2 Apr
2009
2 Apr
'09
7:55 p.m.
Does anyone have an HP262x (2621, 2623, etc) terminal (with a DA15
keyboard connector) and a voltmeter to hand?
If so, could you please open up the keyboard (4 screws on the bottom),
plug the keyboard into the terminal, power up, and measure the voltage
either between ;
Pin 16 (+ve) and pin 8 (-ve) of any of the 3 chips in the keyboard
Or
Across the electrolytic capacitor on the keyboard PCB to the left of the
'Tab' key'
Alternatively, measure the voltage between pins 12 (+ve) and 10 (-ve) of
the DA15 socket on the terminal (temrinal powered up, it doesn't matter
whether the keyboard is plugged in or not)
All 3 should be the same voltage. I am guessing it's 12V, but it might be
5V.
Tnakis in advance for any help.
-tony
2 Apr
2 Apr
8:23 p.m.
Does a HP125 do the trick ?
If so I'll make a measurement for you.
-Rik
-----Oorspronkelijk bericht-----
Van: cctalk-bounces at classiccmp.org
[mailto:cctalk-bounces at classiccmp.org] Namens Tony Duell
Verzonden: donderdag 2 april 2009 21:56
Aan: cctalk at classiccmp.org
Onderwerp: HP262x keyboard voltage
Does anyone have an HP262x (2621, 2623, etc) terminal (with a
DA15 keyboard connector) and a voltmeter to hand?
If so, could you please open up the keyboard (4 screws on the
bottom), plug the keyboard into the terminal, power up, and
measure the voltage either between ;
Pin 16 (+ve) and pin 8 (-ve) of any of the 3 chips in the keyboard
Or
Across the electrolytic capacitor on the keyboard PCB to the
left of the 'Tab' key'
Alternatively, measure the voltage between pins 12 (+ve) and
10 (-ve) of the DA15 socket on the terminal (temrinal powered
up, it doesn't matter whether the keyboard is plugged in or not)
All 3 should be the same voltage. I am guessing it's 12V, but
it might be 5V.
Tnakis in advance for any help.
-tony
3 Apr
3 Apr
6:01 p.m.
Does a HP125 do the trick ?
It should do -- it's the same keyboard I think.
If so I'll make a measurement for you.
Sombody else has already done that HP125 and got 5V (or thereabouts).
This suprises me, the keyboard cotnains 3 chips, all from the 4000 CMOS
family (IIRC, a couple of 4028s and a 4051), along with some discretes.
Why use CMOS chips if you're going to run them at 5V?
Perhaps I should explain what I am trying to do. I have an HP120 without
a keyboard. So of course I need to hack something up. The keyboard
interface on the HP120 is the 6 pin RJ11 one at 12V, it's electrically
the same as the HP150 keyboard intereface, but the keys are in a
different electrical arrangement. My 2 thoughts at the moment are either
to mofify a 'spare' HP150 keyboard or to make an interface to an HP262x
keyboard (I have an HP2623 which I cna 'borrow' the keyboard from).
-tony
7:03 p.m.
And using a 2392A keyboard ?
It also uses the HP150 keyboard interface..
The original keyboard looks a lot like the small HP 9816 keyboard..
Maybe you could modify one of those.
-Rik
-----Oorspronkelijk bericht-----
Van: cctalk-bounces at classiccmp.org
[mailto:cctalk-bounces at classiccmp.org] Namens Tony Duell
Verzonden: vrijdag 3 april 2009 20:02
Aan: cctalk at classiccmp.org
Onderwerp: Re: HP262x keyboard voltage
Does a HP125 do the trick ?
It should do -- it's the same keyboard I think.
If so I'll make a measurement for you.
Sombody else has already done that HP125 and got 5V (or thereabouts).
This suprises me, the keyboard cotnains 3 chips, all from the
4000 CMOS family (IIRC, a couple of 4028s and a 4051), along
with some discretes.
Why use CMOS chips if you're going to run them at 5V?
