Pinout for Compaq Presario 425 modding

I'd like to install a new modern motherboard in a Compaq Presario 425
just for kicks and run Ubuntu. Anyway, the motherboard sort of slides
into a socket which connects it to the integrated crt monitor. Any
ideas on how I could get the schematic or just figure out those "pins"
(It is a one-piece computer so the motherboard lacks normal pins
etc.)




--
Grendal

On Fri, 18 May 2007 04:49:27 +0100, Grendal
put finger to keyboard and
composed:

I'd like to install a new modern motherboard in a Compaq Presario 425
just for kicks and run Ubuntu. Anyway, the motherboard sort of slides
into a socket which connects it to the integrated crt monitor. Any
ideas on how I could get the schematic or just figure out those "pins"
(It is a one-piece computer so the motherboard lacks normal pins
etc.)

The 3 RGB signals should be terminated by 75 ohm resistors inside the
monitor. A DMM should be able to find these. The H & V sync signals
would probably best be seen on an oscilloscope, or maybe a frequency
counter, or you may find a printed legend inside the monitor. Another
way to determine the pinout is to identify the graphics chip and
consult its datasheet. Still another way is to find the monitor's
vertical output chip and work backwards toward the connector.

This site ...

http://home.worldonline.dk/finth/adapters.html

.... states that the Presario 425 uses Cirrus Logic's CL-GD5420
graphics chip.

Here is the datasheet for the "CL-GD542x True Color VGA Family":
http://www.datasheetarchive.com/data...rticle=1162723

See page 12.

Pins 68 and 69 are VSYNC and HSYNC, and 77/76/75 are the RGB pins.

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

Thanks for all that info. It turned out that beneath a glob of glue (or
something now crystalline) I have found markings : r,g,and b in pairs -
so 6 wires are now identified. Also I see an H and V - so those are
identified. Then there is a large Black wire soldered like a ground
wire. So the only remaining unidentified wires are a brown and a white
wire (no markings) which were just above the H and the V wires. I
think the monitor is SVGA. I will go consult all your data and post
more info. My idea is to solder these to a 15 pin video plug so that
this monitor will henceforth be connected to a 1.2 ghz motherboard.
Hope that will work. Any guesses on what the remaining two wires must
be? (This is my first ever look at monitor wiring). Thanks again.




--
Grendal

On Sun, 20 May 2007 18:25:53 +0100, Grendal
put finger to keyboard and
composed:

Thanks for all that info. It turned out that beneath a glob of glue (or
something now crystalline) I have found markings : r,g,and b in pairs -
so 6 wires are now identified. Also I see an H and V - so those are
identified. Then there is a large Black wire soldered like a ground
wire. So the only remaining unidentified wires are a brown and a white
wire (no markings) which were just above the H and the V wires. I
think the monitor is SVGA. I will go consult all your data and post
more info. My idea is to solder these to a 15 pin video plug so that
this monitor will henceforth be connected to a 1.2 ghz motherboard.
Hope that will work. Any guesses on what the remaining two wires must
be? (This is my first ever look at monitor wiring). Thanks again.

One could be digital ground, while the others could be Monitor ID
bits.

See http://pinouts.ru/Video/VGA15_pinout.shtml

The ID pins are either grounded at the monitor end, or not connected.
I'm not sure if these ID bits are actually used. I say this because my
old 486 box has a CL-GD5428 chip which determines the monitor type by
sensing the 75 ohm terminations on the RGB pins. If one of these
terminations is missing, as recently happened to me (open circuit
inductor), then the CL chip thinks that that the monitor is a
monochrome VGA type.

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

So far I have hooked up just four wires, the horizonal, the vert and the
green pair. I was surprised I get this much. I guess my one concern
about the unknown wires is that I don't want to blow out something. But
maybe that is not really likely? I don't have a scope so I may just need
to go with trial-and-error. The last one, (and maybe the middle image),
I guess, is "graphics mode" and needs a bit more info to work.
'[image:
http://img.photobucket.com/albums/v241/pixhopwow/th_DSC04970.jpg]'
(http://img.photobucket.com/albums/v2...w/DSC04970.jpg)
'[image:
http://img.photobucket.com/albums/v241/pixhopwow/th_DSC04973.jpg]'
(http://img.photobucket.com/albums/v2...w/DSC04973.jpg)
'[image:
http://img.photobucket.com/albums/v241/pixhopwow/th_DSC04974.jpg]'
(http://img.photobucket.com/albums/v2...w/DSC04974.jpg)




--
Grendal

On Wed, 30 May 2007 01:15:54 +0100, Grendal
put finger to keyboard and
composed:

So far I have hooked up just four wires, the horizonal, the vert and the
green pair. I was surprised I get this much. I guess my one concern
about the unknown wires is that I don't want to blow out something.

