Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Alt key not working

0 views
Skip to first unread message

Computer User@discussions.microsoft.com Old Computer User

unread,
May 30, 2007, 12:00:00 PM5/30/07
to
Windows 98 with a standard Windows keyboard (Key Tronic). Under Control
Panel - Keyboard the language/layout is English(United States) United States
101. The Alt key does not work (neither left nor right). Any ideas?

Thanks!

Don Phillipson

unread,
May 30, 2007, 1:55:09 PM5/30/07
to
"Old Computer User" <Old Computer Us...@discussions.microsoft.com> wrote in
message news:ACD281C0-C8FC-4F84...@microsoft.com...

First borrow another keyboard, so you can test
that the trouble is in your old KB (not in the
motherboard.) Keytronic hardware was not
built to last for ever.

--
Don Phillipson
Carlsbad Springs
(Ottawa, Canada)


PCR

unread,
Jun 2, 2007, 2:58:39 PM6/2/07
to

I think there likely is nothing wrong with the keyboard, if both SHIFT
keys won't work, unless someone wrecked one with overuse & then went on
to wreck the other. Well, here is what I usually post for that, & I see
you already have answered the questions...

(1) What does it say at "Control Panel, Keyboard. Language tab"?

I am "English (United States) United States 101" there. Perhaps
even look at "Control Panel, Regional Settings", where you may wish to
be "English (United States)".

(2) "START button, Run, SysEdit".

Is the Keyboard mentioned in "Config.sys" or "Autoexec.bat"?
Perhaps post those files, then, or just put "REM " (with blank) in
front of that line(s).

(3) Finally... (though personally, I'd hate to do it)...

(a) Boot to Safe Mode.
(Hold F5 as you boot, or hold CTRL for the Startup Menu, & select
Safe Mode from that. Or turn on the Startup Menu at "START, Run,
MSConfig, Advanced button".)

(b) "START, Settings, Control Panel, System, Device Manager tab".

(c) Open the Keyboard branch, by clicking it's plus sign.

(d) Especially if there are multiple devices in that branch... highlight
each one, & click "Remove"-- each one in the branch, because there's no
telling which is the "ghost".

(e) Close up, & boot to Normal Mode.

Hopefully Windows will find the correct drivers, & all will be well.

| Thanks!

You are welcome.

--
Thanks or Good Luck,
There may be humor in this post, and,
Naturally, you will not sue,
Should things get worse after this,
PCR
pcr...@netzero.net


Franc Zabkar

unread,
Jun 2, 2007, 7:54:04 PM6/2/07
to
On Wed, 30 May 2007 09:00:00 -0700, Old Computer User <Old Computer
Us...@discussions.microsoft.com> put finger to keyboard and composed:

As PCR said, it would be unusual for both Alt keys to fail
mechanically at the same time.

To determine whether you have a strange software issue, I'd get away
from the Windows environment and boot to real DOS mode. Type "edit"
and then try Alt-(F)ile e(X)it.

Alternatively, this old DOS utility reports the scan code of each key:
http://www.users.on.net/~fzabkar/SCAN.COM

There will be two codes, one for make (key press), the second for
break (key release). Hit the spacebar to exit the program. Scan.com
can also be run in a Windows DOS box.

- Franc Zabkar
--
Please remove one 'i' from my address when replying by email.

PCR

unread,
Jun 3, 2007, 1:36:42 PM6/3/07
to
Franc Zabkar wrote:
| On Wed, 30 May 2007 09:00:00 -0700, Old Computer User <Old Computer
| Us...@discussions.microsoft.com> put finger to keyboard and composed:
|
|>Windows 98 with a standard Windows keyboard (Key Tronic). Under
|>Control Panel - Keyboard the language/layout is English(United
|>States) United States 101. The Alt key does not work (neither left
|>nor right). Any ideas?
|>
|>Thanks!
|
| As PCR said, it would be unusual for both Alt keys to fail
| mechanically at the same time.
|
| To determine whether you have a strange software issue, I'd get away
| from the Windows environment and boot to real DOS mode. Type "edit"
| and then try Alt-(F)ile e(X)it.

I should have thought of that!

| Alternatively, this old DOS utility reports the scan code of each key:
| http://www.users.on.net/~fzabkar/SCAN.COM
|
| There will be two codes, one for make (key press), the second for
| break (key release). Hit the spacebar to exit the program. Scan.com
| can also be run in a Windows DOS box.

For the L-Alt key, I get "38 B8". For the R-Alt key key, I get "E0 38 E0
B8"-- that's four! R-Ctrl also gives four!

| - Franc Zabkar
| --
| Please remove one 'i' from my address when replying by email.

