Compression Issue - NeoTec NT-S501

[jpv108]

Hi all,

I'm running a 25" NeoTec NT-S501 standard resolution monitor and I am getting a compression issue.  The far left and right sides of the screen are a bit compressed while the right-middle of the screen is actually a bit stretched.  I can post pictures of the screen with a test grid if needed.

I have already installed a new cap kit that I purchased from Bob Roberts (had a local arcade shop do the install).  While that fixed some of the minor clarity issues, it did not fix the compression issues.

From the research I can gather on the internet, it seems that I may be having a linearity issue.  I have looked at the chassis and I can locate the vertical linearity (V LINE) screw.  However, I cannot locate the horizontal linearity screw, if there is a such thing.  Am I looking for the wrong label? I have found H HOLD and H PHASE on the chassis, but not H LINE.  Below is a PDF of the NeoTec NT-2700/2701/2702 service manual.  While not the same model as the s501 that I have, the chassis diagrams look very close to the NT-S501.

http://www.arcade-museum.com/manuals-monitors/Neotec%202700%20Full%20Family%20version.pdf

Is there a H LINE screw to adjust that I am not seeing?

Also, could this be something else like a horizontal width coil adjustment or maybe I need new horizontal capacitors installed (I don't believe I had those replaced when doing the cap kit)?

Any information you guys could provide would be greatly appreciated!

[jpv108]

Attached is a picture of a test grid on my screen.  As you can see, the left and right sides of the screen are compressed compared to the middle.

[ed12]

thats called skewing or key-stoneing
the horz pahse may or may not help u to a CERITAN RANGE
do not go to far as it will drive the horz freq out of range
skewing is a common problem as is key stoneing
it take's tweaking

to be that exact is just about not worth it

ed

[lilshawn]

on flat/near flat tubes it's just about impossible to get perfect horizontal linearity. It's just the design of the tube.

Any messing around would probably just make it worse in other areas, or worse, blow something on the board by having to change and try to tweak cap values. I've had (still have) 2701's that have perfectly round circles in the middle, and ovals on the left/right sides.

[jpv108]

That's a shame.  Too bad I can't get it just a little bit better.  I was really hoping there was a fix for this out there. 

This monitor is beautiful except for this linearity issue.  I have it hooked up to a PC currently running MAME and it becomes noticeable on games like Mortal Kombat when one fighter is wider than the other.

Thanks for the responses!

[MonMotha]

Indeed, the linearity is about as good as you can hope for on a modern "dynaflat" or "pureflat" type tube.  Just getting it that good takes a surprising amount of effort that I'm a bit surprised somebody managed to pull off with an analog design.  The keystoning could probably be fixed; most digital monitors at least have an adjustment for it, but on an analog monitor it may not be worth dealing with.  Just overscan the thing a hair and live with it, if that's the case.

If you really are using this only with a PC, there's no reason (in theory) you can't have the PC do some transforms to fix all this.  You could even keep the video timing 100% accurate since you're only operating in the horizontal direction.  I don't know of any off-the-shelf software solutions to make this happen, though.  Might be an interesting project for somebody with some time on their hands.

CRTs have some great properties, especially for gaming, but perfect geometry ain't one of 'em.  Even my GDM-FW900 (MSRP when new: ~$2,500) sets don't have perfect geometry despite having a bazillion (not exaggerating) adjustments available.  For perfect geometry, you need a defined-pixel technology like any modern "panel" (plasma, LCD, OLED).

[lilshawn]

the problem is, to get a nice even picture on a flat tube requires some very accurate timing...timing you can get with a digital monitor chassis.

it's something like this:

when the beam is at the left side of the monitor it sweeps towards the right hand side. when it does this on a curved tube, it's a nice steady pace because the beam is always the same distance to the monitor face. on a flat/near flat tube, this isn't the case, the distance to the middle is very much shorter than it is on the left and right sides. because of this, the beam travel speed must be slow at the left, get faster towards the middle, then slow down again as it reaches the right.

usually they (try to) compensate for this by using a bi-polar capacitor to time the circuit. the cap is charged and discharged with positive and negative voltages at a rate that helps get the speed close to what is needed to square everything up. the speed used to maintain the picture size using this system is very linear and it would be a triangle or saw waveform (depending on the system used.)

on a flat tube using this system results in the pic you have...slightly squished on the left and right and fine in the middle. what needs to happen is the timing waveform needs to be logarithmic. it needs to be VERY quick at the beginning, bleed off quickly for the first part and at the same time  slowly discharge through to the middle. makes no sense right? here is a pic:



the yellow wave is a logarithmic waveform. the green, a linear. it is very difficult to create that yellow waveform with a bunch of analog components (capacitors, resistors etc) you need to use a digitally programmed timing circuit to achieve this.