need confirmation from someone with a 840105-01... I might have one, but it's lost its label so I can't be sure!

Here's the signaling diagrams from kbdbabel that akurz made:





It's possible that akurz didn't have complete information when he drew up these diagrams - the WY60 at least, doesn't seem to mind getting more or fewer clock pulses than required and simply ignores any excess, so it's possible that both PCE and WY85 use 160 clocks, and/or that the last few are always reserved for LEDs even on keyboards that don't have them. The output sequence appears to be reset by clock staying low for greater than a certain amount of time (akurz indicates 1ms).


Test Code

I've written a small test program for ATmega32U4 (e.g. Teensy 2.0). It provides a sequence of 160 clock pulses to the keyboard, as in the WY85 diagram above, and prints what comes back to the debug output (visible using hid_listen). It only prints when the data changes to reduce the amount of output.

Testing with a WY60 ASCII reveals that this protocol 'works' for it too. As in, the data output is readable and affected by pressing keys! I don't have a WY85, but I do have a PCE that I'll test soon - which will test setting the LEDs.

The WY60 seems to identify itself by always setting positions 144 and 150 low.

The chip inside the keyboard is clearly little more than a shift register. Holding clock low for more than 34us will reset the output sequence to the start. Otherwise, the one in the WY60 (at least) is very flexible in terms of timing. It can run as fast as I've been able to test so far - about 3.5us per bit for a total scan time of 565us - but could go faster, I'm sure. The only critical thing is to give it a clock pulse that's not too narrow. This is good news for writing a converter, since it means the reading doesn't have to be done on a timer interrupt, or even with interrupts disabled (apart from when writing the clock pulse low-then-high).

Here's the source and compiled hex - note this isn't a converter, just a test!

* wyse_test_1_src.zip (15.57 kB - downloaded 131 times.)

* wyse_test_1_atmega32u4.hex (5.56 kB - downloaded 86 times.)

Pins used are the same as my converter - PD0 for Data, and PD1 for Clock.

If people with keyboards that I haven't got could try it out and report back, that would be cool! Obviously the first important piece of information is whether the test program works at all with a particular model, and shows which keys you're pressing. Then the goal is to fill out this table with IDs (from the second to last byte) and links to matrix layouts:

The ID can be found without even running the test program, just by looking at the back of the PCB. Follow the track from pin 21 of the 28 pin scanner chip to the diode(s) close by, and then from the other end of each diode until the trace returns to one of pins 1 to 8 on the scanner chip. Of course, using a multimeter or continuity tester makes this even easier!


Results

There's a clear progression from WY50 though WY85 to the rest. WY50 takes a 7-bit parallel address of the key to be read and outputs the key's state a microsecond or two later. WY85 adds a counter on the front of that so the interface becomes just clock in and data out, and resets the counter when clock is held low for more than about 34us. The rest package an enhanced version of that circuit into a scanner chip which supports a 14x8 matrix, one column of which is used for an ID byte leaving 104 positions for switches, and (combined with an external circuit) also supports LEDs.

The scanner chip's protocol can be summarised as having 18 bytes of key data (of which the first 13 are used, but see note 4), then 1 byte of ID code, then counting clocks beyond that to set the LEDs. Data output changes to the next bit on the high-to-low transition of the clock input.

ModelKeysPart Nos (old, new/white)ID (hex) Matrix
3083 -1840013-01, (900023-01?)01reply #21
50 ASCII101 -1840059-01n/a 10WY50/75 info
60 ASCII101840338-01, 901867-0141reply #2
60 ANSI101840338-0941reply #2
60 3161106 3840338-0204 (or 05/06/07) 3reply #14
75 ANSI101 -1840059-03n/a 10WY50/75 info
85105 -1840105-01n/a (00) 5same as 85 Gate Array 5
85 Gate Array 1105 -1840366-01, 901879-0160 6clavis/projects/wyse-840366-01
285/520 ANSI108 4(841038-01), 901028-0162intealls' WYSE terminal board mod
285/520 ANSI WPS 2108 4(841038-06), 901028-06???? same as 285/520 ANSI ??
AT Standard84840275-0402reply #14
PCE US, Grey102840358-01, (901865-01)82 7reply #7, clavis/projects/wyse-840358-01
PCE Int'l, Grey103840362-01, (901866-01)83reply #7
PCE US, White102840358-1382 7reply #7
PCE Int'l, White103840362-0483reply #7
PCE US 8102(841135-01), 901865-0182 7reply #7
PCE Int'l 8103(841135-02), 901866-0183reply #7
1100 9101 ??840114-01n/a 10?? same as 50 ASCII ??
2200 9 '286 Style'84 ??840275-01???? same as AT Standard ??

