Sony FV310 - horrible dynamic convergence

[bobrocks95]

I've had this set for a long time, not wanting to do detailed non-service menu adjustments for fear of messing the convergence up worse or hurting myself, and mostly hoping another set would show up near me. But I haven't seen another FV310 locally in over a decade, and at this point even common Sony sets have people asking several hundred dollars for them, and they're pretty much gone from Craigslist.

Overall pictures- here and here

My static/center of the screen convergence looks good to my eyes (slight red misconvergence in the picture isn't really visible in person).

The entire left-hand side dynamic convergence is pretty terrible though, especially in the top-left corner. Vertically, red pulls right, which is the #1 most noticeable problem in games since you get a lot of white UI elements in the top-left corner that end up looking really bad. Blue pulls a little left vertically though only at the extreme edge. Horizontally blue sags downward on pretty much the entire left side. I didn't take close-ups but on the right side blue pulls vertically to the right.



The service manual kind of implies that permalloy strips are only for the extreme corners. I bought some and tried messing with them for a bit, but didn't find any position that was a large help. I didn't spend much time with them though.

If I'm reading right I don't need to do anything with the magnetic rings on the yoke, those are just for static convergence? I think I need to mess with the yoke physically based on the service manual but I'm apprehensive to do that.

Any specific advice? I will say in all the lurking on Fudoh's CRT thread I see a lot of people complaining about convergence and looking at their nearly perfect pictures and wondering how my set is so far off. It is a 32" unit though, so maybe once you hit this size you just really can't correct for these sorts of problems?

EDIT: Rest of the blurb on TLH plate adjustment on the following page here, the TLV variable resistor may be useful and is certainly easier to adjust.

[tongshadow]

Check your purity, if it's not on point then, yes, you'll need to readjust the yoke. Also inspect the yoke externally, look for loose glued magnets, they can cause all sorts of random geometry/convergence issues and tend to fall off due to heat and dust.

[bobrocks95]

Check your purity, if it's not on point then, yes, you'll need to readjust the yoke. Also inspect the yoke externally, look for loose glued magnets, they can cause all sorts of random geometry/convergence issues and tend to fall off due to heat and dust.

Purity is perfect if I understand purity correctly (a solid red/green/blue/white screen would have splotches of incorrect color if purity was off). I can check the epoxy/silicone on the ring magnets and purity magnets, though they looked good to the eye. The TLH plate is supposed to be movable though I'm not sure on which axis.

[matt]

Don't worry, you can fix that! Your photos are pretty standard for large Wegas these days. I've seen a lot worse and haven't found one yet that I couldn't make nice.

First thing you should do is realign the yoke - it's sagged downward over time which is what's causing the horizontal bowing. It needs to be tilted up slightly and secured in place so make the geometry square. This will also affect dynamic convergence, probably for the better, so it's best to do this first. I've found that by now, most the tape that holds the rubber wedges in place has dissolved into a useless goo and needs to be replaced.

Once the yoke has been straightened out, your next steps should be to work with the H Stat pot on the neck board and the dynamic convergence pots on top of the yoke. You can get a lot of mileage out of these, and sometimes it's enough. It's a good idea to mark the pots' original position with a marker to simplify things.

Finally, you can start adding convergence strips to help with the corners. If the TV has strips already, I usually like to remove them and re-seat them - sometimes this can fix things without the need for additional strips.

You probably won't need to adjust the neck rings.

[bobrocks95]

Don't worry, you can fix that! Your photos are pretty standard for large Wegas these days. I've seen a lot worse and haven't found one yet that I couldn't make nice.

First thing you should do is realign the yoke - it's sagged downward over time which is what's causing the horizontal bowing. It needs to be tilted up slightly and secured in place so make the geometry square. This will also affect dynamic convergence, probably for the better, so it's best to do this first. I've found that by now, most the tape that holds the rubber wedges in place has dissolved into a useless goo and needs to be replaced.

Once the yoke has been straightened out, your next steps should be to work with the H Stat pot on the neck board and the dynamic convergence pots on top of the yoke. You can get a lot of mileage out of these, and sometimes it's enough. It's a good idea to mark the pots' original position with a marker to simplify things.

