Old trackball won't work on modern laptops

[Jim Thompson]

Joerg skrev:

On Sat, 20 Jan 2007 15:04:40 -0700, Jim Thompson


On Sat, 20 Jan 2007 13:53:14 -0800, David R Brooks


[snip]

Imho, that just about proves it. Those old laptops did have much meatier
power supplies. If you trawl the specs, you'll likely find the old ones
had RS232C, while the new one is RS232E. That's exactly the difference:
12V vs. about 6V, while the intended receiver load has reduced by an
order of magnitude. Since RS232 defines anything over 3V as a valid
signal, the interface-chip makers consider themselves free to provide
just enough power to stagger over the line, & they are in compliance.

Aha! Which says the solution is to provide 12V power to the
track-ball and add some 1488's and 1489's to buffer the levels ;-)

Or, as Joerg would do it... discretes ;-)

A micro would be a suck solution ;-)

...Jim Thompson


Joerg allowed as to how he'd already run serial data thru the port.
What were the levels?

Ok, guys, thanks for that hint about the voltage levels. Whipped out the
meter: It's 5.9V. Dang! Why on earth did they do that?

rs232 tranceiver with voltage doubler running of 3Volt, I guess

Jim, I am indeed itching to see if a discrete solution sans external 12V
supply could be done. Just for sports. Maybe one of those one transistor
switchers, a few turns around a ferrite bead, hmm....

old fashion rs232 transceiver back to back, running of 5volt taken form
USB or PS2 port?

-Lasse

I think the problem is that the track-ball needs 12V, so you need an
interface:

(laptop) 6V data <=> 12V data (track-ball with 12VDC supply)

Joerg needs to measure what RS232 levels exist on his new laptop, then
measure the trackball levels with one of his old machines.

Then we can concoct an interface with a snap of the fingers.

Of course this episode has already cost Joerg $1K in wasted time spent
scratching his ... ;-)

...Jim Thompson

 

[[email protected]]

Jim Thompson skrev:

Joerg skrev:

Jim Thompson wrote:
[snip]

Joerg allowed as to how he'd already run serial data thru the port.
What were the levels?


Ok, guys, thanks for that hint about the voltage levels. Whipped out the
meter: It's 5.9V. Dang! Why on earth did they do that?

rs232 tranceiver with voltage doubler running of 3Volt, I guess

Jim, I am indeed itching to see if a discrete solution sans external 12V
supply could be done. Just for sports. Maybe one of those one transistor
switchers, a few turns around a ferrite bead, hmm....

old fashion rs232 transceiver back to back, running of 5volt taken form
USB or PS2 port?

-Lasse

On a laptop the available USB current is limited, but still might be
around 100mA.

USB is minimum and default 100mA, a device can ask for up to 500mA

Don't know what's legal from PS/2.

from what I can find 275mA

Why not just a battery pack? Joerg is making the problem too
complicated.

carrying around a battary pack is a bit annoying when you already have
a battery and probably the charger for the laptop

I'm not a "gamer", but what _is_ a game port? How does it behave?

+5V, GND, two switches and two potmeter inputs, don't know ho much
current
you can expect. But you won't find one on many laptops.

-Lasse

 

[Jim Thompson]

Joerg skrev:

Jim Thompson wrote:

On Sat, 20 Jan 2007 15:04:40 -0700, Jim Thompson


On Sat, 20 Jan 2007 13:53:14 -0800, David R Brooks


[snip]

Imho, that just about proves it. Those old laptops did have much meatier
power supplies. If you trawl the specs, you'll likely find the old ones
had RS232C, while the new one is RS232E. That's exactly the difference:
12V vs. about 6V, while the intended receiver load has reduced by an
order of magnitude. Since RS232 defines anything over 3V as a valid
signal, the interface-chip makers consider themselves free to provide
just enough power to stagger over the line, & they are in compliance.

Aha! Which says the solution is to provide 12V power to the
track-ball and add some 1488's and 1489's to buffer the levels ;-)

Or, as Joerg would do it... discretes ;-)

A micro would be a suck solution ;-)

...Jim Thompson


Joerg allowed as to how he'd already run serial data thru the port.
What were the levels?