Perhaps I should explain what I am trying to do. I have an
HP120 without a keyboard. So of course I need to hack
something up. The keyboard interface on the HP120 is the 6
pin RJ11 one at 12V, it's electrically the same as the HP150
keyboard intereface, but the keys are in a different
electrical arrangement. My 2 thoughts at the moment are
either to mofify a 'spare' HP150 keyboard or to make an
interface to an HP262x keyboard (I have an HP2623 which I cna
'borrow' the keyboard from).
-tony
7:25 p.m.
And using a 2392A keyboard ?
I don't think I have one of those. What machine does it go with?
It also uses the HP150 keyboard interface..
The original keyboard looks a lot like the small HP 9816 keyboard..
Maybe you could modify one of those.
No, the HP9816 keyboard is very different. it's a 5V interface on 4 wires
(+5V, ground, clock, data (from keyboard)/reset (to keyboard). It also
has the twiddleknob (which is addressed as 1 column of keys -- 7 bits of
motion and 1 of direction IIRC). Interestingly the 9816 keysiwtches are
not wired as a matrix, rather one side of each switch is grounded, the
other goes to a multiplexer input. Scheamtics are on hpmuseum.net.
The closest keyboard i have electrically is the HP150 one. It is the same
interface, similar circuitry, but with differnt key matrix layout. I
think I can hackl that by replacing the scan counter chip in the keyboard
(a 4024) with a little circuit of about half a dozen chips.
The closest keyboard I have for having the right keys is the one for the
HP2623 terminal -- after all a similar keyboard was used on the HP125,
which as you know is a very similar machine to the HP120 (to the extent
that the firmware ROMs are the same, for example). Of course the
interface is quite differen, but a conversion circuit shouldn't be too
hard to buiold (probably cost more for case/connectors than for logic
chips!).
-tony
7:50 p.m.
And using a 2392A keyboard ?
I don't think I have one of those. What machine does it go with? It also uses the HP150 keyboard interface..
The original keyboard looks a lot like the small HP 9816 keyboard..
Maybe you could modify one of those.
No, the HP9816 keyboard is very different. it's a 5V
interface on 4 wires (+5V, ground, clock, data (from
keyboard)/reset (to keyboard). It also has the twiddleknob
(which is addressed as 1 column of keys -- 7 bits of motion
and 1 of direction IIRC). Interestingly the 9816 keysiwtches
are not wired as a matrix, rather one side of each switch is
grounded, the other goes to a multiplexer input. Scheamtics
are on hpmuseum.net.
The closest keyboard i have electrically is the HP150 one.
And not realy the right style.
It is the same interface, similar circuitry, but with
differnt
key matrix layout. I think I can hackl that by replacing the
scan counter chip in the keyboard (a 4024) with a little
circuit of about half a dozen chips.
The closest keyboard I have for having the right keys is the
one for the
HP2623 terminal -- after all a similar keyboard was used on
the HP125, which as you know is a very similar machine to the
HP120 (to the extent that the firmware ROMs are the same, for
example). Of course the interface is quite differen, but a
conversion circuit shouldn't be too hard to buiold (probably
cost more for case/connectors than for logic chips!).
In a lot of cases the right HP keyboards are difficult to find even HIL
keyboards are not always easy (reasonable priced) to find.
If I was you I should go for the interface and the HP2623 keyboard, it looks
better when using the HP120 ;-)
-Rik
4 Apr
4 Apr
8 p.m.
And using a 2392A keyboard ?
I don't think I have one of those. What machine does it go with? Yes I know and on my harddisk , very conveniant I make
a lot of use of them.
And you're right I forgot the knob, just was thinking about the looks, not
the interface.
As you know by now, the frist requriement is that the system works. The
correct 'looks' are very much secondary to that, particularly in a case
like this where I am not going to modify the HP120 in any way. If by some
great chance I get the correct keyboard I can just plug it in.
The closest keyboard i have electrically is the HP150 one.
And not realy the right style. It is the same interface, similar circuitry, but
with differnt
key matrix layout. I think I can hackl that by replacing the
scan counter chip in the keyboard (a 4024) with a little
circuit of about half a dozen chips.
The closest keyboard I have for having the right keys is the
one for the
HP2623 terminal -- after all a similar keyboard was used on
the HP125, which as you know is a very similar machine to the
HP120 (to the extent that the firmware ROMs are the same, for
example). Of course the interface is quite differen, but a
conversion circuit shouldn't be too hard to buiold (probably
cost more for case/connectors than for logic chips!).