Finally, you can start adding convergence strips to help with the corners. If the TV has strips already, I usually like to remove them and re-seat them - sometimes this can fix things without the need for additional strips.

You probably won't need to adjust the neck rings.

Thanks, I watched some more yoke adjustment videos and felt a little more confident about adjusting it, so I loosened the screw and tried out some small movements. I can correct my slight rotation issue and pull up enough to eliminate the vertical bowing which looks good, but the yoke seems really stiff on there and definitely isn't moving freely, any slight adjustments take more force than it feels like they should and kind of snap right back to where it was. The service manual only mentions one yoke screw and says to just slightly loosen it, and I actually ended up turning it little by little and testing each time and I'm pretty sure I loosened it as much as it could possibly go. So I'm not sure what's got it stuck on there or where I need to apply the force to loosen it correctly? Unless there's like a complete second screw I'm missing.

Whole assembly - yoke screw

I'm currently grabbing at the light gray solid piece on the right where the wedges connect with the large circular wire clipped around it, is this the best approach and a safe point of contact?

[matt]

You probably don't need to loosen the screw. Typically, all you have to do is tilt it up slightly, and there should be enough play on the yoke already. Keeping the screw tight will prevent unwanted rotation or backwards movement.

The screw itself doesn't keep the yoke in place, that's what the rubber wedges are for. If you're having a hard time tilting the yoke up, its probably because the bottom wedge is in the way. The yoke won't stay in place until the wedges are properly set.

Basically, you just have to remove the bottom wedge, lift the yoke until the horizontal lines are straight, and reset the top wedge to hold it in place. Then put back the bottom wedge and reseat the side wedges if needed. The existing tape has most likely degraded and will have to be replaced in the process.

The plastic rim on the edge of he yoke is the safest place to hold it. Don't touch the pins on the neck board or on top of the yoke, they are live and can give a nasty shock.

[Ryeno]

I wrote a guide on how to adjust the convergence. You should start by reading it.

https://www.retrogameboards.com/t/honey ... log/852/61

Based on your pictures. I would start by adjusting XCV. If you're lucky, you'll have a wheel on the top of the yoke that you can spin after you remove the paint. If you're unlucky, you must buy the PLASTIC tv radio alignment trimmer tools. https://www.reddit.com/r/crtgaming/comm ... ers_do_on/

Second I would remove the top and bottom yoke wedges, lossen the yoke bolt and tilt the yoke up or down to adjust YCH. Clean the old adhesive from the wedges using adhessive removal and use 3M mounting tape to reattach them after you adjust the angle. You may want to adjust the 2 rings on the VM neckboard afterwards to improve the geometry.

Third, play around with the H-Stat on the neckboard

These 3 fixes should fix most of your convergence issues. I always do a full yoke and ring calibration on my consumer sets but that requires a lot of skill and a jeweler's loupe. You may also want to use TLV1 to improve H-TRP although it looks fairly good on your set and then TLV2 to set the TLV.

[bobrocks95]

Greatly improved the vertical bowing with a slight yoke adjustment- hopefully they hold well with the 3M 2-sided tape I used, it was the thick red-covered roll stuff. Wasn't as hard as I thought it would be, though I only really did one thing! I think I can tweak the top-right geometry in the service menu a bit more as well.

I have H-TRP, YCH, and TLV available on pots on top of my yoke. H-TRP is also available in the service menu so I didn't touch that, but for YCH and TLV I marked with sharpie where the pots originally were and did my best to adjust them with the Green gun off in the service menu. They kind of have a balance to them though, what looks perfect on one side of the screen can look really bad on the opposite, so you just sort of find a middle ground it seems like.

I am not sure where H-Stat is since the service menu confused me a bit. It has a pot labeled V-Stat, then shows a diagram showing horizontal static convergence, then exclusively talks about moving the magnetic rings for vertical and horizontal static convergence - section here.

