TMK IBM PC Keyboard Converter

[Rayndalf]

I don't have my disassembled one on hand, but comparing the rear labels it (from an old time stamp) to my complete one and they appear to be different revisions.
http://imgur.com/a/2aZwdgY

I have yet to receive one that still has a front sticker/badge, I assume Unitek worked as an OEM and various other companies could slap their own logo on top, but I'm going to try to get some metal badges made and need some inspiration.

Sorry for derailing this thread, I'm ordering some DIN sockets  and will get a convertor made whenever they can arrive.

[Maledicted]

I don't have my disassembled one on hand, but comparing the rear labels it (from an old time stamp) to my complete one and they appear to be different revisions.
http://imgur.com/a/2aZwdgY

I have yet to receive one that still has a front sticker/badge, I assume Unitek worked as an OEM and various other companies could slap their own logo on top, but I'm going to try to get some metal badges made and need some inspiration.

Sorry for derailing this thread, I'm ordering some DIN sockets  and will get a convertor made whenever they can arrive.

Cool, it is in really good shape too.

I have never seen one with a sticker or badge at all myself, lol. I wonder if they designed the case and already had the tooling and just went, "... eh, screw it." If you get any badges made, I would love to hear about it, and maybe get in on that order. You could try to recreate IBM's metal badges. It looks close enough to an F AT for it to fit in well enough.

[hasu]

Keymap Editor is available here now.

http://www.tmk-kbd.com/tmk_keyboard/editor/#ibmpc_usb

Report when you try this firmware with your keyboard.  Check first post.

[Maledicted]

Keymap Editor is available here now.

http://www.tmk-kbd.com/tmk_keyboard/editor/#ibmpc_usb

Report when you try this firmware with your keyboard.  Check first post.

Awesome. I'll take a look at it. I have used the F XT as much as I can stomach and can't say that any typing errors have been the result of your firmware. I thought so at first, but I think all of that has been me missing the oddly-placed and stepped modifiers, so far.

[PancakeMSTR]

Hasu

I just gave this a try with my Northgate Omnikey Ps2. It's almost working perfectly, but I've identified two strange items. Specifically, the 'F11' and 'RALT' keys don't seem to want to fire:



I'm not entirely sure why this would be. It could honestly be an issue with my keyboard. I'm getting RALT to fire intermittently by pressing down on it really hard.

Other than that, looks good!

[EspritBerlin]

Keymap Editor is available here now.

http://www.tmk-kbd.com/tmk_keyboard/editor/#ibmpc_usb

Report when you try this firmware with your keyboard.  Check first post.

Thanks!

A would like to ask you and the community advise how to re-map a 126 key G80-2551. Is has 4 additional keys around the nav cluster, so the 122 key layout would not work I assume. Referring to https://deskthority.net/wiki/Cherry_G80-2551 it is said that "[the] 4 extra keys which send a NULL before the key codes of 1 2 3 and 4 which means that these keys are useless on computers using the PS/2 protocol."

Any help and hints are very, very welcome before I start working on that board.

Thank you all in advance! 

[hasu]

Thanks!

A would like to ask you and the community advise how to re-map a 126 key G80-2551. Is has 4 additional keys around the nav cluster, so the 122 key layout would not work I assume. Referring to https://deskthority.net/wiki/Cherry_G80-2551 it is said that "[the] 4 extra keys which send a NULL before the key codes of 1 2 3 and 4 which means that these keys are useless on computers using the PS/2 protocol."

Any help and hints are very, very welcome before I start working on that board.

Thank you all in advance! 

Use Full-key layout instead of 122-key Keymap Editor can support up to 128 keys at same time. It is different from the keyboard layout and not intuitive for the keyboard, though, you should still remmap all keys with Full-key layout.

Download firmware without editting keys when you test first, though.

I can't understand exactly what the wiki says, especially "before the keycodes of 1 2 3 and 4" part. What does the "1 2 3 and 4" mean?  I may be missing something, it would be helpful if you or someone can rephrase.

Try the firmware with default layout and see if the four keys resister something on computer. If not, use 'hid_listen' tool to see raw scan codes the keyboard spits out, so that I can update firmware for the keys if needed.

Let us know your result.

[EspritBerlin]

Dear All

First: a very big thanks to hasu for pointing me to 'ibmpc_usb_at90usb1286.hex' (https://raw.githubusercontent.com/tmk/tmk_keyboard/master/converter/ibmpc_usb/binary/ibmpc_usb_at90usb1286.hex) to be able to use the G80-2551 with a Teensy 2++.

The "regular" 122 keys work perfectly. (I would like to arrange some keys - later).

I used hid_listen to check the scan codes for the 4 extra keys around the nav cluster (see attached pic);
for the first one "TD=" (10 o'clock in the pic) I get:

r80 !CS2_INIT!
I45570 X46071 r26 W46072 r80 W46074 rF0 W46076 R46076 wF2 rFA rAB r86 ID:AB86(2)
S46083 wED rFA w02 rFA L46090 3r26 333333333333333333333333333333333333333333

for the second "page with edge":

r80 !CS2_INIT!
I61468 X61969 r25 W61970 r80 W61972 rF0 W61974 R61975 wF2 rFA rAB r86 ID:AB86(2)
S61981 wED rFA w02 rFA L61988 4r25 44444444444444444444444444444

for the third "two pages (1/2)":

r80 !CS2_INIT!
I10585 X11087 r16 W11088 r80 W11090 rF0 W11092 R11092 wF2 rFA rAB r86 ID:AB86(2)
S11099 wED rFA w02 rFA L11106 r16 11111111111

for the forth "calc":

r80 !CS2_INIT!
I17872 X18374 r1E W18375 r80 W18377 rF0 W18379 R18380 wF2 rFA rAB r86 ID:AB86(2)
S18386 wED rFA w02 rFA L18394 r1E 22222222222222222222222222222222222

After every key press the 1s, 2s, 3s, and 4s repeat endless until I press another key.