Ok, guys, thanks for that hint about the voltage levels. Whipped out the
meter: It's 5.9V. Dang! Why on earth did they do that?

rs232 tranceiver with voltage doubler running of 3Volt, I guess

Jim, I am indeed itching to see if a discrete solution sans external 12V
supply could be done. Just for sports. Maybe one of those one transistor
switchers, a few turns around a ferrite bead, hmm....

old fashion rs232 transceiver back to back, running of 5volt taken form
USB or PS2 port?

-Lasse

I think the problem is that the track-ball needs 12V, so you need an
interface:

(laptop) 6V data <=> 12V data (track-ball with 12VDC supply)

Joerg needs to measure what RS232 levels exist on his new laptop, then
measure the trackball levels with one of his old machines.

Then we can concoct an interface with a snap of the fingers.

Of course this episode has already cost Joerg $1K in wasted time spent
scratching his ... ;-)

...Jim Thompson

Thinking about the "spec", I suspect all that Joerg needs is the power
supply for the track-ball.

...Jim Thompson

 

[Joerg]

Joerg skrev:

Jim Thompson wrote:


On Sat, 20 Jan 2007 15:04:40 -0700, Jim Thompson



On Sat, 20 Jan 2007 13:53:14 -0800, David R Brooks


[snip]


Imho, that just about proves it. Those old laptops did have much meatier
power supplies. If you trawl the specs, you'll likely find the old ones
had RS232C, while the new one is RS232E. That's exactly the difference:
12V vs. about 6V, while the intended receiver load has reduced by an
order of magnitude. Since RS232 defines anything over 3V as a valid
signal, the interface-chip makers consider themselves free to provide
just enough power to stagger over the line, & they are in compliance.

Aha! Which says the solution is to provide 12V power to the
track-ball and add some 1488's and 1489's to buffer the levels ;-)

Or, as Joerg would do it... discretes ;-)

A micro would be a suck solution ;-)

...Jim Thompson


Joerg allowed as to how he'd already run serial data thru the port.
What were the levels?


Ok, guys, thanks for that hint about the voltage levels. Whipped out the
meter: It's 5.9V. Dang! Why on earth did they do that?

rs232 tranceiver with voltage doubler running of 3Volt, I guess

Jim, I am indeed itching to see if a discrete solution sans external 12V
supply could be done. Just for sports. Maybe one of those one transistor
switchers, a few turns around a ferrite bead, hmm....

old fashion rs232 transceiver back to back, running of 5volt taken form
USB or PS2 port?

Lasse, that's a good idea.

I think the problem is that the track-ball needs 12V, so you need an
interface:

(laptop) 6V data <=> 12V data (track-ball with 12VDC supply)

Joerg needs to measure what RS232 levels exist on his new laptop, then
measure the trackball levels with one of his old machines.

As mentioned above, about 6V.

Then we can concoct an interface with a snap of the fingers.

Of course this episode has already cost Joerg $1K in wasted time spent
scratching his ... ;-)

Sometimes I have to wind down from analog design (I mean the real tough
stuff) and do something non-monetary. Including restoring an old tube
radio for a couple of hours. Which BTW now works for about 15 years. Or
spend 20 hard hours on an old Hammond. It was neglected and totally
seized up when we got it. But now my office time is enhanced by my wife
playing "Oh when the Saints..." and stuff like that on the old Hammond.

 

[Paul Carpenter]

On Sunday, in article
<[email protected]>
[email protected]

Joerg skrev:

Jim Thompson wrote:
53:14 -0800, David R Brooks
[snip]

Imho, that just about proves it. Those old laptops did have much meatier
power supplies. If you trawl the specs, you'll likely find the old ones
had RS232C, while the new one is RS232E. That's exactly the difference:
12V vs. about 6V, while the intended receiver load has reduced by an
order of magnitude. Since RS232 defines anything over 3V as a valid
signal, the interface-chip makers consider themselves free to provide
just enough power to stagger over the line, & they are in compliance.

Aha! Which says the solution is to provide 12V power to the
track-ball and add some 1488's and 1489's to buffer the levels ;-) ....

Joerg allowed as to how he'd already run serial data thru the port.
What were the levels?

Ok, guys, thanks for that hint about the voltage levels. Whipped out the
meter: It's 5.9V. Dang! Why on earth did they do that?

rs232 tranceiver with voltage doubler running of 3Volt, I guess

May even be some form of the laptop battery terminal voltage..