Also no wheel for XCV. Is this the little guy the service manual is calling the XCV "core"? And I need a plastic hex wrench or it'll strip too easily?

[Ryeno]

Links doesn't work.

You aren't supposed to adjust geometry with yoke position. The yoke position is for adjusting purity and convergence.

>it was the thick red-covered roll stuff.

That's fine.

>H-TRP is also available in the service menu so I didn't touch that

Service Menu H-TRP doesn't do much, I would leave it at the middle setting 7 IIRC and adjust H-TRP using TLV1. Then you fix TLV using TLV2.

>but for YCH and TLV I marked with sharpie where the pots originally were and did my best to adjust them with the Green gun off in the service menu. They kind of have a balance to them though, what looks perfect on one side of the screen can look really bad on the opposite, so you just sort of find a middle ground it seems like.

You adjust YCH using the pot on the yoke and by tilting the yoke up and down. If you max out the pot and vertical convergence isn't equal in the upper from center and lower from center sections of the screen then you must tilt the yoke.

TLV adjusts the red/blue convergence horizontal convergence in the upper from center and lower from center sections.

>They kind of have a balance to them though, what looks perfect on one side of the screen can look really bad on the opposite, so you just sort of find a middle ground it seems like

Correct but yokes also have imperfections so you must calibrate around that. The rule of thumb is you want to get good convergence in the center and up, down, left, and right from center areas. Don't try to improve convergence in the corner sections. They can only be modified with strips, sometimes.

>I am not sure where H-Stat is since the service menu confused me a bit. It has a pot labeled V-Stat, then shows a diagram showing horizontal static convergence, then exclusively talks about moving the magnetic rings for vertical and horizontal static convergence - section here.

I think the service manual has a typo. H-Stat is on the neckboard. V-Stat is notched convergence rings.

>XCV "core"? And I need a plastic hex wrench or it'll strip too easily?

Correct.

[matt]

Links doesn't work.

You aren't supposed to adjust geometry with yoke position. The yoke position is for adjusting purity and convergence.

Beg to differ on this. It is indeed listed in Sony's literature as one of the methods of correcting horizontal bowing. For a very exhaustive document on the subject, this is a good read:

https://elektrotanya.com/sony_crt-01_cr ... nload.html

It's somewhat relevant now, as many Wegas have had the rubber wedges come loose over the years and the yoke has sagged downwards. This causes bad convergence and bowing, and readjusting the yoke will improve both of these problems at once. It's always my first step when working on a misconverged Wega.

[Ryeno]

Links doesn't work.

You aren't supposed to adjust geometry with yoke position. The yoke position is for adjusting purity and convergence.

Beg to differ on this. It is indeed listed in Sony's literature as one of the methods of correcting horizontal bowing. For a very exhaustive document on the subject, this is a good read:

https://elektrotanya.com/sony_crt-01_cr ... nload.html

It's somewhat relevant now, as many Wegas have had the rubber wedges come loose over the years and the yoke has sagged downwards. This causes bad convergence and bowing, and readjusting the yoke will improve both of these problems at once. It's always my first step when working on a misconverged Wega.

I've read that manual. If the yoke sags downward then it will affect the geometry but that's irrelevant because, when calibrating, you always set yolk tilt based on dynamic convergence. Generally good dynamic convergence correlates with good geometry but not necessary.

[bobrocks95]

Links corrected, Google Photos is very picky about how you generate your links.

Yoke adjustment made an immediate improvement to the vertical bowing, something I thought was inherently unfixable on the flat screen Wegas, and with no apparent downsides. The top and bottom yoke wedges were very loose when I removed them and the tilt correction felt very natural when doing it, to the point where the yoke just seemed outright incorrectly seated. Compared to every other CRT adjustment I've done so far (service menu, convergence strips, pot adjustment) this one felt the most cut and dry "oh this is where this needs to be" of anything. So I'm grateful for the suggestion, and now hopefully you can see the geometry grid picture.

>H-TRP is also available in the service menu so I didn't touch that

Service Menu H-TRP doesn't do much, I would leave it at the middle setting 7 IIRC and adjust H-TRP using TLV1. Then you fix TLV using TLV2.