(Just in case: here the output after I pressed qwertzuiop which behaves normal on the keyboard:
r15 qrF0 r15 r1D wrF0 r1D re24 rF0 r24 rr2D rF0 r2D r2Ct rF0 r2C r35 zrF0 r35 r3C urF0 r3C r43 irF0 r43 ro44 rF0 r44 r4D prF0 r4D)

It would be great if the 4 extra keys could be used and mapped.

Best, EspritBerlin

[hasu]

EspritBerlin,

That hid_listen log is very helpful to add support of the extra keys.
The keyboard seem to send codes in Scan Code Set 2 like below for the keys.

TD=: 80, 26 / 80, F0, 26
page with edge: 80, 25 / 80, F0, 25
two pages: 80, 16 / 80, F0, 16
calc: 80, 1E / 80, F0, 1E

I added some codes to support the four extra keys. Try attached hex file and can you post hid_listen output again? They should not repeat key any more.

The keys are assigned to Japanese specific keys and Mute by default. This is because of compatibitly for other terminal keyboards. (You can remap the keys with editor later after I add support.)

Also check the keys with xev assuming you are on Linux. Your OS may not recognize the Japanese keys, perhaps.

 ibmpc_usb_at90usb1286.hex (54.2 kB - downloaded 303 times.)

[EspritBerlin]

Dear Hasu

Thanks for the new hex file! It works:

hid_listen reports:

TD=: r80 r26 r80 rF0 r26
page with edge: r80 r25 r80 rF0 r25
two pages: r80 r16 r80 rF0 r16
calc: r80 r1E r80 rF0 r1E

There is no repetition anymore; as you assumed, 'page with edge' does mute/unmute, the other three keys nothing so far, given that I use German keyboard layout.

xev reports:

TD=:

KeyPress event, serial 38, synthetic NO, window 0x5800001,
    root 0x1da, subw 0x0, time 792293, (-469,195), root:(392,650),
    state 0x10, keycode 132 (keysym 0x0, NoSymbol), same_screen YES,
    XLookupString gives 0 bytes:
    XmbLookupString gives 0 bytes:
    XFilterEvent returns: False

Page with edge:
FocusOut event, serial 39, synthetic NO, window 0x5800001,
    mode NotifyGrab, detail NotifyAncestor

FocusIn event, serial 39, synthetic NO, window 0x5800001,
    mode NotifyUngrab, detail NotifyAncestor

KeymapNotify event, serial 39, synthetic NO, window 0x0,
    keys:  2   0   0   0   0   0   0   0   0   0   0   0   0   0   0   0   
           0   0   0   0   0   0   0   0   0   0   0   0   0   0   0   0 

Note: in the sequence above, "keys: 2  0  0.." I got sometimes on the first keypress also 75, 419, after that always 2.

Two pages:
KeyPress event, serial 38, synthetic NO, window 0x5800001,
    root 0x1da, subw 0x0, time 1032317, (-297,-49), root:(564,406),
    state 0x10, keycode 97 (keysym 0x0, NoSymbol), same_screen YES,
    XLookupString gives 0 bytes:
    XmbLookupString gives 0 bytes:
    XFilterEvent returns: False

Calc:
KeyPress event, serial 38, synthetic NO, window 0x5800001,
    root 0x1da, subw 0x0, time 1121468, (-161,127), root:(700,582),
    state 0x10, keycode 101 (keysym 0xff27, Hiragana_Katakana), same_screen YES,
    XLookupString gives 0 bytes:
    XmbLookupString gives 0 bytes:
    XFilterEvent returns: False

Best, EspritBerlin

[hasu]

Great. It seems to work as I expected.
And xev recognizes the four keys somehow but Xorg doesn't seems to have useful keymapping for them by default. You will have to remap the keys on Keymap Editor.

OK. I'll work on Keymap Editor for the extra keys and some other keys specific to the keyboard.

Could you look into scan codes for 10 function keys on left cluster,  6 keys above cusors key and 'Linie' keys?
You can use hid_listen again as you did in previous post. This would be helpful to make reasonable layout of Keymap Editor.

Thanks,

[EspritBerlin]

Great. It seems to work as I expected.
And xev recognizes the four keys somehow but Xorg doesn't seems to have useful keymapping for them by default. You will have to remap the keys on Keymap Editor.

OK. I'll work on Keymap Editor for the extra keys and some other keys specific to the keyboard.

Could you look into scan codes for 10 function keys on left cluster,  6 keys above cusors key and 'Linie' keys?
You can use hid_listen again as you did in previous post. This would be helpful to make reasonable layout of Keymap Editor.

Thanks,

Please find the scan codes attached 

[eruizc]

Hello everyone! I'm new to this, I'm amazed by your work but may need some insights on what to do now...
I've got a model M122 1390250 that is currently not working, this is what the hid_listen says:

Code: [Select]

Waiting for device:...
Listening:

TMK:fefe1028/LUFA

USB configured.

Keyboard start.
I458 X960
Device disconnected.
Waiting for new device:..........
Listening:

TMK:fefe1028/LUFA

USB configured

Keyboard start
I488 X989
Device disconnected.

I've noticed the keyboard has 8 switches at the back, done some research but I'm completely lost on them. The first try they were 11111110 and the second 00000001, because I read somewhere to try them like that.

What could I be missing here?