In a lot of cases the right HP keyboards are difficult to find even HIL
keyboards are not always easy (reasonable priced) to find. If I was you I should go for the interface and the
HP2623 keyboard, it looks
better when using the HP120 ;-)
I'll probably do both in the end. I spent the afternoon pulling HP
keyboards apart and buzzing out connections/ The HP2623 keyboard is much
as I expected, and the switch matrix layout looks to be right for the
HP120. Which means iternfacing it is just a matter of adding a 7 bit
counter (4024 or similar) and maybe an inverter to step through the keys.
Actually, I'll probably add some buffers too, I don't like driving cables
from the output of flip-flop chips...
The HP150 keybaord is also as I suspected. Modifying that one should be a
matter of removing the 4024 scan counter chip and replacing it with a PCB
contianing a 7 bit counter an an EPROM. What makes it a little more work
is that this keyboarsd runs at 12V, all EPROMs ruin at 5V or thereabouts.
So I'll need to add level shifters. But that's not hard.
-tomy
9:06 p.m.
I'll look around for you and will place a small posting on our Dutch
collectorslist (cvml).
May be something comes up (not a great chance).
-Rik
-----Oorspronkelijk bericht-----
Van: cctalk-bounces at classiccmp.org
[mailto:cctalk-bounces at classiccmp.org] Namens Tony Duell
Verzonden: zaterdag 4 april 2009 22:01
Aan: cctalk at classiccmp.org
Onderwerp: Re: HP262x keyboard voltage
HP 2392A Terminal it has the same keyboard interface as the HP150.
Keyboards look a lot like the HP150 keyboards.
Hmmm.. I don't think I have one, but _somewhere_ I have an
obscrue HP keyboard with a 6 wire RJ11 connector. Maybe I
shoukd try to dig it out.
Ture.
find even HIL
And using a 2392A keyboard ?
I don't think I have one of those. What machine does it go with? Yes I know and on my harddisk , very conveniant I
make a
lot of use of them.
And you're right I forgot the knob, just was
thinking about
the looks, not
the interface.
As you know by now, the frist requriement is that the system
works. The
correct 'looks' are very much secondary to that, particularly
in a case
like this where I am not going to modify the HP120 in any
way. If by some
great chance I get the correct keyboard I can just plug it in.
The closest keyboard i have electrically is the HP150 one.
And not realy the right style. It is the same interface, similar circuitry, but
with differnt
key matrix layout. I think I can hackl that by replacing the
scan counter chip in the keyboard (a 4024) with a little
circuit of about half a dozen chips.
The closest keyboard I have for having the right keys is the
one for the
HP2623 terminal -- after all a similar keyboard was used on
the HP125, which as you know is a very similar machine to the
HP120 (to the extent that the firmware ROMs are the same, for
example). Of course the interface is quite differen, but a
conversion circuit shouldn't be too hard to buiold (probably
cost more for case/connectors than for logic chips!).
In a lot of cases the right HP keyboards are difficult to keyboards are not always easy (reasonable priced)
to find.
Now those I am not short of (both the HP46020 (individual
swtiches) and
HP46021 (capacitve membrane assembly).
If I was you I should go for the interface and
the HP2623
keyboard, it looks
better when using the HP120 ;-)
I'll probably do both in the end. I spent the afternoon pulling HP
keyboards apart and buzzing out connections/ The HP2623
keyboard is much
as I expected, and the switch matrix layout looks to be right for the
HP120. Which means iternfacing it is just a matter of adding a 7 bit
counter (4024 or similar) and maybe an inverter to step
through the keys.
Actually, I'll probably add some buffers too, I don't like
driving cables
from the output of flip-flop chips...
The HP150 keybaord is also as I suspected. Modifying that one
should be a
matter of removing the 4024 scan counter chip and replacing
it with a PCB
contianing a 7 bit counter an an EPROM. What makes it a
little more work
is that this keyboarsd runs at 12V, all EPROMs ruin at 5V or
thereabouts.
So I'll need to add level shifters. But that's not hard.
-tomy
5 Apr
5 Apr
4:59 p.m.
