Wide scan angles

The faceplate is flat, so the center is nearer to the gun than the edges so the beam goes out of focus as you move away from the center. The other problem is non lineararity a little basic trig will show you that.

I would have thought focus modulation would deal with that no problem. Even a look up table could be used if it gets that difficult.

Sure, but again thats already successfully dealt with on 110 deg scans, so why not 150 deg? The maths expressions are the same.

The shadowmask would block progressively more of the beam at steep angles, since it must be the same thickness metal, but the beam passes at a widely varying angle. But again todays tubes have plenty of output, and can afford to lose a percentage of it.

What are the intractable problems?

NT

Its not easy to make a gun with a wide focus range, not is it easy to generate the required correction signal.

The expresion is the same but if you punch in the numbers you will see just how bad the problem gets as the widthn goes up.

You want color as well, now you have the added problem of the mask hole casting an eliptical spot at the edges, you gun spot will also allready be eliptical as well, cutting the brightness down considerably so you now need intensity modulation as well. The screen phosfor dots will be spaced out further at the edges unless the shadowmask is closer to the screen at the edges which makes it a hard shape to make.

Could it be that the 'center of deflection' is not a point, i.e. the beam in different directions doesn't originate from a point, but from a place that varies with deflection? And this screws up the color purity?

Thomas

Hi aagin.

Thanks for your feedback. I'm looking to push this one, see what might be doable. Lets see...

snipped-for-privacy@aol.com wrote:

OK. Focus modulation, a lookup table. Its more money, but may be warranted. V_focus only need be within so many %, the needed accuracy determines the amount of data in the table, which would then be low pass filtered to reduce the small errors.

Focus range... lets get optimistic and say we have 6 guns instead of 3. I'd need to think about that one, but since scan can be so accurately controlled as it is on a CRT monitor, I'd think switching the guns during scan might be workable. Fast fading from one gun to the next woudl eliminate light or dark lines with setup errors. But again, the price keeps going up.

yes, its not ideal. 150 deg would be rather optimistic. We can also tweak the video signal to partially compensate.

For example, say we wanted to display a series of pixels of brightness

4,4,7,9,7,5,6,8,4,1,2,2,2 but our beam in fact was 2 pixels wide, not

1. To get the above brightness output, we would drive the beam with

4,4,4,10,8,6,4,8,8,0,2,2,2

Now the wanted brightness output and resolution are obtained for nearly all pixels, but not quite all. Thus a 2 pixel wide beam can give us almost the resolution of a 1 pixel wide beam. We get full resolution except when switching between >50% brightness and dark. Given that the tube will generally be operated well under its peak output, even that appears workable, full resolution transitions between white and black, or at least near white and black.

However... the reason I'm asking is I've thought of something that on the face of it would appear to solve this issue.

that would not be too hard, the multiplier singal could be obtained analoguely from the frame and line scan signals.

Maybe a flat shadowmask with a slight screen curve.

However, this would not be a problem with what I've come up with. But beam focussing would be.

thanks,

NT

It would affect the beam focussing, and the final focus can be dealt with by modulating V_focus. But the focus at the shadowmask is equally important...

thanks, NT

actually much of the flat face plate focus issues and deflection linearity issues are solved in old expired Tektronix patents (from doing it for oscilloscopes).

--
JosephKK