[hasu]

eruizc,
Can you post pic of the switches for the future reference? I couldn't find it with my quick internet search.

I guess the switches cofigure part of keyboard ID and should be '0xBF'. Try below.
10111111
0100000
11111101
0000010

See Identity explanation on this page.
https://www.seasip.info/VintagePC/ibm_1390876.html

Also you better check if your converter work with other keyboard first.


EspritBerlin,
Thanks for the scan codes! I'm working on Keymap Editor and firmware for G80-2551 and will post when it is done.

[hasu]

Can you try attached firmware?
Especially for 10 function keys on left and F13-F24 portion, check if they register anything in xev. It doesn't matter what those keys registers at this point.

Other keys are expected to register as below.

 * G80-2551(Code Set 2) Unimap default mapping                                                                     
 *               ,-----------------------------------------------.
 *               |Gui|Mhk|F23|Hnk|F13|F14|F15|F16|F17|F18|F19|F20|                                                 
 *               |-----------------------------------------------|
 *               | F1| F2| F3| F4| F5| F6| F7| F8| F9|F10|F11|F12|                                                 
 *               `-----------------------------------------------'
 * ,-------. ,-----------------------------------------------------------. ,-----------. ,---------------.         
 * |Mut|F24| |   |   |   |   |   |   |   |   |   |   |   |   |   |       | |  /|PgU|PgD| |Esc|NmL|ScL|  *|
 * |-------| |-----------------------------------------------------------| |-----------| |---------------|         
 * |VoU|Pau| |     |   |   |   |   |   |   |   |   |   |   |   |   |     | |End|Ins|Del| |   |   |   |  +|
 * |-------| |------------------------------------------------------`    | |-----------| |---------------|         
 * |VoD|F22| |      |   |   |   |   |   |   |   |   |   |   |   |   |    | |Yen|   |  #| |   |   |   |  -|
 * |-------| |-----------------------------------------------------------| |-----------| |---------------|         
 * |Psc|App| |    |   |   |   |   |   |   |   |   |   |   |   |          | |   |Hom|   | |   |   |   |   |
 * |-------| |-----------------------------------------------------------| |-----------| |-----------|Ent|         
 * |Gui|F21| |     |   |                                  |    |   |       | RO|   |KAN| |       |  .|   |
 * `-------' `-----'   `---------------------------------------'   `-----' `-----------' `---------------'         
 * 

 ibmpc_usb_at90usb1286.hex (54.78 kB - downloaded 285 times.)

Great. It seems to work as I expected.
And xev recognizes the four keys somehow but Xorg doesn't seems to have useful keymapping for them by default. You will have to remap the keys on Keymap Editor.

OK. I'll work on Keymap Editor for the extra keys and some other keys specific to the keyboard.

Could you look into scan codes for 10 function keys on left cluster,  6 keys above cusors key and 'Linie' keys?
You can use hid_listen again as you did in previous post. This would be helpful to make reasonable layout of Keymap Editor.

Thanks,

Please find the scan codes attached 

[EspritBerlin]

I checked new firmware. I have serious trouble with repeating keys for the 10 function keys on the left.
Please see details attached.

I have not checked F1-F24 etc so far.

Can you try attached firmware?
Especially for 10 function keys on left and F13-F24 portion, check if they register anything in xev. It doesn't matter what those keys registers at this point.

Other keys are expected to register as below.

[hasu]

Thanks for testing.

I think this fixes the problem.

 ibmpc_usb_at90usb1286.hex (54.83 kB - downloaded 449 times.)

[EspritBerlin]

Thank you!

I tested the new firmware esp. regarding the repeating for the left 10 function keys.
It got better, but it seems that they only register about every second key press and get rarely into the repeating mode again.
Very left column of keys (labeled: Abruf=row1, PsAus=row2, ErAus=row3, Druck=row4, blank=row5) behave as follows:

Abruf (row1):sometimes toggles between mute and volume, fast repeating key presses lead to loop/repeating on/off vol

PsAus (row2): every 2nd to 3rd key press increases vol

ErAus (row 3): singe key press has no action, 3 fast key presses = vol - by auto repeat

Druck (row 4): print screen works without problems

blank (row 5): menu key loops/repeats (same as in my post before: pressing d leads to desktop); I can leave this loop only by unplugging Teensy from USB.

I added some information in the attached docs for the 5 keys.
Please let me know how I can help to investigate further.

Thanks for testing.

I think this fixes the problem.

(Attachment Link)

[hasu]

I seems to have missed something to be cared for.
hope new firmware fixes the problems this time.

 ibmpc_usb_at90usb1286.hex (54.83 kB - downloaded 302 times.)

[EspritBerlin]

The loops are gone! The firmware did not hung or crash once, no unexpected behavior!

I have tested the firmware and attached hid_listen output for
- 10 function keys (left)
- 24 F keys (top)
- nav cluster
- num pad (right)

For the 10 function keys I included xev, for the remaining keys I tested if xev output was given.

Since this layout opens the opportunity to move arrow keys to inverted T this would be a good option I think.
Please let me know if further testing/reporting is needed to support you.

Thanks, best, EspritBerlin

I seems to have missed something to be cared for.
hope new firmware fixes the problems this time.

(Attachment Link)

[hasu]

EspritBerlin,

I updated Keymap Editor for G80-2551 support. You can acess it here.
http://www.tmk-kbd.com/tmk_keyboard/editor/unimap/?ibmpc_usb_1286_terminal

Keys are expected to register as below with default mapping. Function keys on left and top row are not straight
due to difference from IBM 122-key. Those keys can be remappable somehow, Editor is not intuitive though.

Try it and report if you find something weird.