I'll look around for you and will place a small posting on our Dutch
collectorslist (cvml).
Thank you.
May be something comes up (not a great chance).
I think it's unlikely. The HP120/HP125 machines don't seem to be that
common (and after looking at the internals, I can see why, I don't think
it's a particulalry good design!).
Anyhow, I think I've worked out how to interface the HP2523 keyboard and
modify the HP150 one. All (!) I need to do is find the bits and solder it
all up.
-tony
8:05 p.m.
It's a strange desing using a terminal to interface with a CP/M computer.
On the otherhand it's very modern with its Z80 dualcore ;-)
-Rik
-----Oorspronkelijk bericht-----
Van: cctalk-bounces at classiccmp.org
[mailto:cctalk-bounces at classiccmp.org] Namens Tony Duell
Verzonden: zondag 5 april 2009 18:59
Aan: cctalk at classiccmp.org
Onderwerp: Re: HP262x keyboard voltage
I'll look around for you and will place a small posting on
our Dutch
collectorslist (cvml).
Thank you.
May be something comes up (not a great chance).
I think it's unlikely. The HP120/HP125 machines don't seem to
be that common (and after looking at the internals, I can see
why, I don't think it's a particulalry good design!).
Anyhow, I think I've worked out how to interface the HP2523
keyboard and modify the HP150 one. All (!) I need to do is
find the bits and solder it all up.
-tony
6 Apr
6 Apr
6:39 p.m.
It's a strange desing using a terminal to interface with a CP/M computer.
A bit like an H89/Z90 :-)
I was looking at the HP120 design at a lower level. There are, as you
said, 2 processors. They communciate by a simple 1-byte parallel
interface, and AFAIK you can't run user code on the 'terminal' processor.
The terminal side uses that strange video chip I asked about a couple of
weeks back. It interrupts the terminal Z80 on every character line, the
Z80 then loads the start address of that line into the processor and the
apporpriate attributes into a latch. The video chip then does DMA and
fetches 80 byes from memory into an octal 80-bit shift register chip.
This is sent to the character generator and then to the CRT.
Note I mentioned that attirbutes are loaded per line. On each line, a
character is either 'normal' or has the atributes applies (some
combination of underlined, blinking, intensified, etc). Note you can't
have some underlined (but normal brightness) and some intesified (but not
underlined) characters on the same line. There's also no grpahics
capability. This would be reasonable if memory was limited, but that
terminal processor has 16K RAM and 32K ROM hung off it. The serial ports
are on the terminal processor bus
Now the user processor has 64K RAM, and an 8K ROM that can be paged out.
It also has the HPIB itnerface on its bus. And from what I can tell, the
HPIB interface is polled, there are no interrupts.
I mentioend the lack of any form of graphics. I also can't find any way
for a program on the user processor to configure the serial ports. You do
that from the setup menu, which is code running on the terminal
processor.
And while there's an exteded CP/M BIOS function to send a byte to an HPIB
device, there doesn't seem to be one to read from an HPIB device. Yes,
OK, since the HPIB controller is in the user processor I/O space it's
possible to get round that, but since the disk drives are hung off that
HPIB port, you could easily get the thing into a state where it couldn't
access the disks, or worse managed to corrupt them.
I find the sales pitch for this series of machines (on hpmusuem.net) to
be worring due to the number of misconceptions and downright lies it
spreads about comepeting machines. I would have though an HP machine
should be able to stand on its own merits (most, IMHO, can!).
-tony
9:16 p.m.
According to the museum site (www.hpmuseum.net) the electronics of the HP125
and HP120 are mostly the same.
It seems the housing and screen size are the biggest difference between
these machines.
I didn't checked my HP125 out just tested and fixed (bad video ram 4116) it
when I got it, and ran some progs CP/M on it.
I like the looks of it, not the internal design solutions.
Playing with a 200 or 300 series is more fun.
-Rik
-----Oorspronkelijk bericht-----
Van: cctalk-bounces at classiccmp.org
[mailto:cctalk-bounces at classiccmp.org] Namens Tony Duell
Verzonden: maandag 6 april 2009 20:39
Aan: cctalk at classiccmp.org
Onderwerp: Re: HP262x keyboard voltage
It's a strange desing using a terminal to interface with a
CP/M computer.