--

Franc Zabkar

unread,
Jun 3, 2007, 5:55:20 PM6/3/07
to
On Sun, 3 Jun 2007 13:36:42 -0400, "PCR" <pcr...@netzero.net> put

finger to keyboard and composed:

>| Alternatively, this old DOS utility reports the scan code of each key:


>| http://www.users.on.net/~fzabkar/SCAN.COM
>|
>| There will be two codes, one for make (key press), the second for
>| break (key release). Hit the spacebar to exit the program. Scan.com
>| can also be run in a Windows DOS box.
>
>For the L-Alt key, I get "38 B8". For the R-Alt key key, I get "E0 38 E0
>B8"-- that's four! R-Ctrl also gives four!

http://www.pcmag.com/article2/0,1895,1163306,00.asp

"Every time a key is pressed or released, it generates a series of one
or more bytes, called scan codes, representing the key in use and
whether it was pressed or released. Standard keys — the keys found on
the original 84-key IBM-PC keyboard — generate only one scan code.
Extended keys, such as the right Ctrl, right Alt, and the Windows logo
key, generate two: a prefix (0xE0 or 0xE1) to indicate that this is an
extended key, followed by a second code to indicate which key."

PCR

unread,
Jun 4, 2007, 6:42:51 PM6/4/07
to

OK, I get it. "38 B8" is the press & release of any Alt key. But the
R-Alt key gets an "E0" preceding each. It might also have been an "E1".
Very good. OK, thanks.

| - Franc Zabkar
| --
| Please remove one 'i' from my address when replying by email.

--

Franc Zabkar

unread,
Jun 9, 2007, 8:37:59 PM6/9/07
to
On Mon, 4 Jun 2007 18:42:51 -0400, "PCR" <pcr...@netzero.net> put

finger to keyboard and composed:

>Franc Zabkar wrote:
>| On Sun, 3 Jun 2007 13:36:42 -0400, "PCR" <pcr...@netzero.net> put
>| finger to keyboard and composed:
>|
>|>| Alternatively, this old DOS utility reports the scan code of each
>|>| key: http://www.users.on.net/~fzabkar/SCAN.COM
>|>|
>|>| There will be two codes, one for make (key press), the second for
>|>| break (key release). Hit the spacebar to exit the program. Scan.com
>|>| can also be run in a Windows DOS box.
>|>
>|>For the L-Alt key, I get "38 B8". For the R-Alt key key, I get "E0 38
>|>E0 B8"-- that's four! R-Ctrl also gives four!
>|
>| http://www.pcmag.com/article2/0,1895,1163306,00.asp
>|
>| "Every time a key is pressed or released, it generates a series of one
>| or more bytes, called scan codes, representing the key in use and
>| whether it was pressed or released. Standard keys — the keys found on
>| the original 84-key IBM-PC keyboard — generate only one scan code.
>| Extended keys, such as the right Ctrl, right Alt, and the Windows logo
>| key, generate two: a prefix (0xE0 or 0xE1) to indicate that this is an
>| extended key, followed by a second code to indicate which key."
>
>OK, I get it. "38 B8" is the press & release of any Alt key. But the
>R-Alt key gets an "E0" preceding each. It might also have been an "E1".
>Very good. OK, thanks.

You may find the following document interesting.

MS Keyboard Scan Code Specification:
http://download.microsoft.com/download/1/6/1/161ba512-40e2-4cc9-843a-923143f3456c/scancode.doc

FWIW, Microsoft's Intellitype software includes an mskey.exe
diagnostic test which displays and logs the scan codes of each key. I
believe it should work with a standard 101 keyboard.

Here are the scan codes for my Microsoft Wireless Desktop Elite
keyboard, model 1011:
http://www.users.on.net/~fzabkar/MSKeyLog.TXT

BTW, the only scan code that begins with E1 appears to be the
Pause/Break key.

PCR

unread,
Jun 10, 2007, 2:51:24 PM6/10/07
to

Uhuh. I thought it must show up somewhere. OK, very interesting, yea,
OK, thanks. OK, Old Computer User has a few ways now to see whether the
Alt keys are good or bad, including reading a 2.41 MB document
(ScanCode.doc). OK, bye!

| - Franc Zabkar
| --
| Please remove one 'i' from my address when replying by email.

--

Franc Zabkar

unread,
Jun 10, 2007, 10:00:55 PM6/10/07
to
On Sun, 10 Jun 2007 14:51:24 -0400, "PCR" <pcr...@netzero.net> put

finger to keyboard and composed:

>Franc Zabkar wrote:

>| FWIW, Microsoft's Intellitype software includes an mskey.exe
>| diagnostic test which displays and logs the scan codes of each key. I
>| believe it should work with a standard 101 keyboard.
>|
>| Here are the scan codes for my Microsoft Wireless Desktop Elite
>| keyboard, model 1011:
>| http://www.users.on.net/~fzabkar/MSKeyLog.TXT
>|
>| BTW, the only scan code that begins with E1 appears to be the
>| Pause/Break key.
>
>Uhuh. I thought it must show up somewhere. OK, very interesting, yea,
>OK, thanks. OK, Old Computer User has a few ways now to see whether the
>Alt keys are good or bad, including reading a 2.41 MB document
>(ScanCode.doc). OK, bye!

Actually I'm having an issue with an MS Wireless Desktop Elite
keyboard right now. It appears to have a design bug (?) in that it
transmits an incorrect scan code in response to the Sleep button. The
scan.com utility tells me that the following three pairs of bytes are
being sent:

E0 5F <1 second pause> E0 59 <1 second pause> E0 DF

However, Microsoft's spec tells me that the correct codes should be E0
5F (make) and E0 DF (break), ie the E0 59 is spurious.

The symptom is that the additional code causes the machine to wake up
within one second of being put to sleep. Without the diagnostic
utilities and the specification document, I'd still be wondering what
was happening.

PCR

unread,
Jun 12, 2007, 5:36:36 PM6/12/07
to
Franc Zabkar wrote:
| On Sun, 10 Jun 2007 14:51:24 -0400, "PCR" <pcr...@netzero.net> put
| finger to keyboard and composed:
|
|>Franc Zabkar wrote:
|
|>| FWIW, Microsoft's Intellitype software includes an mskey.exe
|>| diagnostic test which displays and logs the scan codes of each key.
|>| I believe it should work with a standard 101 keyboard.
|>|
|>| Here are the scan codes for my Microsoft Wireless Desktop Elite
|>| keyboard, model 1011:
|>| http://www.users.on.net/~fzabkar/MSKeyLog.TXT
|>|
|>| BTW, the only scan code that begins with E1 appears to be the
|>| Pause/Break key.
|>
|>Uhuh. I thought it must show up somewhere. OK, very interesting, yea,
|>OK, thanks. OK, Old Computer User has a few ways now to see whether
|>the Alt keys are good or bad, including reading a 2.41 MB document
|>(ScanCode.doc). OK, bye!
|
| Actually I'm having an issue with an MS Wireless Desktop Elite
| keyboard right now. It appears to have a design bug (?) in that it
| transmits an incorrect scan code in response to the Sleep button. The
| scan.com utility tells me that the following three pairs of bytes are
| being sent:
|
| E0 5F <1 second pause> E0 59 <1 second pause> E0 DF
|
| However, Microsoft's spec tells me that the correct codes should be E0
| 5F (make) and E0 DF (break), ie the E0 59 is spurious.

That's right. That's what I get for the Sleep button: E0 5F & E0 DF. But
my Compaq Internet keyboard has a cord.

| The symptom is that the additional code causes the machine to wake up
| within one second of being put to sleep. Without the diagnostic
| utilities and the specification document, I'd still be wondering what
| was happening.

Is "E0 59" the code for anything? Can that be the code for a spider
walking under an optical mouse?

| - Franc Zabkar
| --
| Please remove one 'i' from my address when replying by email.

--

Franc Zabkar

unread,
Jun 13, 2007, 6:06:51 PM6/13/07
to
On Tue, 12 Jun 2007 17:36:36 -0400, "PCR" <pcr...@netzero.net> put

finger to keyboard and composed:

>Is "E0 59" the code for anything?

I haven't been able to find any info on this code.

The following work is an extensive reference, but it, too, is
inconclusive:
http://www.stanford.edu/class/cs140/projects/pintos/specs/kbd/scancodes-5.html

>Can that be the code for a spider walking under an optical mouse?

Heh.

I thought it might be a code for weak batteries (they measured 1.0V),
but fresh batteries made no difference.

PCR

unread,
Jun 14, 2007, 4:13:50 PM6/14/07
to
Franc Zabkar wrote:
| On Tue, 12 Jun 2007 17:36:36 -0400, "PCR" <pcr...@netzero.net> put
| finger to keyboard and composed:
|
|>Is "E0 59" the code for anything?
|
| I haven't been able to find any info on this code.
|
| The following work is an extensive reference, but it, too, is
| inconclusive:
|
http://www.stanford.edu/class/cs140/projects/pintos/specs/kbd/scancodes-5.html

Well, it does show "E0 59" means CD Player "Previous" on a Safeway SW23
keyboard, & it means "Favorites" on a Samsung Ergonomics Keyboard. Let's
see... my own Compaq Internet Keyboard has plenty of extra keys, but
none of them are "E0 59". Even if you have one, I guess it isn't
sticky, if that code shows up only when you press "Sleep".

|>Can that be the code for a spider walking under an optical mouse?
|
| Heh.
|
| I thought it might be a code for weak batteries (they measured 1.0V),
| but fresh batteries made no difference.

Batteries in the keyboard? (I guess it's possible in a portable.)

| - Franc Zabkar
| --
| Please remove one 'i' from my address when replying by email.

--

0 new messages