 * G80-2551(Code Set 2) Unimap default mapping
 *               ,-----------------------------------------------.
 *               |Gui|Mhk|F23|Hnk|F13|F14|F15|F16|F17|F18|F19|F20|
 *               |-----------------------------------------------|
 *               | F1| F2| F3| F4| F5| F6| F7| F8| F9|F10|F11|F12|
 *               `-----------------------------------------------'
 * ,-------. ,-----------------------------------------------------------. ,-----------. ,---------------.
 * |Mut|F24| |   |   |   |   |   |   |   |   |   |   |   |   |   |       | |NmL|PgU|PgD| |Esc|ScL|  *|  /|
 * |-------| |-----------------------------------------------------------| |-----------| |---------------|
 * |VoU|Pau| |     |   |   |   |   |   |   |   |   |   |   |   |   |     | |End|Ins|Del| |   |   |   |  +|
 * |-------| |------------------------------------------------------`    | |-----------| |---------------|
 * |VoD|F22| |      |   |   |   |   |   |   |   |   |   |   |   |   |    | |Yen|   |  #| |   |   |   |  -|
 * |-------| |-----------------------------------------------------------| |-----------| |---------------|
 * |Psc|App| |    |   |   |   |   |   |   |   |   |   |   |   |          | |   |Hom|   | |   |   |   |   |
 * |-------| |-----------------------------------------------------------| |-----------| |-----------|Ent|
 * |Gui|F21| |     |   |                                  |    |   |       | RO|   |KAN| |       |  .|   |
 * `-------' `-----'   `---------------------------------------'   `-----' `-----------' `---------------'
 * JPY: JIS ￥
 * KAN: ひらがな
 * RO: JIS ろ(Japanese RO)
 * Hnk: 変換(Japanese Henkan)
 * Mhk: 無変換(Japanese MuHenkan)

[EspritBerlin]

Hi Hasu

Thank you for all the work! I did work on the nav cluster, worked fine!
As you expected I am did not manage to do some changes to some keys. I attached an example (see hex and pic).
Num, nav, home, end, page up/down worked fine; new Esc (=1), insert (=2), three most right keys of num pad (=3) are resistant [all are mapped in the editor].
Furthermore gui assigment (last row, 10er fun keys) are resistant to change.

This is what his_listen reports for the 'new' keys
- ESC: r5F rF0 r5F
- insert: r1/7 rF0 r17
- old scl/new P/: r7E *rF0 r7E
- old */new Px: r7C -rF0 r7C
- old P() new P/: rE0 r4A ^[[2~rE0 rF0 r4A

Could you give me a hint how to change the behaviour of those 'strage' keys?

Thank you very much, best, EspritBerlin

EspritBerlin,

I updated Keymap Editor for G80-2551 support. You can acess it here.
http://www.tmk-kbd.com/tmk_keyboard/editor/unimap/?ibmpc_usb_1286_terminal

Keys are expected to register as below with default mapping. Function keys on left and top row are not straight
due to difference from IBM 122-key. Those keys can be remappable somehow, Editor is not intuitive though.

Try it and report if you find something weird.

[hasu]

Remmaping the keys is tricky.

For example,
When you want to Key1 in your pic:
The key of G80-2551 is refered as 'F24', see Fig.1 below.
You can remap the Key1 with changing key indicated as 'F24' in Fig.2 on Keymap Editor. i.e. key on left top.

When you want to Key2 in your pic:
The key of G80-2551 is refered as 'NmL'(NumLock), see Fig.1 below.
You can remap the Key2 with changing key indicated as 'NmL' in Fig.2 on Keymap Editor. You can find 'NmL' on keypad portoin, next to 'ESC' in Keymap Editor.

Hope this helps.

Fig.1 Default keymap of G80-2551

 *               ,-----------------------------------------------.
 *               |Gui|Mhk|F23|Hnk|F13|F14|F15|F16|F17|F18|F19|F20|
 *               |-----------------------------------------------|
 *               | F1| F2| F3| F4| F5| F6| F7| F8| F9|F10|F11|F12|
 *               `-----------------------------------------------'
 * ,-------. ,-----------------------------------------------------------. ,-----------. ,---------------.
 * |Mut|F24| |   |   |   |   |   |   |   |   |   |   |   |   |   |       | |NmL|PgU|PgD| |Esc|ScL|  *|  /|
 * |-------| |-----------------------------------------------------------| |-----------| |---------------|
 * |VoU|Pau| |     |   |   |   |   |   |   |   |   |   |   |   |   |     | |End|Ins|Del| |   |   |   |  +|
 * |-------| |------------------------------------------------------`    | |-----------| |---------------|
 * |VoD|F22| |      |   |   |   |   |   |   |   |   |   |   |   |   |    | |JPY|   |  #| |   |   |   |  -|
 * |-------| |-----------------------------------------------------------| |-----------| |---------------|
 * |Psc|App| |    |   |   |   |   |   |   |   |   |   |   |   |          | |   |Hom|   | |   |   |   |   |
 * |-------| |-----------------------------------------------------------| |-----------| |-----------|Ent|
 * |Gui|F21| |     |   |                                  |    |   |       | RO|   |KAN| |       |  .|   |
 * `-------' `-----'   `---------------------------------------'   `-----' `-----------' `---------------'

