I was playing around with the focus magnet on my Fleetwood today and after experimenting for a bit, I was wondering what the common practice is for best picture adjustment.

Essentially you can reach three states - underfocused, overfocused, and exactly focused.

I find that while it's easy to focus exactly (my set, anyway) and get incredibly sharp lines, in practice this produces a rather difficult picture to watch because of the sheer amount of "dark" space between the raster lines. Thus, it's easier on the eyes to slightly de-focus the electron beam so the lines blend together a little better. Therewith, you can overfocus where you have the beam "cross" itself before hitting the screen, or you can "underfocus" and have the focal plane be in front of the actual phosphor screen.

What do the folks here do or prefer for their sets? Feel free to turn this into a discussion on the finer aspects of focusing a B&W, this isn't a particular answer I'm looking for but rather the theory/comments and best practices surrounding the setup of a tube-based tv.

I can't quantify it, but yes, I often back off on the focus a little to make a slightly softer, more natural picture. TVs are not all the same in this regard. On some of them, anyhow, making the focus too crisp gives everything an artificial or slightly unpleasant appearance.

Phil Nelson

Generally I focus it right on, however if the vertical is stretched, the raster trace lines will really be noticeable. If your playing a DVD or watching broadcast with a digital converter box, where the picture is in the 16:9 format it might be that your trying to make the picture fill the screen by turning up the vertical height, stretching the raster way too tall.
Nearly all standard def TVs, will show trace lines when you look at the screen closely. Guess they were meant to be viewed at a distance. Try turning down the vertical height, and give it just a little bit of over-scan. It might be that the vertical height and or vertical lin, is set too high.

Generally I focus it right on, however if the vertical is stretched, the raster trace lines will really be noticeable. If your playing a DVD or watching broadcast with a digital converter box, where the picture is in the 16:9 format it might be that your trying to make the picture fill the screen by turning up the vertical height, stretching the raster way too tall.
Nearly all standard def TVs, will show trace lines when you look at the screen closely. Guess they were meant to be viewed at a distance. Try turning down the vertical height, and give it just a little bit of over-scan. It might be that the vertical height and or vertical lin, is set too high.

Linearity is pretty good, I've been tweaking with that for a while to get it right. As for height, I don't think it's too high - If I crank the brightness I can see the top/bottom of the raster and I have a bit of space on both the top and bottom.

I know someone who worked for Tektronix back in the analog TV days. They had a project to use a very high resolution tube for a standard-definition NTSC color monitor, the idea being that it would show the full capability of the studio signal and not hide any defects in the signal either. They killed the project after the first prototype because the visible line structure was intolerable to users.

When defocussing the thing to watch for is if the spot develops a halo one way or the other, reducing contrast. You also might see if one side produces more even focusing from shadows to highlights than the other. If not, the choice is up to you.

By the way, too much sharpness was even less desirable when making kinescope recordings. If there was a line structure on the kinescope print, then the telecine camera would resolve it and cause a moiré pattern when the show was rebroadcast.

The CBS "EVR" pre-recorded video system that used microfilm to carry images was recorded using an electron beam that was wobbled vertically a precise amount to fill in the spaces between lines.

Too much sharpness also results in poor pictures when taking a still picture off the screen, since you are simultaneously trying to expose for the average brightness while accommodating lines that are much brighter spaced by black. The result is that the resulting picture looks very compressed in contrast because the video lines are actually all overexposed, while the average brightness is correct.

The human visual system does an amazing job of seeing both the average brightness and the line structure, due to parallel processes in the brain; but if the lines are too sharp, they become distracting.

It could be your set has a problem with line pairing, instead of interlacing, the scan lines are more or less on top of each other making them more visible.

It could be your set has a problem with line pairing, instead of interlacing, the scan lines are more or less on top of each other making them more visible.

Well, this thread just got a whole lot more interesting. Now that you mention it, there _does_ seem to be a lack of scan line density that doesn't seem to match up to the 525 standard.

I'll double-check tonight... Am I correct in guessing a problem in the drive circuitry could cause this since the interlacing means you need a pair of sweep sawtooths now instead of a single identical one that repeats?

Sounds like a bear to locate and troubleshoot though as it would involve probing around with a scope...

Well, this thread just got a whole lot more interesting. Now that you mention it, there _does_ seem to be a lack of scan line density that doesn't seem to match up to the 525 standard.

I'll double-check tonight... Am I correct in guessing a problem in the drive circuitry could cause this since the interlacing means you need a pair of sweep sawtooths now instead of a single identical one that repeats?

Sounds like a bear to locate and troubleshoot though as it would involve probing around with a scope...

Interlacing happens automatically with a single continuous horizontal sawtooth and single continuous vertical sawtooth because the ratio of 15750 to 60 is 262.5. Line pairing occurs when some horizontal energy gets into the vertical sync and causes every field to start on a whole line instead of alternating between a half line and a whole line.

You can check for this problem by rotating the vertical hold control very slowly and seeing the line pairing come and go.

Fixing this problem (other than tweaking the vertical hold carefully) can be nearly impossible if it is due to a particular chassis design causing horizontal sweep currents getting into the vertical sync. It can be affected by wire dress, ground points, etc.

In the 1950s, Motorola tried to tout their sets based on quality of interlace:
http://www.ebay.com/itm/Motorola-TV-Standout-Picture-Miracle-Interlace-1953-Vintage-Advertisement-/290962206191#ht_637wt_1223

In the 1950s, Motorola tried to tout their sets based on quality of interlace:
http://www.ebay.com/itm/Motorola-TV-Standout-Picture-Miracle-Interlace-1953-Vintage-Advertisement-/290962206191#ht_637wt_1223

Interesting comments/ad. I'll have a go at this tonight and see whether there is indeed a problem.

I vaguely remembered reading about 'spot wobble' many, many years ago. Dug up a bit of skinny on it...

http://blog.modernmechanix.com/spot-wobble-unlines-tv-picture/

http://www.thevalvepage.com/teletech/wobble/wobble.htm

Seems it never caught on though.

Good news - I tried fiddling with the controls and I can indeed get a uniform field when adjusted in a certain way. There's no actual "overlap" of the interlaced fields, but rather the second one gets close to the first line and moves around a bit when adjusting the hold, so the unevenness is what makes it seem as though there is a piece of the picture missing.

I suspect the set could also do with an alignment as I find myself at the extreme ends of the fine tuning to get a crisp picture... but I'll need to do a bit more work first - there are some IF resistors I didn't replace because it "works" and I don't have replacements on hand - so I'll be ordering some up in the near future and getting those fellas before I attempt any sort of set adjustment.

I'm fascinated by the whole "spot wobble" solution to the scan line problem though - certainly an interesting approach to an eyesore-- especially the homebrew solution which involves making your own coils and adding a tube to oscillate the beam. You'd wonder if there was a more elegant solution to introduce some high frequency, low voltage ripple into the vertical deflection drive instead... somehow adding on your own coils doesn't exactly seem optimal or too "practical" of a solution since positioning would also depend on the set and electron gun design.