Depends how good it is at dealing with control signals and latencies. Just
the same problem as trying to do a parallel port programmer from USB to
Parallel port converter.

I think the problem is that the track-ball needs 12V, so you need an
interface:

By the sound of things it needs MORE than 6V, could be 7V, 7.5V or 9V
as well as 12V.

(laptop) 6V data <=> 12V data (track-ball with 12VDC supply)

I would just try to use a bench power supply current limited with the
unit stand alone, to determine when the unit starts to give reasonable
RS232 levels out. I assume that as the unit is a serailport device working
with RS232 ports it is actually trying to drive something like 0 to 9V
levels and no -ve voltages.

The problem may also be that 6V is enough but the laptop does not provide
any signals sitting at -ve voltage levels and uses 0/6V signalling. Not
having a negative signal to create a negative rail may be sufficient to
get this to foul up.

If you must do this, use a breakout cable/box on the WORKING system and
scope ALL the signals as no doubt some signal may well be sitting at
a negative level on working system but not on failing laptop.

Joerg needs to measure what RS232 levels exist on his new laptop, then
measure the trackball levels with one of his old machines.
Agreed.

Then we can concoct an interface with a snap of the fingers.

Or even work out if he needs a DC/DC of some form to create a negative rail
as well as possibly a positive rail.

Of course this episode has already cost Joerg $1K in wasted time spent
scratching his ... ;-)

Yes but they can be such fun these challenges.

Like the simple one I did for SWMBO, who had a solar powered water feature
and 12V AC garden lights. Adding small box with relay enabled me to create
a pseudo solar panel from the 12V AC feed so water feature worked when
the garden lights were on.

 

[Jim Thompson]

On Sunday, in article
<[email protected]>
[email protected]


Yes but they can be such fun these challenges.

Like the simple one I did for SWMBO, who had a solar powered water feature
and 12V AC garden lights. Adding small box with relay enabled me to create
a pseudo solar panel from the 12V AC feed so water feature worked when
the garden lights were on.

SWMBO ;-) I sure do miss "RUMPOLE OF THE BAILEY". It hasn't been on
the tube here in quite a while :-(

...Jim Thompson

 

[[email protected]]

On Sunday, in article
<[email protected]>
[email protected]

Joerg skrev:
Jim Thompson wrote:
53:14 -0800, David R Brooks
[snip] snip
old fashion rs232 transceiver back to back, running of 5volt taken form
USB or PS2 port?

Depends how good it is at dealing with control signals and latencies. Just
the same problem as trying to do a parallel port programmer from USB to
Parallel port converter.

snip

not even close to being the same, usb to parallel has to translate from
a serialized
packet format to a parallel interface and usually involves a MCU or
something like that.
losing most timing information that might be used in a parallel port
porgrammer

an rs232 tranciever is a simple voltage translator, it will add a tniy
bit of delay but since
rs232 is async and quite slow it will have no impact.

-Lasse

 

[Paul Carpenter]

On Sunday, in article
<[email protected]>
[email protected]

SWMBO ;-) I sure do miss "RUMPOLE OF THE BAILEY". It hasn't been on
the tube here in quite a while :-(

The secret of happy life, remember to say things at the right time like
"yes dear", always with the correct amount of 'sincerity'.....

Then I know my place in the pecking order, I am not hen pecked as I have
been told to say.

 

[Paul Carpenter]

On 21 Jan, in article
<[email protected]>

On Sunday, in article
<[email protected]>
[email protected]

On 21 Jan 2007 08:17:51 -0800, [email protected] wrote:
Joerg skrev:
Jim Thompson wrote:
53:14 -0800, David R Brooks
[snip] snip
old fashion rs232 transceiver back to back, running of 5volt taken form
USB or PS2 port?

Depends how good it is at dealing with control signals and latencies. Just
the same problem as trying to do a parallel port programmer from USB to
Parallel port converter.

snip

not even close to being the same, usb to parallel has to translate from
a serialized
packet format to a parallel interface and usually involves a MCU or
something like that.
losing most timing information that might be used in a parallel port
porgrammer

an rs232 tranciever is a simple voltage translator, it will add a tniy
bit of delay but since
rs232 is async and quite slow it will have no impact.