I'll see how broad H-TRP allows in the service menu and then look. My understanding was that it's broadly speaking a geometric tweak first, that also happens to affect TLV convergence second.

>but for YCH and TLV I marked with sharpie where the pots originally were and did my best to adjust them with the Green gun off in the service menu. They kind of have a balance to them though, what looks perfect on one side of the screen can look really bad on the opposite, so you just sort of find a middle ground it seems like.

You adjust YCH using the pot on the yoke and by tilting the yoke up and down. If you max out the pot and vertical convergence isn't equal in the upper from center and lower from center sections of the screen then you must tilt the yoke.

TLV adjusts the red/blue convergence horizontal convergence in the upper from center and lower from center sections.

Both YCH and TLV were way off at both ends of the pot, the correct setting is definitely close to the middle, so it sounds like the yoke is about where it needs to be for them and the pots can handle the rest? I may have been judging by the wrong portions of the screen, I'll look again. TLV as I recall was either perfect in the lower center or upper center but not both (and horizontal convergence is easy to see on an aperture grille with your face jammed right up to the phosphors). Your advice on focusing on the cardinals should help though.

>I am not sure where H-Stat is since the service menu confused me a bit. It has a pot labeled V-Stat, then shows a diagram showing horizontal static convergence, then exclusively talks about moving the magnetic rings for vertical and horizontal static convergence - section here.

I think the service manual has a typo. H-Stat is on the neckboard. V-Stat is notched convergence rings.

That makes more sense. I'll mark H-Stat's position and see if I can improve it.

I've read that manual. If the yoke sags downward then it will affect the geometry but that's irrelevant because, when calibrating, you always set yolk tilt based on dynamic convergence. Generally good dynamic convergence correlates with good geometry but not necessary.

If I have a separate YCH adjustment available on the pot on the top of the yoke, why not drastically improve geometry by tilting the yoke, and then fine tune YCH with the pot? I saw this added after typing the rest of this reply, but same thing as above- if the sweet spot is roughly in the middle of the pot then the yoke is about where it needs to be I'd think?

And here's the picture of what I think is XCV again

[Ryeno]

>TLV as I recall was either perfect in the lower center or upper center but not both

That's correct. This is because 1) their are imperfections in your yoke and/or 2) because the center rings are slightly off and/or 3) you must tilt the yoke left or right.

>If I have a separate YCH adjustment available on the pot on the top of the yoke, why not drastically improve geometry by tilting the yoke, and then fine tune YCH with the pot? I saw this added after typing the rest of this reply, but same thing as above- if the sweet spot is roughly in the middle of the pot then the yoke is about where it needs to be I'd think?


The sweet spot isn't necessary the middle. It depends on the set.

https://www.arcaderepairtips.com/files/ ... v1.4.0.pdf

Page 30. When you tilt the yoke up and down, it mostly adjusts YCH but it also affects XCV. So the tilt will be determined by the XCV and YCH. That is what you're trying to optimize through tilting the yoke and the geometry will be what it is.

>And here's the picture of what I think is XCV again

I forgot you have the 32" yoke. It's slightly different from the 27" and I keep thinking about the 27" yokes. Take a picture of the top and bottom of the yoke. I think that is the XCV tho.

Also set AXNT to 0 in the service menu and use pro mode with color temp warm.
Reduce the vertical overscan and if experiment with HOSC to improve geometry near the top and bottom edges.

This is my set after I calibrated it.

https://imgur.com/vxSHj8u

[bobrocks95]

Top of yoke - bottom bad angle 1 - bottom bad angle 2 - service manual vaguely pointing to bottom of yoke
It's the metallic looking box closer to the yoke with 2 screw posts coming down and what looks like a wrapped solenoid inside it. Fairly certain this is XCV from what you've been saying so far.

AXNT has been set to 0 for about 8 years to get rid of red push, though I prefer cooler colors unless it's a hindrance while calibrating.

