Greetings,

I just got done resurrecting a Toshiba SV-771 with the help of ya'll at this thread: http://videokarma.org/showthread.php?t=274730
It sat for about 5 years without use in our basement and had erratic playback due to a dirty mode switch. The VCR now plays without issue and generates a good picture, except for one defect...

There is always a white dot/streak on the screen erratically dancing back and forth across the screen and always at the same vertical position (see attached pictures). Sometimes it's very weak or almost gone, other times there are several copies of it. I also noticed that the tapes coming out of the unit had a crease about 1/3 down from the top.

Sleuthing found a hard speck of dirt on a tape guide at just the right spot. I super cleaned that guide as well as the whole tape path and even cleaned the heads with chamois. To my dismay, the streak is still present in the video. No matter what tape I play (tried several from various other VCRs), it's always present at the same horizontal line. The rest of the picture is about as perfect as VHS can get, tracking great and with HiFi audio.

One other point: when I first started working on this VCR, the video head was locked and would not rotate. I think the bearing corroded while sitting in our basement. I broke it free by twisting on it since I had nothing to lose when I first started working on it. I'm not sure if this could cause an electrical fault but it spins freely now.

Is this a sign of a damaged head, or something else? I'm not sure how to interpret this fault.

That's not a head.

My guess: Check the cylinder static grounding. See pic. Might need to clean the surfaces to get a good ground. The little carbon grounding button should be centered too. The way it is now when it spins it's being pushed back which can cause the carbon button to wear. I used to put a tiny trace of conductive grease on those but that stuff is long gone.

Phew, though it was a damaged head...those are surely NLA at this point.

I think you may be right about the ground springy thing. I did some more Googling today and found this article on JVC VCRs with the same issue: The description of the issue matches mine almost to a tee. And mine is definitely cock-eyed.

I will straighten out the spring contact on my deck tonight and make sure it's clean. Is a tiny dot of automotive dielectric grease a bad idea?

Is a tiny dot of automotive dielectric grease a bad idea? That might work as long as it's thick enough so it doesn't sling off. Just a tiny bit.

JVC had a problem with that in some of their later models and issued a bulletin and a grounding kit to install inside the drum. I did a bunch of those. Some would be so bad the playback would have many dots all over the TV screen.

Ugh, straightening the ground spring did not fix it. The white streaks (comets) are still present in the same horizontal line, no worse or better than before. I noticed the center knub is not rotating completely true to center, but even when I lifted the spring completely off the center contactor the output didn't change at all.

When I rotate the head by hand, I hear a slight crunchiness inside. I'm wondering if there are similar slip rings or contactors inside of it that are corroded and part of the reason the head was locked and had to be broken free. I think I'm going to have to figure out how to remove this head and look inside.

I too saw that notice about JVCs and their "comets". This looks extremely close to that, except it's confined to only one or two horizontal lines near the upper 1/3 of the screen. The rest of the picture and the HiFi audio are perfect. There is also a tiny 16V 10 uF capactitor on the bottom of the head assembly. I think I saw JVC or some other machines exhibited this problem when that cap goes bad.

Got busy this week so have not taken the heads apart yet. I'm still debating if I still want to now that I understand how VCR drums are assembled like transformers. I don't think there are any other connections inside of it that need cleaned.

I did do some more testing on Sunday, though. I hooked up the S-video output instead of composite and confirmed the comets still show up via that route. I then recorded a tape using a DTV box and then played that tape in my Panasonic DVD Recorder/VCR combo. Played perfectly with no comets or other defects, so this is just a playback issue. I also measure the resistance of the drum-to-chassis path and measured 2-3 ohms.

I have also attached two better images of what the comets look like using my Diamond VC500 capture card (No I don't watch wrestling, that's just what was on when I turned the DTV box on :rolleyes:). They are always within the same couple of horizontal lines on screen. Sometimes tearing shows up instead of the comets.

I have one of these VCR's with a bad head. If you want the thing for parts, just let me know and I will get it packed up. I would just ask that you pay the shipping costs. I have been saving it for years due to it's rarity, but I will never get around to fixing it.

The tear is the give away. Your heads are switching there. Old school it was called the PG pulse (Only worked on commercial VTRs). Their is a sensor that tells the electronics where the heads are. It has physically moved or a cap is failing affecting the timing. In older machines there was an adjustment for that. The goal is to have the heads switch in the vertical interval so the noise and timing skew is not seen (forgot what line). This adjustment is typically done with a scope or a cross hatch monitor.

Thanks for the info! Hmm, I agree with you on the tearing, and am now familiar with that adjustment after fussing with it on my Sanyo VCR4500. I do have a scope I can use to adjust it once I find it.

Do you think that misaligned switching would also cause the white streaks? After watching this video on YouTube https://www.youtube.com/watch?v=nUW9gPqScts, I'm wondering if the bearing is bad in the drum and causing the streaks. The drum was seized when I pulled it out of our basement a few weeks ago and I had to break it free by rotating it. There's a tiny amount of "crunch" in the bearing at one point during a 360-degree turn.