ONLY if you need TX and RX data, once modem control signals and turn around
times get involved USB to serial does not always work as the control signals
do not get translated time wise always correctly. The latencies and the
like cause all sorts of problems for serial devices when connected via
USB to serial converters. Especially with potential packet latency for a
single control line change.

USB is not always as deterministic as most people think it is.

 

[Frnak McKenney]

Joerg skrev:
Jim Thompson wrote:
53:14 -0800, David R Brooks

[snip]

Imho, that just about proves it. Those old laptops did have much meatier
power supplies. If you trawl the specs, you'll likely find the old ones
had RS232C, while the new one is RS232E. That's exactly the difference:
12V vs. about 6V, while the intended receiver load has reduced by an
order of magnitude. Since RS232 defines anything over 3V as a valid
signal, the interface-chip makers consider themselves free to provide
just enough power to stagger over the line, & they are in compliance.

Aha! Which says the solution is to provide 12V power to the
track-ball and add some 1488's and 1489's to buffer the levels ;-) --snip--
Ok, guys, thanks for that hint about the voltage levels. Whipped out the
meter: It's 5.9V. Dang! Why on earth did they do that?

--snip--
I think the problem is that the track-ball needs 12V, so you need an
interface:

(laptop) 6V data <=> 12V data (track-ball with 12VDC supply)

Joerg needs to measure what RS232 levels exist on his new laptop, then
measure the trackball levels with one of his old machines.

Then we can concoct an interface with a snap of the fingers.

Of course this episode has already cost Joerg $1K in wasted time spent
scratching his ... ;-)
...Jim Thompson

Thinking about the "spec", I suspect all that Joerg needs is the power
supply for the track-ball.

If the laptop port is at least going negative (e.g. is using +/-6V
vs. +6V/0V), could you use a DS275 power-stealing interface?


Frank McKenney, McKenney Associates
Richmond, Virginia / (804) 320-4887
Munged E-mail: frank uscore mckenney ayut minds pring dawt cahm (y'all)

 

[Reinardt Behm]

Joerg ([email protected]) wrote at Sunday 21. January
2007 01:33 in comp.arch.embedded:

I've looked. It wasn't :-(((

Just to verify that the trackball works when connected to your PC:
Open Hyperterminal, set it up for 1200Baud, 8bit, no hardware handshake.
When you move the ball or press a button you should see some char (non
printable) arriving.
If not, I would suspect, that the hardware port drivers are not able to
deliver enough power.

 

[Robert Latest]

["Followup-To:" header set to sci.electronics.design.]

SWMBO ;-) I sure do miss "RUMPOLE OF THE BAILEY". It hasn't been on
the tube here in quite a while :-(

But the books are better still. A new one's just come out:
Rumpole and the Reign of Terror.

 

[[email protected]]

Paul Carpenter skrev:

On 21 Jan, in article
<[email protected]>

On Sunday, in article
<[email protected]>
[email protected]
:
On 21 Jan 2007 08:17:51 -0800, [email protected] wrote:
Joerg skrev:
Jim Thompson wrote:
53:14 -0800, David R Brooks
[snip] snip
old fashion rs232 transceiver back to back, running of 5volt taken form
USB or PS2 port?

Depends how good it is at dealing with control signals and latencies. Just
the same problem as trying to do a parallel port programmer from USB to
Parallel port converter.

snip

not even close to being the same, usb to parallel has to translate from
a serialized
packet format to a parallel interface and usually involves a MCU or
something like that.
losing most timing information that might be used in a parallel port
porgrammer

an rs232 tranciever is a simple voltage translator, it will add a tniy
bit of delay but since
rs232 is async and quite slow it will have no impact.

ONLY if you need TX and RX data, once modem control signals and turn around
times get involved USB to serial does not always work as the control signals
do not get translated time wise always correctly. The latencies and the
like cause all sorts of problems for serial devices when connected via
USB to serial converters. Especially with potential packet latency for a
single control line change.

USB is not always as deterministic as most people think it is.

We are talking about two different things, you are talking about a usb
to serial converter and
yes they change timing.