Notes:
-1. The shift keys are not independant; they are simply wired in parallel. So the matrix has one fewer positions than keys.

1. 840366-01 is listed as 'ANSI Gate Array' on the vecmar site, but has 'WY85 Gate Array' on its label.

2. Some resellers distinguish this as: "Yellow key cap style".

3. ID is 04 with no keys pressed. Keys F15 and F16 are read from bits 1 and 0 of the ID byte, so the ID can read as 05, 06 or 07 if either or both are pressed.

4. The 285 has some extra circuitry to scan another column of keys in addition to the scanner chip. They are output in the 14th byte of the bit stream.

5. Assuming that the keyboard I am testing here is an 840105-01. It does not have the scanner chip, just a few TTL chips. Its matrix matches 840366-01, and it outputs the same as 840366-01 for the first 16 bytes. After that, it repeats the same output, i.e. byte 16 = byte 0 etc. The ID byte will therefore read as 00 only if no keys are pressed in byte 2 (up, numpad 6, numpad star, c, 0, w, backslash, F3), otherwise it could be any value. Note also that pressing other keys in combination could cause ghost key presses in byte 2.

6. Assuming akurz's "WY85" was actually a WY85 Gate Array.

7. By tracing PCB tracks on the pic here.

8. Later revision of PCE US and Int'l have different part numbers, but are drop-in replacements and therefore must be functionally identical.

9. WYSEpc model numbers, not necessarily keyboard model numbers. WYSEpc keyboards with 4P4C connector are believed to use the same protocol as the terminals.

10. Have a 10-pin DIN style connector. See WY50/75 info for interface details.

901866-02 is K/B, IEPC, FRENCH, so perhaps the suffix only indicates language in this case.


TBC

Still require confirmation of the following:
  • Anything marked '??' in the table above
  • WY30 matrix
  • ID code for WY85 Gate Array
  • ID code for PCE US (ANSI Enter)
  • That the unlabeled keyboard I have here is indeed a WY85 p/n 840105-01
  • WYSEpc keyboard info
« Last Edit: Tue, 07 January 2014, 10:45:33 by Soarer »

Offline Daniel Beardsmore

  • Posts: 1806
  • Location: Hertfordshire, England
    • Boring twaddle
Re: Wyse keyboard protocol (4P4C)
« Reply #1 on: Sun, 17 November 2013, 19:11:44 »
I've made you a wiki page for your research, and I've linked to it from the Wyse 50 keyboard and Wyse ASCII pages:

http://deskthority.net/wiki/Wyse_keyboard_protocols

The page also links back here so that people know where to find the discussion.
Bore Awards
Most Boring Person on the Planet 2011 Winner

Offline Soarer

  • * Moderator
  • Thread Starter
  • Posts: 1918
  • Location: UK
Re: Wyse keyboard protocol (4P4C)
« Reply #2 on: Tue, 19 November 2013, 16:07:06 »
I'll just throw info in here for now, until it's more worked out. :thumb:

Here's the 'keyboard scanner' chip, from the WY60 schematics...

44940-0

Pin 19, called 'RC', connects to R5 and C3. The time constant of these two is 36us, so I suspect these are responsible for setting the time required to reset the sequence when clock is held low. My experimentation showed this time to be about 34us, which, allowing for the logic threshold being different to the voltage at the time constant, tallies very closely.

Here's the matrix diagram for WY-60 ASCII (840105-01 / 901867-01) and WY-60 ANSI (840338-09)...

44942-1

Note the IDs - these correspond to the scanner chip reading the last line of the matrix, driven by -LCLR.
41 hex matches the observation that bit positions 144 and 150 were always low for the WY-60 ASCII.