@mgross0: That's a wonderful offer of you, I will PM you about that. I would love to have a spare machine since this is a rare beauty and I may have to take the drum apart on mine to try oiling the bearing.

12volts points out all the important stuff. If your drum motor was seized it is likely the grounding electrode was damaged.

If it is an external graphite pad remove it, clean it with degreaser, then apply a dab of graphite lube to it.

If the grounding is integral to the motor bearings use a syringe to inject graphite into the bearings race.

Check your DOC (dropout compensater). Depending on how hi end your machine is the DOC will inject black (or the previous video line) for the RC duration of the DOC once triggered.

Getting your head switching back into the VI may make any of the other problems livable or moot.

Good Luck
Doogie

Hmm that's what I was figuring. This machine has an external grounding strap on the top of the drum. I already tried cleaning it and verifying continuity from it to the chassis ground (measured 2 ohms), but I can try some graphite lube to see if that helps.

Good idea about the bad capacitors. I found two in the power supply of my Sony SLV-575UC which is about the same vintage as this Toshiba. I have an LCR meter so will look into those as well.

Alrighty, tonight I made some progress in inspecting this unit despite not having a service manual. In my Sony SLV-575UC manual I read how to adjust the switching pulse (same thing as Pulse generator?) against the Line out signal. Assuming all VHS machines follow a pattern, I found Video Out and SW P test points on the Toshiba so scoped them while playing a tape.

I need some help deciphering what I am seeing in the attached oscilloscope captures since I'm still figuring how NTSC video works. I tried three different tapes (1 commercial movie, 2 tapes recorded on other machines) and believe the switching point to be in the correct point compared to the diagram given in the Sony manual. However, the first two "big" pulses going from right to left are truncated compared to the diagram Sony gives. Also, I see a spike in the Video Out signal (yellow) right in line with the Switching Pulse (blue).

I'm not sure what 6.5 +/- 0.5H in the Sony is referring to. What does "H" mean in these measurements?

H refers to the horizontal line period.

Your first two traces show start of switching pulse exactly 6.5H before vertical sync. The third trace shows 7.5 H. This should not be a terrible problem, it just means the switching transient is slightly higher above the bottom of the active picture than specified (by one half a horizontal line).
[EDIT - I think this could vary if the recording machine for the tape was slightly off.]

A noise spike at the switching instant is probably normal.

I'm not following which big pulses you are referring to, and do you mean truncated amplitude or time? Can you post an image with a pointer to what you mean? Are you referring to the last few lines of active video, which vary with picture content?

In any case, this proves that the switching transient is not what you are seeing one third of the way down the picture.

Crumbs, looks like I'm back to taking the upper drum off and seeing if I can do something about the bearing. That or will try graphite lube on the grounding strap up top and see if that helps.

Thank you for the explanation about the H value. Learning more and more as I go here which is great. I reuploaded one of the images showing the truncation in the final two pulses before the end of the signal. Perhaps this is normal because it's real video and not an alignment tape?

I have no idea if that's deliberate or accidental, but it seems non-critical because it's the last two video lines at the bottom of the picture, hidden by overscan even more than the switching transient is.

Comparing to your other plots, it looks like one has clean black video and all the others have some bursts of noise in those lines. Doesn't make a lot of sense since the preceding video lines are full amplitude. Maybe it's an artifact in the original source, don't know.

Looking again, it seems those two lines have the video blanked during recording, maybe deliberate in VHS machines for some reason? The noise bursts are possibly because the tape is not making the best possible contact with the heads near that edge. Since this is likely due to the particular tape and not the machine (note you have one clean tape), I would not try to chase it.

A professional analog video monitor would have a pulse-cross display to show what's going on in/near the blanking intervals.

https://en.wikipedia.org/wiki/Pulse-cross

Thanks for the additional information; I agree with your assessment so will not be touching the switching point or anything related to it at this point.

I tried some graphite lube for key locks last night on the ground strap contactor and it didn't make a lick of difference. The white streaks were still present at the same location on the screen. I see the grounding contactor has a divot worn in it and it's not rotating on center, but if I pick the strap completely off the head while it's spinning it doesn't make any difference either.

What is intriguing me about this defect is the white streaks appear in just about the same exact horizontal line no matter what tape I play. All other references I've seen for white streaks in playback exhibited randomness in their appearance. Admittedly I'm not sure where to go here, but I think I either need to check every capacitor on the board and see if one is bad, or take the upper drum off and see if anything jumps out at me.

There are two magnets mounted on the flywheel of the lower drum. One is a north-pole magnet and the other is a south-pole. As the drum is spinning, those two magnets pass over the Pulse Generator sensor and thus the PG signal is formed which is that square-wave signal you see on your oscilloscope. The PG signal is used to tell the servo circuit which head is in contact with the tape. If the relation/alignment of the magnets vs the position of the video heads on the upper drum is messed up, the switching point will occur at the wrong timing, causing the symptom seen here.