A bit like an H89/Z90 :-)
I was looking at the HP120 design at a lower level. There
are, as you said, 2 processors. They communciate by a simple
1-byte parallel interface, and AFAIK you can't run user code
on the 'terminal' processor.
The terminal side uses that strange video chip I asked about
a couple of weeks back. It interrupts the terminal Z80 on
every character line, the Z80 then loads the start address of
that line into the processor and the apporpriate attributes
into a latch. The video chip then does DMA and fetches 80
byes from memory into an octal 80-bit shift register chip.
This is sent to the character generator and then to the CRT.
Note I mentioned that attirbutes are loaded per line. On each
line, a character is either 'normal' or has the atributes
applies (some combination of underlined, blinking,
intensified, etc). Note you can't have some underlined (but
normal brightness) and some intesified (but not
underlined) characters on the same line. There's also no
grpahics capability. This would be reasonable if memory was
limited, but that terminal processor has 16K RAM and 32K ROM
hung off it. The serial ports are on the terminal processor bus
Now the user processor has 64K RAM, and an 8K ROM that can be
paged out.
It also has the HPIB itnerface on its bus. And from what I
can tell, the HPIB interface is polled, there are no interrupts.
I mentioend the lack of any form of graphics. I also can't
find any way for a program on the user processor to configure
the serial ports. You do that from the setup menu, which is
code running on the terminal processor.
And while there's an exteded CP/M BIOS function to send a
byte to an HPIB device, there doesn't seem to be one to read
from an HPIB device. Yes, OK, since the HPIB controller is in
the user processor I/O space it's possible to get round that,
but since the disk drives are hung off that HPIB port, you
could easily get the thing into a state where it couldn't
access the disks, or worse managed to corrupt them.
I find the sales pitch for this series of machines (on
hpmusuem.net) to be worring due to the number of
misconceptions and downright lies it spreads about comepeting
machines. I would have though an HP machine should be able to
stand on its own merits (most, IMHO, can!).
-tony
7 Apr
7 Apr
8:07 p.m.
According to the museum site (www.hpmuseum.net) the electronics of the HP125
and HP120 are mostly the same.
Well, the physical layout is different -- the HP125 ahs one large PCB
with all the digital stuff on it, the HP120 has a pair of PCBs, one with
most of the terminal circuitry on it, the other with all the user
processor stuff, the serial ports and the interprocessor communication
latches, etc. But, yes, AFAIK the circuity is very much the same.
The firmware ROMs are the same (at least between a version C HP125 (uses
64K DRAMs like an HP120) and an HP120). The only major component to be
different is the keyboard interface microcontroller. And that's mainly
because hte keyboard interface is different. The HP125 has the DA15
keyboard connecotr, the keyswitch scan is generated by, AFAIK, that
microcontroller. The HP120 has the 6 pin RJ11 socket, carrying clock and
reset lines fro a counter in the keyboard to scan the keys, these signals
coming from the microcontroller.
It seems the housing and screen size are the biggest
difference between
these machines.
I think the PSU and video monitor PCBs are different, but the latter at
least might be similar in design.
Of course the HP125 can take an internal thermal printer, the HP120 is
not documented as being able to do so. But the interface (for an HP2674)
seems to be there, at the same I/O addresses as the interface on an
HP125. And since the terminal firmware is the same, it should work.
Sometime I'll try the cabling from an HP150 and see what hapopens (yes,
I'll check it carefully first...)
I didn't checked my HP125 out just tested and
fixed (bad video ram 4116) it
when I got it, and ran some progs CP/M on it.
I like the looks of it, not the internal design solutions.
YEs, the 'ET head' styling of the HP125 is rather unusual...
The HP120 looks very like an HP9816 (or an HP15 (without the space for
the printer)
Playing with a 200 or 300 series is more fun.
I don;t like the 9000/300 too much, they've got too much custom silicon
in them (or at least the ones I've seen inside do). But I do like the
HP9000/200 machines. Very clean designs...
-tony
5745
days inactive
5750
days old
12 comments
2 participants
participants (2)
-
ard＠p850ug1.demon.co.uk -
hp-fix＠xs4all.nl