CurtPalme.com Home Theater Forum

[JustGreg]

I'm taking my sweet a$$time this go round setting up my 8500 so am nitpicking and questioning every setup step. I want to get the most out of all the cap and resistor updates I've done over the last three months.

I've run into a difference of opinion between my 52 yo eyeballs and a centering template cut out of a manilla office folder. I've checked and rechecked the template and it hits center of the tube mount housing in both H & V directions. Why then, when I look at the grid on each tube does the crosshatch pattern sit farther to the left on each and every tube but center of the pattern is still smack dab on the centering mark? (7 in from each side and 6 up and 6 down) WTF OVER!!??

To clarify, the pattern doesn't 'look' to be in the middle of the available tube scan area from edge to edge. The grids can all be moved more to the right IMO but going by the centering template the grid is up to the edge of the tube on the left side with room to move more on the right side.

Am I missing something in the way the tubes are fixed into the mounts that would cause this visual aberration?? Am I wrong in using a centering template due to inconsistent methods of tube mounting? All 3 images sure look like they could be moved to the right. (looking at the tubes with lenses off and scan reversed) I suppose it could be something with the linearity needing adjustment at this stage of the setup?
Grrrrr.

[Nashou66]

Greg. I don't use a template at all. Reason is C and S linearity can move the center of the grid off center before you even touch them. What you should use is a full field pattern from an external source you plan to use as your main source and put up the full grid or full field pattern and center it not by center of the image but by the sides.
So center the full field edges to the edges of the tube face. Timing from your source are probably not going to be the same as from your internal grids.

The only thing bad about this is sometimes you want to check say the greyscale step pattern or the other internals and they will be off to one side and sometimes off the tube face. This is ok if the image from the source is centered. what i do is just shring the image of the internal down so they fit on the face. do what i have to do and shut of the PJ and re start it and my saved size goes back( do the shrink with the source lock on of course ).

One more point. after your done and actually start to do the geometry and use C and S linearity you might have to use phase again to shift the image back to center. its a back and forth issue sometimes, sometimes you get lucky and do not have to do anything else.

Athanasios

[JustGreg]

Thanks for the reply Nash...you can disregard half of the PM I just sent you.

I DL'd and burned the BR HD patterns from AVS and looked at them earlier and although I'm sure there's enough of them with fullfield to use I wasn't crazy about using patterns designed for color and greyscale adjustment to achieve basic setup needs.
Time to break down and buy HD BR DVE or Avia. Crap. I was cooking right along too!

[Nashou66]

That disc is fine. i would use it. there is a 2.35 image also that has the 16x9 white border but the 2,35 is in black. it also has circles in the four corners to help a bit with geometry. Also i love using the small grid pattern for final convergence, especially for the blend. the smaller the boxes the better you can see if you need to go into the super secrete service menu and make convergence timing adjustments.

Athanasios

[cmjohnson]

Linearity has to be set correctly FIRST or your centering can well be off.


But most people don't know how to set linearity correctly.


It's easy if you have the right test pattern.


What IS the right test pattern?


It's nothing but a simple bit X across the screen from corner to corner.

The way this works is simple: If the linearity isn't perfect, then the legs of the X will not be straight.

The best way to see any imperfections is by standing near the screen and looking at the lines as they go across the screen.

You can accomplish the same thing with a test pattern made of a diagonal crosshatch pattern but I find that a simple X pattern
is simpler to work with. Less confusing to the eye.

Hook up your PC to your projector and project an X pattern that you can make up in PAINT in about 20 seconds.


This essential test pattern is not present on any test pattern DVD or Blu-Ray disc I've ever seen. Some of them do have patterns
that INCLUDE some diagonal lines, but nothing beats the clarity and simplicity of a simple full-screen X.


CJ

[JustGreg]

Thanks guys. I'll take as many tools as I can get.

Nash, I found the patterns you're talking about. I burned the wrong ones to the disc. It wasn't what I thought it was. There was a nominal difference in image center after locking to the correct scanrate.

At this point, before actually making any physical changes (magnetics), I'll segue off on a learning junket and test out Chris' suggestion and see how a corner to corner X looks.

I wonder when I was happier with a CRT pj...before I knew anything at all, or now, when I know just enough to know that I don't know.

Back to the Dungeon.

[cmjohnson]

You should also know that hidden deep in the factory service menus, there are menus which set the position of the test pattern crosshatches. You might need to use them if they're far off.

If you need the password, PM me for it. But be aware that there are a lot of settings in those menus that are not documented at all
and their operation is not necessary self-explanatory. You can get into big trouble if you go in there and start changing values around
at random.

If you need to get in there, I'll give you a guide as to where to find the crosshatch centering adjustments.

There are other very useful adjustments in there such as one which allows you to align the convergence signals to the on-screen zones
that they should be in. The peak of a large vertical adjustment should alway be centered up on a vertical line in the test grid.
If it's not, it can be adjusted.


CJ