Fig.2 Default keymap of Keymap Editor

 *               ,-----------------------------------------------.
 *               |F13|F14|F15|F16|F17|F18|F19|F20|F21|F22|F23|F24|
 *               |-----------------------------------------------|
 *               |F1 |F2 |F3 |F4 |F5 |F6 |F7 |F8 |F9 |F10|F11|F12|
 *               `-----------------------------------------------'
 * ,-------. ,-----------------------------------------------------------. ,-----------. ,---------------.
 * |Mut|Hkn| |  `|  1|  2|  3|  4|  5|  6|  7|  8|  9|  0|  -|  =|Yen| BS| |  /|PgU|PgD| |Esc|NmL|ScL|  *|
 * |-------| |-----------------------------------------------------------| |-----------| |---------------|
 * |VoU|Pau| |Tab  |  Q|  W|  E|  R|  T|  Y|  U|  I|  O|  P|  [|  ]|    \| |End|Ins|Del| |  7|  8|  9|  +|
 * |-------| |-----------------------------------------------------------| |-----------| |-----------|---|
 * |VoD|Mhk| |CapsLo|  A|  S|  D|  F|  G|  H|  J|  K|  L|  ;|  '|  #| Ret| |JPY|Up |  #| |  4|  5|  6|  -|
 * |-------| |-----------------------------------------------------------| ,-----------| |---------------|
 * |Psc|App| |Shif|  \|  Z|  X|  C|  V|  B|  N|  M|  ,|  ,|  /| RO| Shift| |Lef|Hom|Rig| |  1|  2|  3|Ent|
 * |-------| |-----------------------------------------------------------| |-----------| |-----------|---|
 * |Gui|Gui| |Ctrl|    |Alt |          Space              |Alt |    |Ctrl| | RO|Dow|KAN| |  ,|  0|  .|  =|
 * `-------' `----'    `---------------------------------------'    `----' `-----------' `---------------'
 * JPY: JIS ￥
 * KAN: ひらがな
 * RO: JIS ろ(Japanese RO)
 * Hnk: 変換(Japanese Henkan)
 * Mhk: 無変換(Japanese MuHenkan)

[hasu]

EspritBerlin,
Can you try this temporary firmware for test if you have time?

 ibmpc_usb_at90usb1286.hex (56.78 kB - downloaded 312 times.)

I got some idea of handling G80-2551 as a Terminal keyboard(Code Set 3). Current firmware sees it as weird PC keyboards with 126 keys.

I like to know whether G80-2551 supports decently 'Code Set 3' and its scan codes are correspondent with genuine IBM ones.
We can have better/straight mapping support in Keymap Editor perhaps if the keyboard does.

It would be helpful if you can share hid_listen output of firmware startup.

Genuine terminal keyboard should send scan codes like below. Can you check left 10-key function portion and  F13-F24 especially? If they look all good look into four extra keys around cursors also.

 *               ,-----------------------------------------------.
 *               | 08| 10| 18| 20| 28| 30| 38| 40| 48| 50| 57| 5F|
 *               |-----------------------------------------------|
 *               | 07| 0F| 17| 1F| 27| 2F| 37| 3F| 47| 4F| 56| 5E|
 *               `-----------------------------------------------'
 * ,-------. ,-----------------------------------------------------------. ,-----------. ,---------------.
 * | 05| 06| | 0E| 16| 1E| 26| 25| 2E| 36| 3D| 3E| 46| 45| 4E| 55| 5D| 66| | 67| 6E| 6F| | 76| 77| 7E|*84|
 * |-------| |-----------------------------------------------------------| |-----------| |---------------|
 * | 04| 0C| | 0D  | 15| 1D| 24| 2D| 2C| 35| 3C| 43| 44| 4D| 54| 5B|  5C | | 64| 65| 6D| | 6C| 75| 7D| 7C|
 * |-------| |-----------------------------------------------------------| `-----------' |---------------|
 * | 03| 0B| | 14   | 1C| 1B| 23| 2B| 34| 33| 3B| 42| 4B| 4C| 52| 53| 5A |     | 63|     | 6B| 73| 74| 7B|
 * |-------| |-----------------------------------------------------------| ,-----------. |---------------|
 * |*83| 0A| | 12 | 13| 1A| 22| 21| 2A| 32| 31| 3A| 41| 49| 4A| 51|  59  | | 61| 62| 6A| | 69| 72| 7A| 79|
 * |-------| |-----------------------------------------------------------| `-----------' |---------------|
 * | 01| 09| | 11  |   |19  |        29                   |39  |   | 58  |     | 60|     | 68| 70| 71| 78|
 * `-------' `-----'   `---------------------------------------'   `-----'     `---'     `---------------'

[EspritBerlin]

Dear Hasu

Please find attached the hid_listen codes for 10 F keys on the left, nav cluster, F1-F24, and num pad.

Please let me know if I can help further

EspritBerlin,
Can you try this temporary firmware for test if you have time?

[hasu]

EspritBerlin,
Great. The keyboard seems to be very straight terminal keyboard except for the extra keys.

I just updated firmware in Keymap Editor to support G80-2551 in better way. This update breaks your current keymap probably and you will have to start mapping from scratch.

To get new version Editor you may need to reload Keymap Editor page completely in web browser with Shift+Control+R or summat.

Try mapping especially on the extra keys in Keymap Editor. Mapping keys should be very straight now.

Thanks,

[EspritBerlin]

Dear Hasu

I can confirm that all keys can be mapped. I am running the prior proposed mapping

- nav as inverted T
- num pad 'modern style'
- ESC on left 10 F keys
- changed position of menu key

Thank you very, very much for your help and patience!

Kind regards, EspritBerlin

EspritBerlin,
Great. The keyboard seems to be very straight terminal keyboard except for the extra keys.

I just updated firmware in Keymap Editor to support G80-2551 in better way. This update breaks your current keymap probably and you will have to start mapping from scratch.

To get new version Editor you may need to reload Keymap Editor page completely in web browser with Shift+Control+R or summat.