Examining my WY-PCE, I can see three diodes connected to -LCLR. These set an ID of 83 hex, resulting in bit positions 144, 145 and 151 always being read as low - which doesn't match kbdbabel's signaling diagram. That's probably a good thing as far as a converter is concerned, because it will be able to tell the PCE and WY-85 keyboards apart! However, my PCE has an ISO enter key, so perhaps the ID is different for the one with ANSI enter.

So, of the 20 byte (160 bit) sequence, there appears to be 18 bytes (144 bits) of key state (even though the scanner chip couldn't possibly scan that many), then a byte for the ID, then a byte for LEDs.

I was expecting to see a shift register driving the LEDs on the PCE. But no! They are driven by a counter (74HC163), which is driven from the -LCLR and -LCLK pins of the scanner chip. Since the scanner chip presumably has no clock of its own, I'm not at all sure how this arrangement works yet! edit: Aha! Now I understand what "Clock-Bits 152-158 used for LED-signaling, -4: Caps, -2: Num, -1: Scroll" means. After clock pulse 151, clock pulses are simply fed through to the counter, so stopping the sequence short sets the LEDs.
« Last Edit: Sat, 14 December 2013, 09:16:48 by Soarer »

Offline Soarer

  • * Moderator
  • Thread Starter
  • Posts: 1918
  • Location: UK
Re: Wyse keyboard protocol (4P4C)
« Reply #3 on: Wed, 20 November 2013, 08:12:11 »
The protocol is essentially worked out and I've edited the previous posts.

There's now a table at the end of the first post which shows what information is still needed. Please, if you have one of the keyboards which has any '?' then try to test it for us! Or just send it to me, even if it's just the PCB :D

Offline jdcarpe

  • Literally geekhack
  • * Curator
  • Posts: 8572
  • Location: Odessa, TX
  • God mode active
Re: Wyse keyboard protocol (4P4C)
« Reply #4 on: Wed, 20 November 2013, 09:44:45 »
I have a WYSE PCE (ANSI Enter) which I really want to get working, so I will try and test it with this when I have a chance.

Also, I believe I have a WY85 I can test, somewhere.

And a WY50, but that's another beast. :)

Offline metalliqaz

  • * Maker
  • Posts: 4698
  • Location: the Making Stuff subforum
  • Leopold fanboy
Re: Wyse keyboard protocol (4P4C)
« Reply #5 on: Wed, 20 November 2013, 10:38:15 »
Sweet I total want to get my PCE working
Notice: I am no longer active as of August 2016
buy/sell/trade thread | keyboard collection | Epsilon project | Qazpad project | CompactTKL project | Sigma project | Easy firmware/keymapper
SmallFry will be remembered

Offline Soarer

  • * Moderator
  • Thread Starter
  • Posts: 1918
  • Location: UK
Re: Wyse keyboard protocol (4P4C)
« Reply #6 on: Wed, 20 November 2013, 16:07:10 »
I have a WYSE PCE (ANSI Enter) which I really want to get working, so I will try and test it with this when I have a chance.

Also, I believe I have a WY85 I can test, somewhere.

And a WY50, but that's another beast. :)

That would be great! I'll do the PCE (ISO) matrix later, so you might not have much to do on that one... we'll see!

The WY-50 probably works with my controller with a config similar to the bigfoot - except without the muxstrobe_gate, and possibly with more sense_delay. Not sure about sense_polarity.


Offline Soarer

  • * Moderator
  • Thread Starter
  • Posts: 1918
  • Location: UK
Re: Wyse keyboard protocol (4P4C)
« Reply #7 on: Wed, 20 November 2013, 19:07:36 »
Wyse PCE Int'l (ISO) matrix...

45179-0

* Wyse-PCE_ISO.xls (18.5 kB - downloaded 86 times.)

Notes:
Position 05 is backslash on the ANSI layout. Corresponds to HID code 31 hex, BACKSLASH, is both cases.
Position 17 is omitted from the ANSI layout. Corresponds to HID code 64 hex, EUROPE_2.

Scancodes start at the bottom left and increase up the columns, like this...

