Drive line Question
 

I am completing the recap of the ctc4B chassis for my Cheltenham.

I was plagued with a strong vertical bar in the center of the screen which I believe is referred to as a "drive line" Adjusting the horizontal drive control to minimum would not make the drive line go away. As it turned out the 6CB5 HOT (which tested very good) was the cause. I had to try 3 NOS tubes before I found a tube that cured the drive line issue. Just by luck one of the NOS tubes made the drive line disappear.

I am not terribly well versed from an electronic theory standpoint. I know that there are many of you who understand electronic theory way better than I do. Can someone explain what causes the drive line, and what the reason was that I had to try several known good NOS tubes to find just the right one to make the drive line go away?

Any enlightenment would be gratefully appreciated.

I am not good at pure engineering theory. Just bench
theory, there are places for both.

To us a drive line was a bit of hoz fold over caused by the
H. osc or H out. Usually on the right IIRC.

You may have had a snivet, IIRC caused by oscillations in the
H out tube. Reason for multiple tubes either the tubes were
low quality or maybe something else is off a little. Look to the
screen grid decoupling cap.
Again IIRC another cure was a magnet around the H out tube.
Very common in the olden days.

73 Zeno:smoke:

Not sure, but checking the H oscillator/driver may be worthwhile in case it is contributing to the touchiness of H out selection.

It may have been Barkhausen interference, I've usually seen it on the left third of the screen, but I had an RCA with what I'm assuming was Barkhausen interference dead center. It was a large bright noisy vertical line. Changing the HOT fixed it.

This sounds like a Barkhausen oscillation in the HOT. I have seen ION traps installed on the HO tube and rotated until the vertical line disappears to cure this problem.

Hope someone else can chime in with more details...

A while back someone wrote a very good explanation. IIRC it has to do with screen grid conduction causing oscillation. When the plate becomes less positive than the screen grid the screen begins to conduct. IIRC at those transition points the tube will go into high frequency oscillation changing the plate current slightly for a fraction of each line.

Had the same problem with the ctc7. It was a burned up horz lin/Eff coil. Did you replace the damper tube?

I have several books and read more than a few HO tube data sheets that talk about it and what causes it, but never seen it in real life on any of my sets. But then I'm a nazi when it comes to horizontal circuit set up and adjustment, so maybe my being anal about it is why it hasn't come up. A few data sheets I've read (for compactrons) state that applying a small positive (~30 volt) potential to the beam forming plates can counteract the phenomena, but obviously this is not possible in tubes having the beam forming plates tied to the cathode. I've heard of the ion trap trick as well, and even encountered a chassis with a seemingly factory installed magnet around the HO tube. Common wisdom these days says go tube rolling till you find one the chassis likes, but not everyone has piles of sweep tubes laying around their house.

Read here about "reaction scanning" describing the phases of the horizontal scan process on a CT-100 (and likely most other tvs with a flyback transformer).

http://www.earlytelevision.org/Deksn...implified.html

I suspect that the "drive line" is visible when the transition between the various scan states is not correct due to leaky caps out of spec resistors, weak damper or HO tube, miss adjusted grid drive, while BO is a self oscilation of some HO tubes, which can indeed be stopped by a magnetic field from an ion trap mounted on the tube. Two different causes of lines, IMHO.

jr

Exactly. Two different things. Both hard to describe. A drive line
as we knew it was a true failure. Often seen on SS sets when the
1.5 ohm base resistor changed value. It has a fold over look just
like vert fold over. You can see backwards elements of the scene
moving about in a apx 1 inch area. Damage WILL occur eventually.

Snivets did not distort the pix geometry wise. They come & go
& are usually narrow ( < 1/2 inch ). May or may not go top to bottom.
Usually have squigglys in them. Move around sometimes. They
also radiate & can bust up a lesser set like a GE, Emerson etc.
One of our by-outs was a wanna be radio shack. They loaded this
poor cat with thousands of tubes. Al were IEC service master, Ratheon
& Lindals. Every HO tube gave snivets........ We used them in
"last time" repairs on junkers to save the customer $$.

73 Zeno:smoke:

Yes, a horizontal "drive line" arises from a different cause and it looks different than a Barkhausen line.

A "drive line" is white and it's caused by excessive drive in the horizontal output tube. Here's an illustration from one of my TV books:

http://antiqueradio.org/art/HorizontalDriveLine.jpg

TV literature often tells you to watch for the "drive line" while adjusting the horizontal drive control. Turn it up until the line appears, then turn it down until the line just disappears.

A Barkhausen line is dark. One of my books describes Barkhausen interference as "a parasitic oscillation in the horizontal output stage, radiating to the antenna input terminals of the receiver." The name alludes to similarities in a Barkhausen-Kurtz ultra-high frequency oscillator.

My Hallicrafters T-67 had a classic case of Barkhausen interference: the squiggly vertical interference line in the left part of this screen, going right past the woman's nose:

http://antiqueradio.org/art/hallit-6708.jpg

I cured it by replacing the 6BG6 horizontal output tube. Swapping a used 6BG6 didn't do the trick, but a new HOT cured it for good. (Footnote: all of those tubes tested equally "good" on my tube tester, but that's irrelevant in these cases, since the tester tells you nothing about oscillation.)