PDF link is broken but my point on YCH and XCV isn't that the middle of the pot is the sweet spot. It's that I'm not maxing out the pot in either direction. If tilting the yoke vertically affects vertical bowing, YCH, and XCV, I don't understand why I shouldn't adjust it for the vertical bowing, which is the one thing I have no other way of adjusting.

If the yoke was so far off that the pot/core couldn't correct YCH or XCV fully, then of course I'd have to make the compromise. But unless the yoke is adjusting YCH and XCV in some way different than the pots are, I don't see why I'd use it for adjustment when the pots are available? Am I making sense now?

EDIT: Since I've now noticed my vertical static convergence wasn't that great to begin with, I've taken the paint off the 4-pole V.Stat magnets. Spreading them apart appears to affect vertical convergence, but rotating them both around the neck affects horizontal as well? Is this effectively the same thing as H.Stat and I can just tweak H.Stat again afterwards?

[Ryeno]

I have this set of these for adjusting the XCV

htt1ps://www.ebay.com/itm/254880368168?chn=ps&mkevt=1&mkcid=28

PDF link is broken but my point on YCH and XCV isn't that the middle of the pot is the sweet spot. It's that I'm not maxing out the pot in either direction. If tilting the yoke vertically affects vertical bowing, YCH, and XCV, I don't understand why I shouldn't adjust it for the vertical bowing, which is the one thing I have no other way of adjusting.


If the yoke was so far off that the pot/core couldn't correct YCH or XCV fully, then of course I'd have to make the compromise. But unless the yoke is adjusting YCH and XCV in some way different than the pots are, I don't see why I'd use it for adjustment when the pots are available? Am I making sense now?

No because if you try to move the yoke for geometry you will end up getting outside the YCH/XCV range. You haven't adjusted XCV yet, which was also the number 1 biggest problem on your set, so you haven't realized this yet. I've calibrated over a dozen sets and I only think about dynamic convergence and purity when I move the yoke. If you spend enough time calibrating the set, you'll see what I mean. FYI, these are some of the hardest TVs to calibrate because the sweat spot is so small and there are usually defects in the yoke. You have to go in with a plan and you have to know what you're doing or else you'll spin your tires doing nothing or make things worse.

EDIT: Since I've now noticed my vertical static convergence wasn't that great to begin with, I've taken the paint off the 4-pole V.Stat magnets. Spreading them apart appears to affect vertical convergence, but rotating them both around the neck affects horizontal as well? Is this effectively the same thing as H.Stat and I can just tweak H.Stat again afterwards?

You can't rotate both rings around the neck on a Sony Trinitron. You can only rotate the notched green/magneta rings around the neck freely. The ribbed red/blue rings must be equally spaced at or around 12 o clock (e.g. 11 and 1, 10 and 2) just like the purity rings. The ribbed rings are only used to adjust red/blue center vertical alignment. Red/blue horizontal alignment is controlled through the H-stat.

[bobrocks95]

Delrin hex tools are on the way so we'll see if I'm outside of the XCV adjustment range. I am not outside of a good YCH range and you said yourself that vertical tilting of the yoke mostly affected YCH. If I have to tilt the yoke again it's not the end of the world but it's infinitely harder to fine tune vs twisting a pot.

Don't know why you think the V.Stat rings can't be rotated around the neck. The service manual shows both rings rotating clear as day, and once I got the paint off I confirmed that rotation affects red/blue H.Stat just like the pot on the neckboard, and spacing between them affects red/blue V.Stat. From the factory they were definitely not vertical, they were about 8 o'clock and 10 o'clock.

Spoiler

Regardless, so far everything is improved from where I started. Geometry, static convergence, and dynamic convergence are better across most areas. My blue horizontal lines are still sagging on the left since I can't adjust XCV yet, but red pulling right is mostly unnoticeable now. I think with XCV and reseating the convergence strips it will look almost as good as it possibly could.

I have a couple of other questions about this set but I will save them for when I've completed convergence adjustments. I definitely appreciate all advice offered so far.