45181-2
« Last Edit: Thu, 21 November 2013, 06:36:17 by Soarer »

Offline Soarer

  • * Moderator
  • Thread Starter
  • Posts: 1918
  • Location: UK
Re: Wyse keyboard protocol (4P4C)
« Reply #8 on: Wed, 20 November 2013, 20:33:49 »
Urbanus' clavis project contains matrix layouts for PCE (ANSI) and ANSI Gate Array, which confirms that the PCE (ANSI) has effectively the same layout as the PCE (ISO) minus the extra ISO key next to left shift.

ANSI Gate Array and WY-85 may well have the same matrix, since they share a layout - still not sure what difference there is between those two!

From Sandy's PCE (ANSI) page, I can see that it has 2 diodes connected to pin 21 (-LCLR), and that one of them is connected to C1 meaning that bit 1 of its ID (bit 145 of the full sequence) would be low. Can't tell where the other one connects to; it's out of frame...

« Last Edit: Thu, 21 November 2013, 06:27:00 by Soarer »

Offline dorkvader

  • Posts: 6260
  • Location: Boston area
  • all about the "hack" in "geekhack"
Re: Wyse keyboard protocol (4P4C)
« Reply #9 on: Tue, 26 November 2013, 20:50:46 »
This may be of some use here
http://geekhack.org/index.php?topic=38260
IBM nut and bolt mod guide Keyboard Photos "I sort of want to have 1/4 of a grand staircase connecting two halves of a split level"
I am happy to answer honest PMs from concerned parties

Offline Soarer

  • * Moderator
  • Thread Starter
  • Posts: 1918
  • Location: UK
Re: Wyse keyboard protocol (4P4C)
« Reply #10 on: Wed, 27 November 2013, 06:55:23 »
Hmm, interesting that the WYSEpc 286 could take either a 4P4C, or a regular AT keyboard using an adapter from DIN to 4P4C:

Quote from: Robert Casey;6282;3.57;$0201  06.09.89
copied from packet:
 Msg# TSP  Size #Rd Date/Time MsgID        From   To
  2364 BN$  1442   0 0902/1001 1817_W6PW    WD6EOS ALL@WA2RKN.NYNET
   Sb: USING STD PC KBD WITH WYSE
   
    I would be very surprised if there were not other people out there, like
    myself, who own a Wyse 2112 or similar PC-compatible computer, and would
    like to use a different keyboard than what Wyse supplies.

It can be done; any IBM PC, XT or AT style keyboard using a DIN
connector will work with these machines. However, since mine was
missing the 4-pin modular to 5-pin DIN adaptor, I had to build one. The
interconnections are as follows.

WYSE MODULAR CORD WIRE COLOR:    FUNCTION:    CONNECT TO DIN PIN:
RED                    +5V                  5
GREEN                  CLOCK                1
BLACK                  DATA                 2
YELLOW                GROUND                4


Any electronic supplier should have the 5-pin DIN female. I have used
this mod very successfully with my 2112 and a Mitsumi enhanced AT-style
keyboard.  In order to preserve the original functions of the Wyse
keyboard, I do recommend the use of a 101-key enhanced AT version.
Best of luck!
                                  73,
                                  DE
                                  WD6EOS,
                                  Bruce,
                                  in
                                  Berkeley,
                                  CA.

(The link to that was found in this post).

It wouldn't surprise me at all to find that the 286's 4P4C keyboard (model PCET?) used the same protocol as the rest of these. The mysterious U57 on the motherboard would then either translate it to AT protocol for the 8042, or pass through standard AT, depending on which keyboard type was attached.

« Last Edit: Wed, 27 November 2013, 07:05:46 by Soarer »

Offline Soarer

  • * Moderator
  • Thread Starter
  • Posts: 1918
  • Location: UK
Re: Wyse keyboard protocol (4P4C)
« Reply #11 on: Wed, 27 November 2013, 07:28:43 »
60 ANSI, 85, 285... Incoming!!

The 285 intrigues me - it has more than 13*8 keys - what will be inside?!