There are two magnets mounted on the flywheel of the lower drum. One is a north-pole magnet and the other is a south-pole. As the drum is spinning, those two magnets pass over the Pulse Generator sensor and thus the PG signal is formed which is that square-wave signal you see on your oscilloscope. The PG signal is used to tell the servo circuit which head is in contact with the tape. If the relation/alignment of the magnets vs the position of the video heads on the upper drum is messed up, the switching point will occur at the wrong timing, causing the symptom seen here.

The scope traces show correct timing near the end of the video field. But just to be sure the pulse is stable for the whole field, It would be good to slow the scope trace rate to view two complete fields and make sure there are no glitches during active video.

OK sounds good, I will work on getting those this weekend. I've been coming home late due to work, but can set it up again to scope it.

Alrighty, as promised please see that attached images. I took one of two fields of video, and then one of about four to give a larger picture. This was taken while playing "Apollo 13", a commercial movie.

Just to get terminology staight, you mean 2 frames or 4 frames of video, not fields.

I see nothing in the 2-frame capture, but see some tall narrow spike in the 2nd, 4th, and 6th field of the video in the 4-frame capture.

I see nothing in the switching waveform of either.

Do you agree?

The odd thing is that the timing seems so precise. Is that true when viewing the picture, that it is always the same exact distance from the top of the frame? This does not seem likely to be caused by an intermittent rotary contact, but I have no idea what in a VCR circuit could behave like that.

I didn't think of something: is your scope capturing the two traces alternately or chopping between the two inputs? If alternate traces, you wouldn't see the same glitch on both traces. Even if chopping, the glitch might show up in one and not the other. Try looking at the switching waveform a few more times just to make sure it's absolutely clean.

Hey thanks for the additional input. I agree with your terminology, you're correct since I am still tying what I see on scopes to NTSC video terminology. Did I capture what you were asking for correctly? I do see the spikes in video you mention.

To answer your other questions, yes the streaks/tearing are always in the exact same line down from the top of the screen. I agree that an intermittent contact is unlikely but I too have no idea what else would fail at one line only. The rest of the picture is perfect: no noise, clear image, good color, and even HiFi tracks every tape with aplomb. At least it sounds like I'm not the only one confuzzled. :lmao:

I am using a Rigol DS1054Z, a digital scope, so I don't believe it would alternate between the inputs like an analog scope. I can take a closer look at the waveforms again to be sure they are clean. I will try to see what the other test points may show me, there are a few near the head output board and a bunch on the video board.

Yes, these are good captures.
Still confuzzled! You could try looking at power supplies, I guess.

That might work as long as it's thick enough so it doesn't sling off. Just a tiny bit.

JVC had a problem with that in some of their later models and issued a bulletin and a grounding kit to install inside the drum. I did a bunch of those. Some would be so bad the playback would have many dots all over the TV screen.
I would try cleaning the ground brush and its drum contact with alcohol and reinstall dry without any grease.

Did that last night, no dice. There is a small divot worn into the carbon brush and the drum part is not rotating on center, but no amount of pushing, pulling, or repositioning this brush makes any difference in the white streaks.

I was thinking about this last night, trying to map what I see on screen to the signal coming out of the VCR. I know VCRs record video as diagonal stripes on the tape, alternating between two heads. The question is, what is recorded within one stripe? Is it a complete set of 262.5 horizontal lines worth of information (aka, a field)? Since the head swipes from bottom to top of the tape, is the bottom of the swipe the beginning of a field? I wondered if it was dirt on the tape path but I'm not seeing any, nor any tape path issues during playback.

One field per swipe.
If it's on the tape, it would be part way between the top edge and the bottom, edge, proportional to the vertical position in the picture.

Tape damage would also show up in both fields, but the examples you posted seem to be in alternate fields.

I think this is too minor to be tape damage as such. Damaged tapes typically produce a much wider noise stripe and really wreck the picture as the wrinkle passes by, and produce a buzzing sound as the heads run over the wrinkle.

I still can't come up with an idea of what's going on.

Grasping at straws: Is there a stationary part of the drum that you can get to and ground?

I too am really coming up short, but appreciate all of the suggestions so far. I did attempt grounding the stationary part of the dump using a test lead but it had no affect. I also tried finding the RF output to see if something shows up there but haven't found that pin yet.

I finally got mgross0's spare SV-771 so now I have a spare machine to swap the Video and RF boards out if and see if that makes a difference. That, or I need to send it to 12voltvids on YouTube and let him take a crack at it. :scratch2:

This issue is common on 21st century VHS machines. I've been able to clear it by skewing the A/C head adjust boss. However, the issue returns after the cassette has been ejected and reinserted.