I was talking about trancievers e.g. max232, wired back to back and
_powered_ from USB 5volt
to go from rs232 to TTL and back to rs232 levels with a more powerful
tranciever

-Lasse

 

[Tom Lucas]

Wine?

In a bit of a thread hijack, how good is Wine with odd legacy hardware?
I've got an ancient Noral 80188 emulator that needs a proprietory ISA
card in the PC and will only run under DOS. Would Wine let me use it on
a new PC or do you reckon that it would a be a bit too proprietory for
that?

 

[Paul Burke]

In a bit of a thread hijack, how good is Wine with odd legacy hardware?
I've got an ancient Noral 80188 emulator that needs a proprietory ISA
card in the PC and will only run under DOS. Would Wine let me use it on
a new PC or do you reckon that it would a be a bit too proprietory for
that?

I don't think even Linux software will add an ISA slot to your motherboard.

 

[Tom Lucas]

I don't think even Linux software will add an ISA slot to your
motherboard.

<mutter, mutter> Well yes, I shall have to cross the hardware bridge too
but assuming that I have an ISA slot then is Wine likely to allow me use
it for the emulator? Of course, I am now less certain that I could
purchase a motherboard with ISA on it - do they still make them?

 

[Tom Lucas]

<mutter, mutter> Well yes, I shall have to cross the hardware bridge
too but assuming that I have an ISA slot then is Wine likely to allow
me use it for the emulator? Of course, I am now less certain that I
could purchase a motherboard with ISA on it - do they still make them?

Oh yes, a quick Google has revealed quite a few.

 

[Jan Panteltje]

<mutter, mutter> Well yes, I shall have to cross the hardware bridge too
but assuming that I have an ISA slot then is Wine likely to allow me use
it for the emulator? Of course, I am now less certain that I could
purchase a motherboard with ISA on it - do they still make them?

There is a third way, no guarantees, and slow, but I installed
http://www.virtualbox.org/
virtualizer in Linux, started it, and did a full win98 (from original disk)
install.
(Note you have to remove any other 'real' win98 then, as MS allows only one
copy per PC).
That worked, also with MS explorer, and could connect to the internet
via the gateway.
Then I installed DRDOS 7.1 (I think it was 7.1), and it runs too.
The CDRom access does not work for me from win98 in the virtualizer,
but I am sure that is some config issue, as I installed from it.

So you get Linux, with in a window MSDOS, and in an other window MS Windows.

As to 'wine' in Linux, it has no problem with IO to the only ISA card I have.
(But that one only talks via one IO address).
To run the virtualizer you need a _d*mn_ fast PC, else it is too slow.

 

[Tom Lucas]

On a sunny day (Mon, 22 Jan 2007 11:11:48 -0000) it happened "Tom
Lucas"


There is a third way, no guarantees, and slow, but I installed
http://www.virtualbox.org/
virtualizer in Linux, started it, and did a full win98 (from original
disk)
install.
(Note you have to remove any other 'real' win98 then, as MS allows
only one
copy per PC).
That worked, also with MS explorer, and could connect to the internet
via the gateway.
Then I installed DRDOS 7.1 (I think it was 7.1), and it runs too.
The CDRom access does not work for me from win98 in the virtualizer,
but I am sure that is some config issue, as I installed from it.

So you get Linux, with in a window MSDOS, and in an other window MS
Windows.

As to 'wine' in Linux, it has no problem with IO to the only ISA card
I have.
(But that one only talks via one IO address).
To run the virtualizer you need a _d*mn_ fast PC, else it is too slow.

Sounds interesting, I still have original MS-DOS 6.22 floppies around
somewhere so I might even be able to avoid having to do a Win98 install
at all. Maybe that would ease some of the speed requirements but the PC
I plan to run it on is pretty nippy anyway.

 

[Chuck Harris]

Sounds interesting, I still have original MS-DOS 6.22 floppies around
somewhere so I might even be able to avoid having to do a Win98 install
at all. Maybe that would ease some of the speed requirements but the PC
I plan to run it on is pretty nippy anyway.

If you are running a dos program, use linux, and dosemu, which comes with
linux. I use it to run all of the old OrCAD tools from linux. dosemu can
use MSDOG, or it can use FreeDOS. I use FreeDOS, as it is open source, and
works better than the real thing.

-Chuck