Try mapping especially on the extra keys in Keymap Editor. Mapping keys should be very straight now.

Thanks,

[Maledicted]

At long last, I stopped being lazy and installed wine on a Linux install I have lying around for that beige label Zenith Z-150. scla works on there, at least. I'll attach the full output from sctrace, and a zip file of every individual screenshot of a zoomed-in visualization of that output from scla. Here's a selection from the output, in order:











 Zenith Z-150 Beige Label scla output.zip (238.13 kB - downloaded 302 times.)

I hope that's everything you need to crack the case. Otherwise, let me know.

I think I mentioned it last time, but the portB version of sctrace basically did nothing, by the way. This was with the portD version of sctrace.

[hasu]

Maledicted,
Great contribution! Thank you for your time.

Zenith Z-150 Beige asserts stop bit as low level(0) for some reason, this is obvious violation of IBM PC AT protocol.




That is why many PS/2 converters and PS/2 port on PC fail to talk with the keyboard. Converter and PC with decent implementation of AT keyboard protocol won't accept the keyboard.

https://github.com/tmk/tmk_keyboard/wiki/IBM-PC-AT-Keyboard-Protocol#whitebeige-label-zenith-z-150

TMK IBM PC Keyboard Conveter doesn't check validity of stop bit, I don't think the stop bit prevents the converter from working. I may miss something else.

What ouputs do you get on hid_listen with the latest firmware?

[Maledicted]

I wish I could help more than what little I may have done. With the latest version of TMK it acts much the same:



That's all I can get because none of the keys register, the beeper doesn't even work. I tested with an old Fujitsu board, which is most likely AT, as a control and that board worked as it should. Even on the HP 8200 system that the board does work with (beeper and all keys), it still doesn't seem to always initiate caps lock, num lock, etc as you would normally expect from a keyboard, and on some other computers mashing caps lock over and over again would sometimes get the rest of the keys to register.

[hasu]

hid_listen log is almos always really helpful for me to debug.
'Copy and paste' log text is good enough unless taking screenshot is easier for you.

The converter can read powerup test result code(rAA) from the keyboard at startup as I expected.
But the keyboard doesn't response with 'read ID' command(wF2) at all and seems to stop working since then.

I just changed firmware temporarily to skip the 'read ID' and recognize it as AT(Code Set 2) automatically. You can type keys with firmware perhaps. And I guess it stops after pressing Caps Lock or Num Lock.

Can you try pressing Q, W, Caps Lock, A and S in this order? hid_listen log would be helpul again.

 ibmpc_usb_atmega32u4.hex (54.39 kB - downloaded 303 times.)

[Maledicted]

hid_listen log is almos always really helpful for me to debug.
'Copy and paste' log text is good enough unless taking screenshot is easier for you.

The converter can read powerup test result code(rAA) from the keyboard at startup as I expected.
But the keyboard doesn't response with 'read ID' command(wF2) at all and seems to stop working since then.

I just changed firmware temporarily to skip the 'read ID' and recognize it as AT(Code Set 2) automatically. You can type keys with firmware perhaps. And I guess it stops after pressing Caps Lock or Num Lock.

Can you try pressing Q, W, Caps Lock, A and S in this order? hid_listen log would be helpul again.

(Attachment Link)

Here you go:

Keyboard start.
I267 X769 rAA W1213 R1714 ID:0000(2)
S1714 wED w00 L1750
ERR:17
r15 rF0 r15 r1D rF0 r1D r58 wED w04
ERR:17
rF0 r58 r1C rF0 r1C r1B rF0 r1B

It looks like it actually has full function with that modified firmware, it even seems to toggle caps lock num lock and scroll lock normally, and continues to type. Interestingly, the LED status lights for the lock keys do not work though.

[hasu]

Can you try this and check if the LED indicators works?

 ibmpc_usb_atmega32u4.hex (54.26 kB - downloaded 318 times.)

According to this post in reddit,
it seems that the indicators are controlled internally by the keyboard, instead of the converter.

https://www.reddit.com/r/MechanicalKeyboards/comments/16woai/having_troubles_with_my_newly_acquired_zenith/

I recently picked up a Zenith data systems keyboard from 1984. Its in great condition. It has an AT plug so i got an AT(dim 5) to PS/2 adapter. I plug it in and it not registering with windows. It does however light up the num lck and caps lock when pressed and the keys make faint beeps when pressed.

[Maledicted]

That seems to have brought no functional change. I'm using the board right now to type this. These green Alps are wonderful. I noticed that Sys Req muted my computer's audio, lol. Is that a normal feature of TMK? The weird thing is that the lock lights do work on every direct ps2 connection computer that I could get the board to somewhat work on, even the ones where I had to mash the caps lock key many times on to get the rest of the keys to register.

With the HP 8200, literally everything works normally, though the lock keys don't always necessarily engage their corresponding lock mode on every press.

[hasu]

It depends on what scan code Sys Req key sends. What does log output show when pressing the key?
EDIT: And what does the key register on HP 8200?


WIth internet search I found people said many computers and PS/2 converters couldn't handle with the keyboard.
PS/2 port on HP 8200 can handle with beige Z-150. Interesting. Do you have other computer that can do with the keyboard?

When plugin USB keyboard with HP 8200 as well and press lock key on the USB keyboard, does indicator of Z-150 on PS/2 port change?
If yes the it means keyboard can receive commands from computer.

That seems to have brought no functional change. I'm using the board right now to type this. These green Alps are wonderful. I noticed that Sys Req muted my computer's audio, lol. Is that a normal feature of TMK? The weird thing is that the lock lights do work on every direct ps2 connection computer that I could get the board to somewhat work on, even the ones where I had to mash the caps lock key many times on to get the rest of the keys to register.

