AVS Forum banner

Status
Not open for further replies.
1 - 20 of 25 Posts

·
Registered
Joined
·
2,088 Posts
Discussion Starter · #1 ·
I have zeroed all converence and alignment values and the image I get on the tube face looks nothing like a rectangle...nothing at all! I can correct it to a rectangle using heavy pincushion and keystone but I shouldn't need to use 100% on any of these adjustments...should I??


I know all my plugs are correctly plugged in, I'm pretty stumped on what the issue could be and I'm considering a eeprom flush!!


WHat did your raster look like after zeroing everything?


Ben
 

·
Registered
Joined
·
2,837 Posts
It will look nothing like a rectangle. Expect a trapezoid with unequal, curved edges when everything is zerod on a NEC. You're going down the right road. The idea is to take that awful shape and turn it into something usable while at the same time minimizing the amount of adjustment of each control. Do it stepwise. I'll rough it out. If you have more specific questions about each step, somebody here will probably be able to help you out.


Register a new setting memory of your input signal.

Set phase

Center the rasters in the tubes.

Mechanically aim the tubes.

Focus the guns and lenses


Work on the green gun only and center a "center cross" pattern on your screen


Adjust vertical size and linearity while paying attention only the center vertical line. Make the top and bottom extent of the center vertical line equal.


Adjust horizontal size and linearity while paying attention to only the center horizontal line. Make the left and right extents equal.


Use the bow controls to remove curvature of the center vertical and center horizontal lines.


Use tilt to make the center vertical line well.... vertical. The center horizontal lines needs to be horizontal.


Now that you've established the centering, size, linearity, bow, and tilt of the center cross you can move on to a crosshatch pattern and begin work on the periphery of the image.


Use the keystone and keystone balance controls to make the trapezoid rectangular.


Use pincushion and pincushion balance to straighten the edges.


Turn on the red gun and go through the same steps again. Only this time also align the red on top of the green.


Turn off the green and do the blue gun. Align blue on red.


Finally, recheck with all three guns on and tweak as needed.


As the final step, use point convergence SPARINGLY to fix leftover geometry and convergence issues. You will not achieve perfect geometry. Some compromise is acceptable. If you push for too much perfection you will overstress the convergence circuits and invite early projector demise.


The above is a very rough outline. Keep at it. Don't run controls up and down too far. Don't do too much work in one setting. Every time you redo a CRT you'll get better at it. You will undoubtedly "suck" at setup the first time. Stay calm. Be methodical and stepwise. Learn how each control affects the picture. You will gain considerable skill over time. Read this forum for more advanced tips once you get some idea of the basics. This is an ongoing relationship you'll have with your CRT. It's going to take time. A lot more time than plunking down a digital and twisting a lens.
 

·
Registered
Joined
·
8,079 Posts
Guy,


Do you ever find that you have to go back and re-centre the raster after your've roughed out the linearity? (Which moves the edges of the raster).


I did it basically the way you posted above, centered the raster with setting's zeroed, but found that the rasters were not centered after alignment/convergence.


Then I did it again, but roughed out linearity early on, and recentered the rasters after that. It all fell into place this time.


Is this normal? Or did I just make it up as I went along?...


Sorry to slightly hijack your thread Ben :)


Mark
 

·
Registered
Joined
·
2,837 Posts
Yes. The controls interact, but the centering controls have a greater effect on the raster centering than the linearity controls. It's an iterative process. What you did is fine.
 

·
Registered
Joined
·
2,088 Posts
Discussion Starter · #5 ·
The problem I have is that the raster goes up and to the left in an almost uncorrectable fashion. I have to use almost 100% for pin and key to fix it. If I zero everything and look at the cross hair adjustment you can see that on green the horizontal line moves up near the left edge and the vertical line moves left near the top edge, it's almost like a magnet is pulling the image up and away. I really think this is not normal and I'm wondering what other options I have.


The one adjustment that I don't understand is phase, what am I looking for when adjusting it?


Ben
 

·
Registered
Joined
·
2,837 Posts
Those extreme edge distortions on the built-in crosshatches are typically outside the active imaging area. Don't worry about them.


Phase does two things.


1. Aligns the internal pattern generator to match in current video signal's timing charcteristics. You achieve this by making the vertical lines of the pattern go left/right until the center vertical line is centered on the screen.