Offline Soarer

  • * Moderator
  • Thread Starter
  • Posts: 1918
  • Location: UK
Re: Wyse keyboard protocol (4P4C)
« Reply #12 on: Wed, 27 November 2013, 10:08:37 »
Interesting list of part numbers here for the Wyse PCs:

WYSEpc ModelDescriptionPart Number
WY-1100KEYBOARD840114-01
WY-1400KEYBOARD----
WY-2108ENHANCED KEYBOARD (102 KEY)840358-13
WY-2108KEYBOARD (84 KEYS)900130-01
WY-2112ENHANCED KEYBOARD (102 KEY)840358-13
WY-2112KEYBOARD (84 KEYS)900130-01
WY-2114ENHANCED KEYBOARD (102 KEY)840358-13
WY-2114KEYBOARD (84 KEYS)900130-01
WY-2200KEYBOARD, 286 STYLE840275-01
WY-2200KEYBOARD, ENHANCED840358-01/901865-01
WY-2200KEYBOARD, AT STYLE990088-01

840114-01, dunno. This thread is already coming up second in my google searches, after the link above. Could be a variant of the WY50/75 keyboard, see here and this picture from that thread:



840275-01 has the same base part number as the AT-Style keyboard (840275-04). The suffix can indicate anything from different keycaps to a completely different layout, as seen in the WY60 variants! (However, they are all compatible with the same terminal in that case).

840358-01 and 901865-01 are numbers already listed as terminal compatible.

840358-13 is the 'PCE US, White' keyboard.

900130-01 I suspect is a 'single unit' part number for the AT-Style keyboard. (See here where a similar pattern can be seen for other keyboards, i.e. 900xxx, 840xxx). A few resellers list it as a WY60/99GT keyboard.

990088-01 is the part number for the PCB of the AT-Style keyboard mentioned in the WY-60 service manual.
« Last Edit: Sun, 22 December 2013, 16:00:37 by Soarer »

Offline Soarer

  • * Moderator
  • Thread Starter
  • Posts: 1918
  • Location: UK
Re: Wyse keyboard protocol (4P4C)
« Reply #13 on: Thu, 28 November 2013, 22:48:26 »
60 ANSI, 85, 285... Incoming!!

What I got wasn't quite what I expected!

The 60 ANSI was actually the 3161 variant (840338-02) with over 104 keys - nothing like the 60 ANSI 840338-09. It borrows a couple of the positions used for ID for the excess keys.

Still not sure exactly what the '85' is. The PCB says 980062-01, which does appear to be a part number for a keyboard, but none of the resellers had any pictures. It has the same key layout as WY85, but it does not use the 28-pin scanner chip. Instead, it has the same 4 chip circuit that the WY-50 has, plus a 74LS393 counter to cycle through the strobes. The interface is just clock and data, and the output bit sequence is the same as the first 16 bytes of the scanner chip's sequence for the 840366-01 WY85 Gate Array. It was listed as "840105-01 85 Keyboard", which I'm assuming it is, pending confirmation (it's lost the label on the back).


The 285 intrigues me - it has more than 13*8 keys - what will be inside?!

An ingenious bit of circuitry to add an extra strobe after R12:

46195-0
« Last Edit: Sun, 22 December 2013, 07:58:59 by Soarer »

Offline Soarer

  • * Moderator
  • Thread Starter
  • Posts: 1918
  • Location: UK
Re: Wyse keyboard protocol (4P4C)
« Reply #14 on: Thu, 28 November 2013, 22:52:28 »
Couple more matrix layouts from the WY60 Maintenance Manual.

WY60 3161...

46197-0

WY60 AT-Style...

46199-1

Offline Ethaniel

  • Posts: 15
Re: Wyse keyboard protocol (4P4C)
« Reply #15 on: Sat, 07 December 2013, 11:29:13 »
I've bought four of these today. If you need me to butcher one or two for the cause, let me know. :)

Offline Soarer

  • * Moderator
  • Thread Starter
  • Posts: 1918
  • Location: UK
Re: Wyse keyboard protocol (4P4C)
« Reply #16 on: Sat, 07 December 2013, 12:08:35 »
None of this involves actual butchering! :D

Which ones did you get? There's still a few assumptions that need confirmation ('TBC' at the end of the first post). And other variants might yet appear!

