Color convergence problem on NEC 4FG
Omer van der Horst Jansen
I've been using a NEC 4FG for over a year now. It works great in 640*480 and
800*600 modes, but at 1024*768 a color convergence problem shows up. The blue
beam is a fraction of a millimeter to the left compared to the red and green
beams, creating a very fuzzy image compared to the ultra-cheap monitor I
have at work.
I actually sent the monitor back to NEC for replacement when I bought the
monitor, but they couldn't reproduce the problem. I guess they simply ran it
at 640*480, or tested only with yellow text.
I recently had the opportunity to peek at an opened 4FG. I noticed a few
control screws on the outside of the inside of the monitor (at a safe distance
from the high-voltage parts, presumably). Can anybody tell me how I can change
the alignment of the blue beam? Which screws need to be turned?
BTW, I'm running the monitor at 72Hz in 640*480 and 800*600, and at 70Hz
vertical refresh rate in 1024*768, conforming to the monitor's specs. I've
tried various video boards, so that probably isn't the problem.
Omer
John P. Mechalas
ago. As a result, I don't know the full capabilities of the monitor, and
was wondering if anyone out here can provide me with some technical
information. Mainly, I am curious as to what the maximum display resolution
is. I know it is capable of 1024x768x256, since that is what I am using
now, but I have never owned a board capable of higher resolutions (1280x1024),
so I don't know if the monitor can handle it.
Second, the MultiSync Plus is capable of receiving a BNC input via the four
jacks located in back labeled "R", "G/SYNC", "B", and "CS". Does anyone
know what the CS stands for? I assume the monitor synchs on green given the
labeling. I would like to hook up a Megapel adapter board from an IBM RT,
but have not been able to get the two to mate (perhaps the Plus isn't capable
of the 1024x1024 resolution?)...and the Megapel adapter only has three BNC
lines, R, G, and B (with synch on green). That leaves me with not enough
cables, so to speak. Or does it?
I've tried the comp.sys.ibm.pc.rt group, but haven't been able to get a
definite answer on the monitor or the card. :) I thought I'd try here,
just in case. :)
Any ideas?
Cheers,
John
--
John Mechalas "I'm not an actor, but
mech...@gn.ecn.purdue.edu I play one on TV."
Aero Engineering, Purdue University #include disclaimer.h
Bob Breivogel
>Hi netters,
>I've been using a NEC 4FG for over a year now. It works great in 640*480 and
>800*600 modes, but at 1024*768 a color convergence problem shows up. The blue
>beam is a fraction of a millimeter to the left compared to the red and green
>beams, creating a very fuzzy image compared to the ultra-cheap monitor I
>have at work.
This is probably within "spec". I have noticed that convergence measurements
are often poorly defined. From the user's standpoint, he is concerned with
largest misconvergene in the area of viewing. If one looks at tests by
PC magazine, they give misconvergence in RMS, over center and "corners"
(wherever that is.) One would expect that there would be places on the
screen that show 3 times this amount. "typical" misconv. figures were
in the range of .1 to .2 mm, giving a very noticable .3 to .6 mm worst
case.
>I actually sent the monitor back to NEC for replacement when I bought the
>monitor, but they couldn't reproduce the problem. I guess they simply ran it
>at 640*480, or tested only with yellow text.
They probably don't see anything wrong with it. I think that at the factory,
they take no more time then needed to achieve barely acceptable conv. Thus
they are used to seeing marginal results (like yours.)
>I recently had the opportunity to peek at an opened 4FG. I noticed a few
>control screws on the outside of the inside of the monitor (at a safe distance
>from the high-voltage parts, presumably). Can anybody tell me how I can change
>the alignment of the blue beam? Which screws need to be turned?
Doing convergence set-up is an art (thus the so-so results on most stock
monitors.) Even Nanao monitors can be tweaked to give better results than
the factory, if you know what you are doing and take lots of time (I speak
from experience.) I don't know about the 4FG, but I have adjusted a 4D.
There are magnetic rings with tabs that stick out on the neck of the tube
just behind the yoke. These rings are in stacked pairs. Moving the tabs
of a pair of rings together decreases the magnetic intensity, while moving
them apart increases it. Rotating the ring pair (leaving angle between the
tabs alone, rotates the field. There are several ring pairs, which have
different effects on convergence. There is also a ring closer to the rear of
the tube that effects "purity", i.e. a red screen being red everywhere
without blotches.
These rings effect "static" convergence. This is a global adjustment that
effects the whole screen. If convergence is poor only on one area, such
as top or left, there are dynamic convergence adjustments. These are
electrical rather than mechanical (pots and coils.) If you want to
adjust convergence, start with the static.
Also, be aware that having the cover off the monitor can expose you to
dangerous voltages (although the 25,000 volts supply is well shielded,
there are other 100+ volt points around.) You might want to wear a pair of
dishwashing rubber gloves for safety.
You will need a screen of white text ("+" symbol) on black background. The
text must cover all areas of the screen that are important. This also
should be at the highest resolution you will run at.
Remove the case of the monitor to gain access to the neck of the CRT and the adjustment rings. Sketch the locations so that you can return them to the original positions if necessary. Get a mirror and face the monitor towards it so you can see the screen while you work from the rear. Turn the power on and let the monitor warm up for 1 hour. Display the text screen at moderate brightness. To turn the rings, you will have to overcome the factory seal, which is a bit of paint on them to prevent movement due
to vibration. Slowly turn one ring-pair at a time (leave the tabs where they are) and watch
the results on the screen. Return to the start position. Do this with the
other rings. By doing this you get some idea which ring does what. You
can then increase the angle between tabs on a given ring pair and can rotate
the PAIR again. Theres is some combination of adjustments that will optimize
the display.
-------------------------------------------------------------------------------
Bob Breivogel "Not the opinions of Intel Corp, just my own"
Intel Corp - Aloha Oregon
brei...@ptd.intel.com
-------------------------------------------------------------------------------
Mark J Marlow
: I have an NEC MultiSync Plus monitor that was purchased used a few years
: ago. As a result, I don't know the full capabilities of the monitor, and
: was wondering if anyone out here can provide me with some technical
: information. Mainly, I am curious as to what the maximum display resolution
: is. I know it is capable of 1024x768x256, since that is what I am using
: now, but I have never owned a board capable of higher resolutions (1280x1024),
: so I don't know if the monitor can handle it.
1024x768 @60 Hz is the max resolution and refresh rate.
: Second, the MultiSync Plus is capable of receiving a BNC input via the four
: jacks located in back labeled "R", "G/SYNC", "B", and "CS". Does anyone
: know what the CS stands for?
CS = composite sync. both horizontal and vertical sync signals share one
channel.
: I assume the monitor synchs on green given the
: labeling. I would like to hook up a Megapel adapter board from an IBM RT,
: but have not been able to get the two to mate (perhaps the Plus isn't capable
: of the 1024x1024 resolution?)...and the Megapel adapter only has three BNC
^ correct.
: lines, R, G, and B (with synch on green). That leaves me with not enough
: cables, so to speak. Or does it?
XL does sync on green, composite or separate. The later being available
only through the multipin connector.
: I've tried the comp.sys.ibm.pc.rt group, but haven't been able to get a
: definite answer on the monitor or the card. :) I thought I'd try here,
: just in case. :)
--
==========================
& Mark Marlow &
& mar...@world.std.com &
==========================