2. Aligns the convergence system timing to match the current video signal. You achieve this by moving the "blip" left/right until it is centered on the center vertical line of the test pattern.
 

·
Registered
Joined
·
2,088 Posts
Discussion Starter · #7 ·
Guy,


thanks for the tips. But I believe this geometric distortion to be well within the active image. It's always there, as if a magnet was pulling the whole left side of the screen up and away. At any rate, I'll be redoing my setup from scratch again, I hope that will solve the problem but I have my doubts.


Do you think a table mounted NEC should require 96% keystone and 100% pincushion? I should hope not....I won't rule out an eeprom reset at some point, this problem is very frustrating!



Thanks,


Ben
 

·
Registered
Joined
·
609 Posts
Is there any reason why you shouldn't use an external test pattern direct from your video source to do the initial set up your NEC PG?
 

·
Registered
Joined
·
2,837 Posts
No. An external pattern is preferable if available.
 

·
Registered
Joined
·
337 Posts
Guy,


Comforting words !

My current setup is my third, and though not completely awful it is still far from perfect.

In the process I think I have learned a bit more every time I've done it, but, like you say, it does require a lot of time. And patience.


Thomas
 

·
Registered
Joined
·
2,041 Posts
How is the shape of your dots?

A totally misadjusted astig magnet can cause really weird geometric distortions.

Roland

P.S.:

Is your PJ in level or is it tilt upwards?
 

·
Registered
Joined
·
2,088 Posts
Discussion Starter · #12 ·
The pj is level and there was some misadjustment to the astig magnets, but after correcting that I still had the upper left pull. Also, how is it possible that I need 100% keystone? the image on my screen looks like a flowerpot or something....I know it's not right, yet I can't figure out how to fix it!


Ben
 

·
Registered
Joined
·
2,837 Posts
A high horizontal keystone setting is common. It also increases the closer you position the projector to the screen. Too close and you run out of range.
 

·
Registered
Joined
·
165 Posts
benareeno,

The distortion you describe sounds exaclty like what you would experience if the projector was not perpendicular to the screen. For instance if the projector were tilted higher left than right then angled slightly.

In other words if the distance from the top left corner of the screen to the centerpoint of the green lens was longer than the distance from the top right corner of the screen to the centerpoint of the green lens. It creates a flaring of the image in one corner.

This is how I discovered that the first step in calibration is placement of projector to the screen, simple but very important.

*My most important lesson was don't correct electronically that which can be corrected mechanically.
 

·
Registered
Joined
·
299 Posts
Ben,


Are you looking in the tube or on the screen. The image will not look correct on the tube face. If you can't get it correct on the screen, then you might have to make a different adjustment (magnets). Also, make sure that you have the orientation correct. The projector is mounted for front floor projection. Make sure the sweep plugs and the software settings are correct.


Pete
 

·
Registered
Joined
·
2,088 Posts
Discussion Starter · #16 ·
Doktor,


I agree 100% except this distortion is also present when I'm looking at the raster on the tubes....


As for the plugs, I've checked them many times, I even reversed them to make sure that I wasn't wrong. Orientation in the menu is desk front...I really think that I'm building on corrupt memory here, I disabled point, do you think I should pull the point board?


Do any of you think it's the c-drive? Perhaps yoke adjustment?


I think an eeprom reset will be in my future if I can't resolve this issue.


Ben
 

·
Registered
Joined
·
299 Posts
I think you can zero out all point by highlighting the point convergence in the menu (don't select, just highlight) and then selecting control-init. You have to do this for Green, Red, and Blue.


Someone want to verify this??
 

·
Registered
Joined
·
8,079 Posts
crtl+normal, then select the cancel option, I think you'll find.


It does sound like the c-drive plugs are wrong to me or....


Perhaps you may have to reset the eeprom.....
 

·
Registered
Joined
·
2,088 Posts
Discussion Starter · #19 ·
If I pull the point board out, that must be good enough...right?


I agree, it does sound like the c-drive plugs are wrong, but they're not...I wish they were.


eeprom reset is not an appealing option, but I'll do it if I have to.


Also, what about a bad stk chip? What are the symptoms for that?


Ben
 

·
Registered
Joined
·
8,079 Posts
Disable it in the menu first - settings menu I think.
 
1 - 20 of 25 Posts
Status
Not open for further replies.
Top