You really should have a separate digital ground for the sync pulses.
As it is now, you are using the green analogue ground for this
purpose.

But maybe that is not really likely? I don't have a scope so I may just need
to go with trial-and-error. The last one, (and maybe the middle image),
I guess, is "graphics mode" and needs a bit more info to work.
'[image:
http://img.photobucket.com/albums/v241/pixhopwow/th_DSC04970.jpg]'
(http://img.photobucket.com/albums/v2...w/DSC04970.jpg)
'[image:
http://img.photobucket.com/albums/v241/pixhopwow/th_DSC04973.jpg]'
(http://img.photobucket.com/albums/v2...w/DSC04973.jpg)
'[image:

http://img.photobucket.com/albums/v241/pixhopwow/th_DSC04974.jpg]'
(http://img.photobucket.com/albums/v2...w/DSC04974.jpg)

The latter looks like a horizontal sync issue.

I'm assuming that you haven't chosen a resolution that drives the
monitor outside its allowable H and V limits. If so, then you could be
observing the same problem that I had when I attempted to hook up an
older monitor to a newer graphics card. The problem in my case was
that the newer card couldn't handle the load on the sync signals when
their polarity switched from positive to negative, as it does during
the change from text to graphics modes. You may need an oscilloscope
to see this.

In my case the fix was to change a resistor and/or capacitor inside
the monitor, as described in the first URL:

http://groups.google.com/group/sci.e...e=source&hl=en
http://groups.google.com/group/comp....e=source&hl=en

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

Here it is with the blue and red pairs too at 640x480 resolution (That
fixed the random mess). I still haven't been able to figure out the
other two wires. I read all the info you provided, though comprehension
was around 30% ! Also, I am wondering if I can tweak the little pots in
the monitor to stretch the image to fill the screen. Or is that not a
good idea?

'[image:
http://img.photobucket.com/albums/v241/pixhopwow/th_DSC04977.jpg]'
(http://img.photobucket.com/albums/v2...w/DSC04977.jpg)
'[image:
http://img.photobucket.com/albums/v241/pixhopwow/th_DSC04981.jpg]'
(http://img.photobucket.com/albums/v2...w/DSC04981.jpg)




--
Grendal

On Fri, 1 Jun 2007 03:37:05 +0100, Grendal
put finger to keyboard and
composed:

Here it is with the blue and red pairs too at 640x480 resolution (That
fixed the random mess).

Sorry, it appears that I may have misled you. I thought your "random
mess" looked like a horizontal sync problem.

I still haven't been able to figure out the
other two wires. I read all the info you provided, though comprehension
was around 30% !

The horizontal and vertical sync signals can be either positive or
negative. I believe that's how the earlier monitors were able to
detect the video mode.

___ ___ ___ +5V ___ ___ +5V
| | | | |_| ~2V
___| |___ |_| 0V

positive negative faulty
sync sync sync

In my particular case, the sync polarity changed from positive to
negative when the graphics card switched from text to graphics mode.
Unfortunately the new card was unable to pull the sync signal down to
0V, so that created a "horizontal tearing" issue for me. I resolved
this by modifying the monitor's circuitry.

Also, I am wondering if I can tweak the little pots in
the monitor to stretch the image to fill the screen. Or is that not a
good idea?

'[image:
http://img.photobucket.com/albums/v241/pixhopwow/th_DSC04977.jpg]'
(http://img.photobucket.com/albums/v2...w/DSC04977.jpg)
'[image:
http://img.photobucket.com/albums/v241/pixhopwow/th_DSC04981.jpg]'
(http://img.photobucket.com/albums/v2...w/DSC04981.jpg)

Tweaking the "vertical size/height" and "horizontal width" should be
OK, although be aware that multisync monitors may have more than one
height or width setting to accomodate the different H and V
frequencies. I wouldn't touch the linearity controls, though. You may
want to use circles and a crosshatch in your test pattern if you are
making big changes. Be aware that the width is sometimes adjusted by
means of a coil with a brittle ferrite slug.

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

Ok, Franc, I'll look for that slug. So far this is working out pretty
well. The monitor can now run at 600X800 and I am trying to figure out
if it can go to a higher resolution or not. Anyway, it looks like it
will be a usable Linux box soon.




--
Grendal