Offline Ethaniel

  • Posts: 15
Re: Wyse keyboard protocol (4P4C)
« Reply #17 on: Sun, 08 December 2013, 14:27:44 »
The four of them are 840358-01, a common model it seems. I just grabbed them because of switches, and because they were absurdly cheap. 4 WYSEs and 7 model Ms for 2.50, I couldn't let that pass. Lost my arms and shoulders bringing them back, though.  ;D

Offline Soarer

  • * Moderator
  • Thread Starter
  • Posts: 1918
  • Location: UK
Re: Wyse keyboard protocol (4P4C)
« Reply #18 on: Sun, 08 December 2013, 19:00:41 »
How cheap?! Nice haul! :cool:

Might you convert one of the WYSE, or are they just switch and cap donors?

Offline Ethaniel

  • Posts: 15
Re: Wyse keyboard protocol (4P4C)
« Reply #19 on: Mon, 09 December 2013, 19:04:59 »
Let's see: 4 WYSE 840358-01 keyboards, four PS/2 Model M's (1391506, need a lot of work), and three "terminal" 1390250 Model M's, for a grand total of USD 2.50 (I think the recycling guy was happy to see them go). I want to convert at least one of the WYSE. My "clone-ATMega32U4-boards-that-are-not-Teensys" are trapped in the mail right now (on strike until Wednesday). I'd love to use real proper Teensys, but I can't bring them down here. Customs will intercept and the cost will jump 50 per cent. :(

Looks like you need something called "ID code for PCE US (ANSI Enter)", that would be one of my 840358-01, correct?

Offline Soarer

  • * Moderator
  • Thread Starter
  • Posts: 1918
  • Location: UK
Re: Wyse keyboard protocol (4P4C)
« Reply #20 on: Mon, 09 December 2013, 19:57:51 »
Oh cool, you could start deciding how you'll wire it up then :D

Since the PCEs have the cable on a socket, you could replace the cable (or butcher one you have already) and solder one end of it direct to the Teensy. Telephone handset cables have the right connector, and are cheap and easy enough to find.

Or, you could make it with a 4P4C socket, but they're generally designed to mount on a PCB, so you might need a breakout board for it as well.

The cable route is the easiest by far, but then it wouldn't work with any WYSE which have non-removable cables.

The ID code, yes, that would be one of yours... I'm 90% sure of it, but if you wire one up and run the test .hex firmware we can be 100% sure :D

Offline Soarer

  • * Moderator
  • Thread Starter
  • Posts: 1918
  • Location: UK
Re: Wyse keyboard protocol (4P4C)
« Reply #21 on: Sat, 14 December 2013, 09:12:19 »
The WY30 (840013-01) matrix took a while to get straight :(

From the maintenance manual:

47961-0

Completely illegible in places! :eek:

Found a post on kbdmania which had this table:



Unfortunately that seems to have some errors compared to the maintenance manual.

Moving a few blocks of cells down one cell makes it match:



(The rows and columns are still in a different order, and the labeling doesn't correspond to correct pin numbers).

Final clean version that matches the other matrix diagrams:

47967-3

WY30 PCB pics by dorkvader also used to verify the deductions above and spot test a few other positions :thumb:
« Last Edit: Fri, 10 January 2014, 14:51:06 by Soarer »

Offline Soarer

  • * Moderator
  • Thread Starter
  • Posts: 1918
  • Location: UK
Re: Wyse keyboard protocol (4P4C)
« Reply #22 on: Wed, 18 December 2013, 19:59:49 »
Currently trying to work out how to protect you from yourselves, when some of you inevitably wire the connector backwards :p

When wired in reverse, +5V goes onto the keyboard's clock line, GND onto Data, Clock onto 5V, and Data onto GND.

Since the Clock and Data lines have protection diodes on them, and the keyboard's Clock has 5V on it, the keyboard's 5V is also at 5V. Thus, driving Clock low to the keyboard's 5V would probably fry that pin of the Teensy, or worse!

Offline Soarer

  • * Moderator
  • Thread Starter
  • Posts: 1918
  • Location: UK
Re: Wyse keyboard protocol (4P4C)
« Reply #23 on: Sat, 21 December 2013, 15:49:31 »
First WYSEverter version is now available!