Here's more:

http://antiqueradio.org/art/BarkhausenExplanation.jpg

The "Barkhausen eliminator" mentioned in that book looks like a miniature ion trap magnet.

My DuMont RA-103 has a small permanent magnet mounted in the HV cage near the HOT, presumably to minimize Barkhausen interference.

http://antiqueradio.org/art/DuMontRA-103Cleanup04.jpg

I don't know if the RA-103 was especially susceptible to Barkhausen interference, or if that magnet was just typical Dumont over-engineering. My newer RA-113 has no such magnet, so perhaps it was dropped as a nice-but-not-necessary feature, just like the RA-103's B+ delay circuit that required a relay and extra tube.

Regards,

Phil Nelson
Phil's Old Radios
http://antiqueradio.org/index.html

Great discussion guys. Thanks!

To more enlighten you as to if it's drive line or barkenhausen: The line looks like the "drive line" in the photo above. It gets stronger as you turn up the H drive control, and get dimmer when you turn it down, however it would not go away with the control at minimum.

All caps were replaced so it is probably not that, however I did not check voltages on the HO tube to see if they were in line with what is expected.

In any event a good NOS Sylvania cured the issue, but several "International" brand imports had the same issue as the original defective tube which was also a Sylvania. My biggest question is why the difference between various NOS tubes? That just dosent make much sense. You would logically think there has to be an underlying cause as to why several new tubes would not work properly.

FYI the thing that led me to replacing the HO tubes in the first place was that I noticed the getter flashes looked a bit off color, a tad bit brown around the edges of the flash. So I took a chance in swaped tubes and got lucky with the NOS Sylvania.

BTW, the line was right in the center of the screen left to right and ran all the way top to bottom

Silly thing is you can probably put the 'bad' tube in another set, and the line won't be there. Just one of those things with TVs, gotta roll tubes and see what it likes.

A drive line effect will occur from top to bottom since it is high-level and the same effect on every horizontal trace. Barkhausen is an oscillation that is at RF and gets back into the tuner. It is sensitive to conditions such as bias voltages (and magnetic fields, as shown by the magnet trick), and could come and go between top and bottom.

I think the distance from the left to the drive line can depend on the particular circuit, since a more efficient H scan circuit means the damper will be supplying retrace current for a longer time before the stored energy is used up.

Aside:
I never worked on tube H outputs, only transistor monochrome. I don't recall having to make any special accommodations in new designs to prevent drive lines, but that may have been because Moto already had a driver design with a duty cycle that was OK for different output stages. We did have to optimize the turn-off current for the H drive to the H output transistor. There needed to be a negative pulse at the leading edge of the turnoff to pull charge out of the junction of the (saturated) H out transistor (HOT), but its magnitude and width had to be right so as to turn the HOT off as quickly as possible, but quit before causing reverse breakdown. This meant a different component choice depending on the particular model of transistor and sweep current. If the turn-off wasn't fast enough, the dissipation in the HOT would go up, due to the HOT still drawing some current while the collector pulse voltage was rising. A hotter HOT would switch less efficiently, and a thermal runaway would be possible. If the drive current waveform design was really wrong, the HOT turn off could be so slow that the combination of rising pulse voltage and slowly falling current could exceed the safe area of the HOT and kill it instantly. We did a lot of x-y scope tracing with the HOT collector voltage and current as the x and y axes while adjusting the drive design for fastest switching. You could always tell which scopes were used in sweep design because they were the ones with a hole burned in the phosphor at the lower left corner of the graticule; someone would always forget to turn down the scope beam current when turning off the set.

Thanks Wayne, that's interesting stuff!

So today I re-connected the chassis to the test jig again. First thing I did was measure the HOT current and it is running nice at 175ma. I am happy with that. And with the good HOT installed, I never get the drive line on the screen, regardless of how I adjust the H-drive control.

So then I still was not satisfied with the way the picture was presenting. I was having trouble with my ability to adjust the screen and background colors. And to add to the issue, with normal brightness I couldn't get the screen properly adjusted because it always had a pinkish cast. And when I turned down the red screen, my red color nearly disapeard in the video.

I checked voltages on the screen grids and they were looking within normal range, so I decided to start working my way back from the crt to see if there was a signal problem in the video. Sure enough I was getting very low signal on the red grid of the crt. Tracing back to the terminal board at the rear of the chassis where the crt cable connects up to, I found a bad 1/2 watt resistor in the red grid line. Should have been a 150K and it was over 3 megs. It was one of those poor quality resistors that have the rather coarse, rough looking body, not nice quality like Allen Bradley or Ohmite. I have found these cheesy resistors to be much more problematic than the good quality brands. Seems difficult to understand why RCA would allow the use of any poor quality components in their sets, when for the most part, they used good quality materials.

AFter replacing the bad red grid resistor, a beautiful picture developed with plenty of red. :yippy:

IN the end I now have a very nice looking picture on the crt even without doing much setup work.

AFter the Cheltenham project is completed (after convention) I will post a writeup with photos on my web site. I will post a notice here when the Cheltenham pages are up.