With the HP 8200, literally everything works normally, though the lock keys don't always necessarily engage their corresponding lock mode on every press.

[Maledicted]

It took me a while to even find a tool that would work with the Sys Req key. Nirsoft always seems to have an awesome utility for everything. Sys Req is functioning as print screen by default. Relatedly, the actual print screen key functions only as *, regardless of number lock status, etc.

I looked back through a thread I started in February when I first tested all of this. My HP may actually be an 8300. I think they're basically the same, HP just added USB 3.0 ports to the 8300 vs 8200. I tested on quite a few computers, though I think I may have more I could try now. A lot of the others I didn't at the time because they don't even have operating systems on them.

It had perfect function on my:

HP 8200/8300
Compaq SR1920nx (yeah, that's an oldie, my first desktop computer)

It had partial function by mashing the caps lock key many times on an old Dell XPS 630, in Linux Mint ... maybe 17 or so. The same trick didn't work in Windows 10.

It didn't work at all on any other computer I tried.

The lock keys of a separate keyboard do indeed turn on the corresponding indicator LED of the Zenith when connected to the HP.

[hasu]

Then, the keyboard supports commands somewhat.

I tweaked timing of command signal a bit.
Can you try attached firmware and post 'hid_listen' log with pressing the 'Sys Req' key, Caps Lock and Num Lock?

 ibmpc_usb_atmega32u4.hex (54.65 kB - downloaded 299 times.)

The lock keys of a separate keyboard do indeed turn on the corresponding indicator LED of the Zenith when connected to the HP.

As for the '*/Print screen' key, I think it worked as Print screen on original XT AT computer when pressed with Shift key. But mordern  OS and USB keyboards don't have the feature. The key just registers 'numpad *' with the converter and modern PCs.
And I think the 'Sys Req' key on IBM AT 84-key keyboard should be registered as Print screen. Its scan code is 0x84 but Z-150 seems to use diffrent code perhaps.

[Maledicted]

That new firmware does not work with the keyboard at all. On my first reboot after flashing it, the keys would still beep, but it wasn't sending any codes to the computer. Every other try after the keyboard would do nothing, so I flashed it back to the previous firmware and everything but the lock lights is working again. The caps lock and num lock keys, although they function normally, do not send any codes that hid_listen detects. I think I could get the specific codes with that Nirsoft utility I found though.

Here it is with sys req:

TMK:2e5e6a69+/LUFA


USB configured.

Keyboard start.
I279 X781 rAA W1224 R1725 ID:0000(2)
S1725 L1725 r7F rF0 r7F r58 rF0 r58 r77 rF0 r77

The utility is KeyboardStateView

Key Name   Description   Key Code   Hex Key Code   Key Pressed   Key Toggled
VK_CAPITAL   CAPS LOCK key   20   0x14   *   *   5/2/2020 12:03:25 AM

Key Name   Description   Key Code   Hex Key Code   Key Pressed   Key Toggled   Last Key Press Time   
VK_NUMLOCK   NUM LOCK key   144   0x90   *   *   5/2/2020 12:04:03 AM   

Those are, all standard, I imagine.

Here's Sys Req, with the converter:

Key Name   Description   Key Code   Hex Key Code   Key Pressed   Key Toggled   Last Key Press Time   
VK_VOLUME_MUTE   Volume Mute key   173   0xad   *   *   5/2/2020 12:08:34 AM   

[hasu]

Sorry. It seems I attached wrong or corrupt firmware for no reason.

Can you try this instead and check if the indicators work with Caps Lock and Num Lock?

 ibmpc_usb_atmega32u4.hex (54.56 kB - downloaded 296 times.)

[Maledicted]

That firmware is doing the same thing, with the Fujitsu control board from before as well.

[hasu]

That firmware is doing the same thing, with the Fujitsu control board from before as well.

I believe the firmware should display hid_listen log at least and wonder old firmware is sitll in the converter.
Could you check the file and flash it again?

[Maledicted]

That firmware is doing the same thing, with the Fujitsu control board from before as well.

I believe the firmware should display hid_listen log at least and wonder old firmware is sitll in the converter.
Could you check the file and flash it again?

I deleted the old firmware file from my computer, copied the latest to its place, flashed it, and tested. Then I flashed to the firmware prior to it, tested, flashed back to the latest and tested. That's what I did before posting the previous message. Now I deleted it once more, downloaded it again, renamed it just in case the Teensy flashing utility had cached it somewhere or something, and flashed it once more.

The results are all the same.

Here's what hid_listen says with this firmware:

Keyboard start.
I232 wED
ERR:11
X733 rAA W1178 R1679 ID:0000(2)
S1679 wED
ERR:11
L1750

[hasu]

The keyboard still doesn't like how the converter sends command and the request ends in timeout.

Attached firmware revised for signal timing again. Also fixed for 'SysRq' key, it should register Print Screen now.

 ibmpc_usb_atmega32u4.hex (54.72 kB - downloaded 282 times.)

[Maledicted]

Same results, functionally. No beeps, no key codes sent.

TMK:da82c5f3+/LUFA


USB configured.

Keyboard start.
I297 X799 rAA W1243 R1744 ID:0000(2)
S1744 wED
ERR:11
L1797

[hasu]

I looked into some ps/2 port of computers and PS/2 converters kicking around and found some variations of sginal handling there but got no definite useful finding. We will have to do a bunch of trials and errors on timing values without any insight

What does hid_listen show with attached firmware when pressing CapsLock?

 ibmpc_usb_atmega32u4.hex (54.69 kB - downloaded 294 times.)


I wonder what Soarer's converter display on its debug log with the Z-150. Can you try it too?

[acid2000]

I’ve been testing the latest TMK firmware from the github link you shared. The newer ones posted in the forum just spew out garbage.

Overall it is better than when I was testing in September but still has lots of issues.

I’m using an XT 83 key 5150 UK layout with a 5V Pro Micro without reset connected. It’s either connected into a powered USB hub or with a Y cable into a 2.1amp PSU to ensure the voltage doesn’t drop. It has strong pull-ups and the cable to the board itself is short.

I have built a few pro micros with Soarers before. With XTs I usually power it up with the RAW pin and bypass the regulator all together to make sure the power doesn’t dip. I can’t remember if I did this in this one.

It works well with Soarers and is my daily driver.

I’ve got some hid listen output for a few different states.
* No input, power on and don’t touch anything. I left it for around 5m. I was expecting to see nothing but got some things that look like errors.
* Holding down Q. I have two different outputs with slightly different results. I’ve tried a few keys and Q always seems to show issues.
* General typing. I’ve included the ASCII output from the keyboard and the corresponding hid_listen results.

Testing is a PITA for me. For some reason I can’t get the Soarers fw into the bootloader from Windows. I need to bridge the reset pin by hand. I have a few 5150 keyboards and in all of them my adapter is inside the keyboard glued in place and wrapped in electrical tape.

Back in September there were lots of problems with stuck keys. This looks like that is solved (only did a bit of testing) but instead I do get an incorrect keypresses. Is this a timing error?

I’m really hoping that one day soon these issues can be resolved. I’m quite familiar with the TMK codebase now and already have some mods I’ve written I’d like to push up when they’ve been tested.

[hasu]

acid2000,
Thanks a lot for the report.
It sems to read a lot of garbage and phantom key press of Q and not completely miss signals.
The converter read signal at diffrent timing from Soarer's and it can be fussy on signal cleanness. But I don't know why Soarer's can work even with this situation. The setup of same keyboard with same Pro Micro works well with Soarer's firmware?

My fIrmware can still have a silly bug or room to improve but I'd suspect its power line first. I'm not totally sure power voltage drop causes  serious problem like these.

I’m using an XT 83 key 5150 UK layout with a 5V Pro Micro without reset connected. It’s either connected into a powered USB hub or with a Y cable into a 2.1amp PSU to ensure the voltage doesn’t drop. It has strong pull-ups and the cable to the board itself is short.

I have built a few pro micros with Soarers before. With XTs I usually power it up with the RAW pin and bypass the regulator all together to make sure the power doesn’t dip. I can’t remember if I did this in this one.

I got Pro Micro and looked into closely the other day and found something people are likely to misunderstand.

DId you check voltage with multimeter?
First, RAW pin is not for 5V power supply directly from USB line. The name may be not easy to understand but it is used for external power input, not output. You get voltage with diode's forward voltage drop(0.4V typically) on RAW pin.

You have to use VCC pin to power your keyboard, not RAW pin.

And you have to make J1 solder bridged, without this you will get same voltage as RAW pin or lower.

With J1 soder bridge you can get raw power(possibly 5V) from USB line on VCC pin, note that RAW pin doesn't give 5V even with the solder bridge.

Check this wiki for detail.
https://github.com/tmk/tmk_keyboard/wiki/Pro-Micro#j1-solder-bridge

[Maledicted]

I looked into some ps/2 port of computers and PS/2 converters kicking around and found some variations of sginal handling there but got no definite useful finding. We will have to do a bunch of trials and errors on timing values without any insight

What does hid_listen show with attached firmware when pressing CapsLock?

(Attachment Link)


I wonder what Soarer's converter display on its debug log with the Z-150. Can you try it too?

With the firmware you attached, this is the result in hid_listen after having pressed caps lock, which caused the board to stop responding. Before I press caps lock with that firmware, the regular keys seem to function and beep like normal. After pressing it, they stop working:

TMK:b9de6db4+/LUFA


USB configured.

Keyboard start.
I287 X789 rAA W1234 R1735 ID:0000(2)
S1735 L1735 r58 rF0 r58

The keyboard doesn't work at all with Soarer's firmware, so this is all hid_listen reports:

Listening:
R08 wEE

remaining: FFFC


Keyboard ID: 0000
Code Set: 1
Mode: PC/XT

[hasu]

With the firmware you attached, this is the result in hid_listen after having pressed caps lock, which caused the board to stop responding. Before I press caps lock with that firmware, the regular keys seem to function and beep like normal. After pressing it, they stop working:

TMK:b9de6db4+/LUFA


USB configured.

Keyboard start.
I287 X789 rAA W1234 R1735 ID:0000(2)
S1735 L1735 r58 rF0 r58

Didn't you see any output like 'wED' or 'ERR:11' as before?
I expect the error output when seding command of 'set indicator'(ED). Debug prints is not very reliable and you have to try several times perhaps?



I see. R08 means stop bit error. Soarer's do very right thing, it is totally the keyboard's fault.
I wonder why Zenith made the keyboard not compliant to AT protocol

The keyboard doesn't work at all with Soarer's firmware, so this is all hid_listen reports:

Listening:
R08 wEE

remaining: FFFC


Keyboard ID: 0000
Code Set: 1
Mode: PC/XT

[acid2000]

My fIrmware can still have a silly bug or room to improve but I'd suspect its power line first. I'm not totally sure power voltage drop causes  serious problem like these.

I checked again a few things again, I’m not doing any power shenanigans.

Soarers converter works without issue, I swap the fw to TMK and have issues.
I’ve tried on 2 different XT keyboards both built around the same time using two different ProMicros. Same issues.