Line pairing

[lnx64]

Having a slight issue.. On my Hitachi, adjusting vertical hold would fix line pairing, but on my Samsung, no matter what I try to do, the scan lines are still paired, and I can't get rid of the line pairing that is going on.

Any ideas on how to fix that?

[Username1]

What exactly do you mean.... Line Pairing ? What is that....?

[wa2ise]

Originally Posted by Username1

Line Pairing ? What is that....?

That's when the interlacing doesn't happen correctly. Ideally the deflection circuits should get the raster lines of one field fall between the other field's. But sometimes the sync separate circuit doesn't do a good enough job of getting the horizontal out of the vertical, or some other circuit (maybe B+ bypassing) lets some horizontal get into the vertical.

[lnx64]

Originally Posted by Username1

What exactly do you mean.... Line Pairing ? What is that....?

This. These are two fields that are bundled closely together, causing no uniformity.

[Username1]

So, If I understand this correctly, the lines are not interlacing...? (like wa2ise said) So if that is it, then the scan lines are on top of one another, you should see horizontal objects loose definition as the tops or bottoms of horizontal items seem to jitter, or bounce up and down, or be unfocused.... ?

Or as you move your finger down the screen and focus your eye on it, you fail to be able to follow scan lines...?

I really don't see anything in those pictures above.... I believe I have seen the phenomenon before, but as some limited interference where the picture seems to jump up and down by a small bit, possibly by the size of a scan line....?

I guess a photo will not show it well..... So does it do this with several signal sources....? I guess you should look at old picture stability circuits and find that part of your tv and look for a failure in that area.....

[lnx64]

They are close to being on top of each other, but not quite.. Like, scanlines are in pairs of 2, if that makes any sense..

Happens on all sources, EXCEPT my Sencore LO-BOY generator.. That one oddly doesn't do it. But a Sencore VP300, it does it.

[lnx64]

Maybe this will make more sense.

[ChrisW6ATV]

Inx64-

Your photographs show what I would call decent but not perfect interlace; really bad interlace (or "line pairing", which was a new term to me as well starting a couple of years ago or so) that I have seen causes the lines to be completely on top of each other with a dark space between them, a true loss of half the vertical resolution. I have seen many sets that can go from bad to medium to excellent interlace repeatedly as I adjust the vertical-hold control within its locked range.

I have not seen any descriptions of "fixes" for these limitations; good versus poor interlace was just something that separated one brand/model from another. If there are truly ways to improve this (other than the fiddling with the vertical-hold control as mentioned), I would like to learn about them, too.

[wa2ise]

Many years ago at some small company I worked for, we rigged up a circuit using one shots and flip flops to act as a sync separator to get the lines to not pair.

[old_coot88]

=ChrisW6ATV;3088246]...If there are truly ways to improve this (other than the fiddling with the vertical-hold control as mentioned), I would like to learn about them, too.

Interlace occurs 'naturally' (or it's supposed to, anyway) due to the even/odd mathematical relationship 'tween the H and V scan frequencies. But as Wa2ise suggested, it can go wonky if there's some kinda crosstalk between H and V. As for a fix, who knows.

[old_tv_nut]

If the pairing is caused by poor v sync filtering, it could be fixed by changing part values, but more often than not it is caused by ground currents from H sweep, which is unfortunately tied up with the physical chassis design and grounding. Special countdown or flipflop circuits can force interlace, but this can actually cause trouble with cheap random interlaced sources.

Sent from my LG-LS970 using Tapatalk

[Kevin Kuehn]

This Zenith patent description is an interesting read on the common problem and the circuit designed to cure it.

http://www.google.com/patents/US3855496

[lnx64]

Now these circuits that are designed to fix and force proper interlacing, what happens if I feed the TV a 240p source like an old game system that had a non interlaced output?

[old_tv_nut]

May or may not work well. Later integrated circuits incorporating internal countdown also contained complex non-interlaced source detection as well as accommodation for vcr head-switching glitches.

Sent from my LG-LS970 using Tapatalk

[ChrisW6ATV]

Originally Posted by wa2ise

Many years ago at some small company I worked for, we rigged up a circuit using one shots and flip flops to act as a sync separator to get the lines to not pair.

Years ago, after I had built a Heathkit 12" B&W TV kit (model GD-104C if I remember right) that had lousy interlace, I found some type of digital sync-separator circuit in a book, made of about a dozen TTL integrated circuits, and I started to build it hoping to hack into the TV to fix it. I never finished the project, but it is good to hear that maybe I was not crazy for trying. (-: