Video inputs Tube TV

[ARC Tech-109]

Thanks for the heads-up on those. Analog Devices is not exactly sample friendly as I've found out but I've had good luck finding what I need on the 'bay.

Just an update; I added the LM1881 sync separator to the Magnavox with mixed results. Vertical is rock solid and it takes almost 40 degrees of control rotation to break the lock but the horizontal is being rather finicky with respect to the signal phasing. If I flip the phase the horizontal locks but the color is off by 180 and is very unstable, running in-phase the color is good but horizontal is skittish. A work in progress... little progress.

[kf4rca]

Another chip that shows promise is the LM359. It comes in a dip package, is affordable and plentiful. It operates off a 12V single supply (requires a-c coupling) , bandwidth is 400 MHz, and has an 18 ma. output. 18ma. output is important since it takes 13.33ma to produce 1Volt into a 75 ohm load. It does not have AGC however but there are TWO op amps in the chip.

[Alex KL-1]

BTW... the original sync separator is poor with the video input amplitude?
Or is difficult to use/unstable with mod?
I had some issues with vertical when injecting video input in some old TV's, due to some issues. Some sets needed some different layout/some changing for sync separator. But the one most perfect one resulted to be a tube one, since sync separator is after the video output. Textbook perfect sync separating (for a single stage), with a 6CS6.
For my delta gun one, from the TBA IC's era, I used the LM1881, due to horrible verical behaviour of the original sync separator.

[ARC Tech-109]

More progress on the video inputs today and the sync separator has pulled everything together, I had to do an NPN inverter on the composite sync line from the LM1881 and feed that to the horiz AFC while feeding the vertical sync to the vert section at R515 thru an 0.1uf cap. It's enough to trigger the MV and has a 2/3 swing on the vert hold control before it breaks lock.

The actual video input circuit is Phil Nelson design but with 500 ohm pots on the 12AT7 cathodes and 0.047uf cap between the stages, it then feeds the 1st video amp of the Magnavox which now runs a 6GH8. The 500 ohm pots help get the levels spot on without overloading, not every set will need the high drive level and these will eventually be replaced with fixed values once I have them dialed in.

I'm still dealing with the banding and DC restoration issues and not exactly sure how to solve this. It's very similar to when the RF AGC is too high.

[old_tv_nut]

I've been following this with curiosity, because I don't understand why an added sync separator is required if a video jeep circuit is properly designed to provide the peak-to-peak and DC levels required by a TV chassis. "Banding" (whatever you mean by that) and appearing like the the AGC is too high are signs that the basic input circuit isn't right.

[ARC Tech-109]

Being a compost input it doesn't need the AGC to control the overall gain of the first RF amp. The issue here is the Magnavox T940 was never designed with the external composite video input in mind and everything was taylored to work together this one reason for the external sync separator as it's not so finicky with the variations of the levels and line voltage which would normally be countered by the AGC.

[kf4rca]

I agree oldtvnut. I have never had to add a sync separator to any circuit I built. Composite video has all the timing you need.
I'd like to see his schematic for all this.
Two other chips, that may be hard to find are the CA3450 (by RCA) and the HA5004 (by Harris). They have high output drive but both require a bipolar power supply.

[old_tv_nut]

Originally Posted by ARC Tech-109

Being a compost input it doesn't need the AGC to control the overall gain of the first RF amp. The issue here is the Magnavox T940 was never designed with the external composite video input in mind and everything was taylored to work together this one reason for the external sync separator as it's not so finicky with the variations of the levels and line voltage which would normally be countered by the AGC.

Thanks. I see this makes sense as the jeep circuit DC restored output is probably not following B+ variations the same as the set's video detector. However, I would be curious to put the whole thing on a variac and investigaate what the variations are, to find if it's possible to make the jeep circuit work by itself.

[ARC Tech-109]

LM1881 with an NPN based inverter feeding the horiz AFC. I'm not sure if something was missed in the string of posts but the circuit is only to work in place of the original sync separator not to generate sync. The one issue I'm dealing with is the capacative coupling of the "Phil Nelson design" video input interface amplifier posted on page one of the thread and the lack of both DC restoration and linearity needed for the Magnavox separator to work correctly. The LM1881 is fed right off the 12AT7 input amp and breaks down the vertical, horizontal and burst/back porch to their own outputs, there is a timing resistor/capacitor that is used to set the pulse widths and this has proven to be an asset with unstable sources like VHS tape that doesn't pass thru my TBC, it also stabilizes the vertical during DVD playback so it's not affected by the Macrovision garbage in the blanking bar.
I did post the schematics back a few, they're links to the outside host imgur.com as videokarma will not allow me to post anything large enough for the needed details, I will post more progress pics and screen shots once I find my USB cable.

[ARC Tech-109]

Okay time for a much delayed update on the T940 video input progress.

I won't say the set is complete by any means but it's good enough to watch. I completed the LM1881 sync separator board and have successfully interfaced it to the Magnavox. Original vertical feed resistor (R-515) was removed and this is where I feed the separator via 0.047uf 400V capacitor reduced from the 0.1Uf, the multivibrator triggers off the edge anyway so it's not super critical but too large a cap caused a retrace and foldover at the top, vertical output tube is a 6GF7 twin triode. This arrangement gets me almost 40 degrees of swing on the control before jitter and will not fully unlock going low, it also maintains a full lock with a macrovision signal.

Horizontal uses the composite sync out of the LM1881 via NPN inverter 2SC945 and feeds the original 6GH8 sync separator via emitter thru a 0.1Uf cap. The LM1881 follows the recommended layout from Texas Instruments and uses a 1-meg pot to set the pulse widths. Horizontal stability is very good and it's tolerant of timebase errors of home grade VHS sources. Everything I have is slaved to a master Videotek STG-6000 reference, playback formats are DigiBeta, BetaSP, DVCPRO Type-C and Newtek TC-550 Tricaster so timebase errors are not an issue.

Once I had the sync separator in place a few new issues came up with the chroma, color killer and 3.58 osc itself and I ended up replacing all of the 1/2 watt carbon comps on both the video and sweep boards as most were well outside of their 10% and 20% tolerance. I also cleared off the video IF board and it was repurposed for the 12AT7 input amplifier that was described earlier. Response is far more linear without the ringing I had earlier and it's passing better than 5.5 MHz of luma, still having issues with the chroma washing out on the highlights and I'm not sure what is responsible for this. I spent over 6 hours working with the bandpass and other coils for the chroma circuit but I'm still at a loss. At first it appeared to be a saturation issue with the input amp so I added 500 ohm pots to the cathodes of the 12AT7 to soften the gain but the less saturated areas seem to wash out of any color regardless of the luma. Plenty of room with the color level control but even at full drive the less saturated areas wash out to B&W. Burst amp and the related are all keyed correctly as the horizontal is timed ahead of the input amp and the pulse width is allowing the full burst thru without opening the bandpass amp late.

Other issues. Convergence has been a pain with what appears to be a harmonic of the horizontal leaking into the blue, on setup the blue has a distinct wave over the red & green which are flat and I'm at a loss for where it's coming from. Everything will converge with the controls but that wave or warp leaks into the blue horizontal line. Diodes all check good and everything responds well when adjusted.

Overall it looks good enough to watch without complaint and I've buttoned the back on for the time being, I have countless hours into replacing resistors and scrubbing off the rosin flux from the board in between the recent tornadoes we had, over a dozen trees damaged or completely uprooted plus a pine that broke the trunk and ended up upside down in the V of a nearby mature oak. It took over a gallon of gas and burned up two chains on a Stihl 024 chainsaw to slice this one up. No injuries or property damage but that pine was a priority removal.

So here are the links to the images for your entertainment, I've kept the resolution high as I don't feel like spending hours in GIMP to squeeze them down. They're large and detailed.


LM1881 sync board https://imgur.com/QY3eEc6

Repurposed Video IF board https://imgur.com/QXyJRES

Image in operation https://imgur.com/NWea574

Front Perspective https://imgur.com/PuzCXkv

I have numerous other images of the set & progress posted on imgur.com, search for images by brhatweed to see them all... I'm being lazy tonight. The link: https://imgur.com/gallery/magnavox-t...onsole-shuTJoT

[kf4rca]

That's a pretty set. Hope you can keep sweep tubes in it. I'd put a couple of fans on the back cover to keep it running cool and prolong set life.

[ARC Tech-109]

Thanks. I have a small EDM-Pabst whisper fan running on the bottom pushing a draft up thru the chassis and the other holes are covered with cardboard, it's strong enough to feel coming out of the cabinet back and the internal temp is down a good 20F according to my Fluke type-k thermometer.

I do have a good supply of 6LQ6 sweepers plus a number of the 6LF6 and 6LB6 that would only require a socket and screen resistor change if needed. The bigger problem I've been having is with gassy damper tubes, two have failed recently with one taking the primary fuse out. Most of my used tube stock goes back to the 1980's but the most recent 6DW4 was a NOS pulled from the carton and dated 1977... new when I was in the 3rd grade. We have a big surplus store called AxMan in St. Paul that has a good stock of NOS TV types for a couple bucks each so I'm going to grab a handful of those and the 6GH8s. They even have NOS 6CW4 and 6DS4 Nuvisters if anyone is interested, I'm not looking to make a buck just cover the purchase and postage and I'm open to trades.

New thermal issue has cropped up, takes a few minutes for the horizontal to drop down in frequency enough for HV and sweep, I suspect one of the "new" resistors is bad but it's not running the horiz output in full conduction from a lack of drive. Other issue is the color killer seems to also have a thermal problem killing the chroma until things warm then it returns to full saturation, guess I need to invest in a can of R-152 known as keyboard duster.

It's been quite a journey and honestly I didn't think it would still be going to this day, a one owner set from 1970 still in the family? Might just outlive me.

[kf4rca]

What year is that set? How much was it when new?
My mother had a Sylvania entertainment console years ago. It was hybrid- solid state RF, IF, and video. Tube type sweep. It had a phenomenal sounding AM radio! I think she liked it more for the furniture appeal. I remember changing some caps in it.
I made a board for the LM359 but the chip is on a slow boat from China. I made some changes to the original circuit. I changed the gain from10x to 2.5x and installed a pot for fine adjustment. Also I installed a trimmer cap to adjust the response somewhat. Its a 12V chip but I'm hoping it will work on 9V.
Another chip that might be useful is the LH0002. Its really not an op-amp but a buffer. It comes in a 10 pin dip and operates off a bipolar supply. There is also a TO-5 can version available.

[ARC Tech-109]

The magnavox arrived about 9 months after I did in 1970 and I have no idea how much it cost. I was told it was on a "clearance sale" at a dealer on Lake & Cedar. It was quite reliable through the years with only the sweep tubes being replaced, might have been a 6GH8 or two and has countless hours of Saturday morning cartoons, football games and by the 1980's ATARI-800 time until it was replaced with a Zenith S-3 sometime around 1985 and I ended up with what I though was an albatross in my room.
I blew out all the output transistors in the original amplifier rockin' it hard and it sat until 2005 when I pulled it out of the basement and began to sort things out, the real estate crash a few years later got it stuffed into a large refrigerator box and stored until this past July when I decided it was time to get it going again. Back in 2005 I replaced the old amp with a Heathkit AA-100 that didn't have a cabinet, the plan with that was to harvest the iron and 7591's but instead it got a re-cap and bias pot mod and grafted into the set, gotta tell ya those alinco magnet 15's on the tube amp really have some sound to them. The tuner was never that great on sensitivity but did sound good, the AM section was never worth a dime and the EM84 tuning eye is weak. The stereo section is tube hybrid, I don't know if this was a leftover at the factory because it's not the one depicted in the service manual set Sams 1109-folder-1 and I can't find the original tag or chassis stamp.

So right now it's working but there are still a number of challenges to sort out, one is the color killer wants to cut the bandpass amp off intermittently, the horizontal osc starts off very high then comes down with heat and the color is still "shallow" meaning it goes away on areas of the scene with low chroma saturation regardless of luma level. At some point in the near future I will photobomb my imgur page with the internals and various other pics of the set, I did do some damage to the finish of one of the sliding doors that turned it a milky white, I know nothing of wood finishes and will post this damage later on.

[Electronic M]

Mohawk makes a blush eliminator (lacquer thinner and lacquer retarder) in an aerosol can that is intended to fix white blush in recently dried finishes that were sprayed in high humidity (which causes the white blush). I have had luck removing white blush in vintage finishes, hiding scratches, and rejuvenating tired original finishes by orienting the cabinet such that the surface to be treated faces up and is level (otherwise the toner lacquer runs and you get dark runs) and spraying enough to temporarily make that finish liquid. It'll slowly re-dry and be fine.