[Ryeno]

The service manual for the 27FV310 isn't correct and it isn't complete. It doesn't even mention H-STAT. The PVM service manual describes the process better.

https://imgur.com/Fxnp13Q

[bobrocks95]

The V. Stat rings were absolutely not vertically oriented from the factory though. I could put them vertically but it would cause a drastic change in H.Stat, possibly more than the pot could account for. What year is that PVM manual from, the FV310 is a 2003 set.

[Ryeno]

The V. Stat rings were absolutely not vertically oriented from the factory though. I could put them vertically but it would cause a drastic change in H.Stat, possibly more than the pot could account for. What year is that PVM manual from, the FV310 is a 2003 set.

M Series (14M4U, 20M2U, etc). Any Triniton set that uses a separate H-Stat adjustment will be calibrated this way. Non Trinitron TVs don't have an H-Stat so they are calibrated by freely rotating all rings. On Sony TVs only the middle green ring can be freely rotated.

Most sets are properly calibrated from the factory but some are not. The 1st set of rings are purity. The 2nd set of rings adjust green. The 3rd set of rings adjust red/blue. The 3rd set should be used to only set red/blue vertical alignment and should be set equally spaced from 12 o clock as shown in the service manual then H-Stat is used to converge red/blue horizontal convergence.

[tongshadow]

The 1st set of rings are purity. The 2nd set of rings adjust green. The 3rd set of rings adjust red/blue. The 3rd set should be used to only set red/blue vertical alignment and should be set equally spaced from 12 o clock as shown in the service manual then H-Stat is used to converge red/blue horizontal convergence.

Not always, even within the same manufacturer they could be in a different order, the best way to check it to actually move them. Red/Blue convergence is the most important and should be calibrated first, because aligning the green beam is just putting those two together, so if R/B is off, so will the overall convergence.

H-Stat potentiometer does indeed make fine tuning easier though, and Sony recommends keeping the convergence tabs at a certain angle (H. Convergence will be off) to adjust V. Convergence, and then use the H.Stat potentiometer to join the beams.

[bobrocks95]

H-Stat potentiometer does indeed make fine tuning easier though, and Sony recommends keeping the convergence tabs at a certain angle (H. Convergence will be off) to adjust V. Convergence, and then use the H.Stat potentiometer to join the beams.

Both H.Stat and V.Stat are dialed in nearly perfectly right now, do I really need to redo the adjustments just so that the V.Stat tabs are vertically oriented? Again, they were horizontal from the factory.

[tongshadow]

H-Stat potentiometer does indeed make fine tuning easier though, and Sony recommends keeping the convergence tabs at a certain angle (H. Convergence will be off) to adjust V. Convergence, and then use the H.Stat potentiometer to join the beams.

Both H.Stat and V.Stat are dialed in nearly perfectly right now, do I really need to redo the adjustments just so that the V.Stat tabs are vertically oriented? Again, they were horizontal from the factory.

Oh I wasnt talking about your case, but if the middle look perfect then leave the tabs alone.

[bobrocks95]

Can anyone explain how you're supposed to move the TLH plate? Every service manual says to "tilt" it, but in which direction? I can slide it in or out, but I can't really tilt it up/down or towards the front/back. Maybe I need to clear out the epoxy on it but it doesn't seem to want to do anything other than slide. Don't seem to be any videos, pictures, or discussions anywhere on google of someone adjusting it.

EDIT: Found this:

Spoiler

So I'll just have to mess with it more since in and out seems to be correct (in what world is that "tilt"?)

My main problem right now is NOT XCV, It's XBV. My vertical balance between red and blue is good, but blue is sagging down on both the left and right sides of the screen. I found this in this same manual:

Spoiler

So perhaps the orientation of the 4-pole magnets does matter. Unfortunately this manual doesn't explain how the magnets should be adjusted, so I get to guess...

EDIT: For XBV you either get static vertical convergence looking good with the 4 pole magnets, or you get dynamic vertical convergence looking good with the 4 pole magnets. Maybe I should have gone for something a little more balanced but I opted for perfect static over dynamic. We'll see if I get annoyed and pop the set back open, but I'm done tweaking for now it's been hours on this set.