CurtPalme.com Home Theater Forum

cmjohnson · Joined: 03 Apr 2006 Posts: 5180 Location: Buried under G90s

I'm bumping this topic back up to the top because it's relevant to my Barquee Cine 9500LCHR project.

Using the focus yokes from a Cine 9, and ALSO, after modification, using the deflection and convergence yokes from a Cine 9 as well, on a very special and unique set of high res tubes made for the cancelled VDC Marquee HR (High Resolution) projector project, I'm hoping to push the Marquee to a level of sharpness that is at least equal to that presented by a perfectly tuned Cine 9.

The reason for using the Barco deflection and convergence coils is a matter of ROOM on the tube neck. The Barco deflection and convergence coil set is almost an inch shorter than the Marquee deflection and convergence coil set, which interferes with proper positioning of the larger Barco focus coil. By using all Barco coils, then the focus coil will be where it needs to be.

My plan, however, is to make no modifications to the yokes other than to change connectors as needed. Since these magnetics are known to deliver incredible results on 9" Panny tubes, I want to leave them in an unmodified state as much as possible.

Instead, I will try to address any electrical compatibility issues by suitably modifying the drive circuitry instead. Rather than change the inductance of a coil, change the capacitance in the drive circuit instead.

I don't think that the required changes will be huge in any event, as the application dictates the design, and a given focus or deflection coil that works well on one high performance 9" CRT should not be TOO different from a different one made to work in the same kind of CRT in a different projector that nonetheless uses the same principles.

The thing about the focus yokes, though, is that the Cine 9/909 yokes, which are KD-3212s, have extra coils in them. They have a 13 pin connector, one of them being ground, and the other 12 are for H and V focus (2 coils, 4 pins), H and V 4 pole astig, (2 coils, 4 pins), and a pair of 6 pole astig windings as well, using up the last 4 pins.

Since the Marquee doesn't have drive circuitry for two sets of astig windings, I'll be trying out both the 4 pole and 6 pole astig windings individually and seeing which are the best to use in the Marquee. I think the 6 pole coils have the most potential for improved sharpness.

As part of this process I will be making measurements of every coil to be used and/or modified, for future reference.

cmjohnson · Joined: 03 Apr 2006 Posts: 5180 Location: Buried under G90s

As I was browsing the Barco 909 service manual, I encountered this regarding the focusing operation.

It helps to explain why we have noticed that with non-stock focus yokes, we often see that the picture is rotated and we have to rotate the focus and/or deflection coils to straighten the image out.

mp20748 · Posted: Wed Mar 16, 2016 6:21 pm Post subject:

Here ya go CJ. Barco 909 coils using the wires and connectors from the Marquee coils. Follow the color scheme on the Marquee Stig/Focus wires...

cmjohnson · Joined: 03 Apr 2006 Posts: 5180 Location: Buried under G90s

Kewl. Thank you!

cmjohnson · Joined: 03 Apr 2006 Posts: 5180 Location: Buried under G90s

I need to add something really important that I just learned today.

My LVPS crapped out. So I put in a spare. Had the projector up and running in about two minutes.

I've ALWAYS fought unstable focus with changes in contrast on this projector. It's why I've pursued the frankenyone concept
so long, because I found the focus stability on my projector to be absolutely atrocious.

It's currently running STOCK yokes as I don't have a full set of "optimized" frankenyokes made up for it yet. Well, I'll soon be running Cine 9 yokes with some MP suggested mods to the focus board so I expect some improvements there.

But, anyway, a funny thing happened when I put the replacement LVPS in.

My focus stability issues went away. Shocked

I can take the focus to 90 before it blooms AT ALL.

With stock (but calibrated) Thomson yokes.

I've been able to achieve sharper focus than ever before, too. And not only because I have done all
focus and astig adjustments using a video camera pointed at the screen to see what's happening more clearly than
I can see with the unaided eye.

If you're having ANY kind of stability issues, don't rule out the idea that your power supply may be weak or unstable.

It seems that most of the issues I've had with getting the kind of focus I've always thought possible, but never really achieved,
were related to a weak LVPS all along and it wasn't until I swapped it for a better one that this became clear.

I can now easily count scan lines at 1080p in 16:9 squeeze mode. In 4:3 mode the scan lines have almost a full line's width worth of gap between them. You'd want to defocus that!

The lesson: Picture quality STARTS with a good power supply. It may be more important than anything else.

**AnalogRocks** · Posted: Mon Mar 28, 2016 3:32 am Post subject:

Sounds like you are onto something there CM. I found the same with PC's and cheap power supplies vs High Quality ones.

cmjohnson · Joined: 03 Apr 2006 Posts: 5180 Location: Buried under G90s

It would appear now that I've been giving the Thomson focus yokes a bad rap for several years now, when it was my LVPS that was really to blame.

I'm still absolutely sure that there are better focus yokes that can be adapted to use in a Marquee, and in fact, I'm sure that the right ones are an essential component if you want to resolve much higher than 1080p, assuming you have the bandwidth to do that. But the Thomsons do get the job done at and below 1080p. I thought they were garbage, now I know better.

My machine is resolving the picture well enough to see the jaggies in diagonal lines in the full HD picture. That's resolving every pixel.

The MP modded VIM and neck cards make that resolution possible. Best focus makes it possible to SEE it.

mp20748 · Posted: Mon Mar 28, 2016 6:35 pm Post subject:

The LVPS noise level is very critical to higher resolution performance. I hope to post more on this later, but much like the thread I started on my LVPS and the results I got after making certain changes to it.

I've since went further and have fixed the 5 volts rails (+/-) to also greatly reduce noise on the 15 volt rails (+/-). To now working on the 85 volt rails with great success also, also need to add something to the neck boards to finish the job.

The 14 volt rails have also been fixed, but only at the VIM. My goal would be to also tighten them up, that way I hope to achieve a more stable vertical deflection. Because there is a little movement there.

And all of the little leaded 0.1 decoupling (or pull-up) yellow caps in the LVPS are basically doing almost nothing, because putting any value in their place makes a change for the better whereas, the larger in size the cap is the better the results. I'm thinking those little yellow caps would do great in other stages or circuits, where the voltages they are tasked to do something about are not so aggressive and outside of the clock rates of processors.

I tell ya, I've been having a ball sweeping these circuits.. Very Happy

cmjohnson · Joined: 03 Apr 2006 Posts: 5180 Location: Buried under G90s

Mike, I just came into a pretty good sized sample kit of noise suppression components and other goodies that I think will be handy for quieting down the DC rails. It causes me to suggest that you might want to request similar sample kits, which are either free or at least cheap, as a low cost way of quickly getting a large variety of components for your experiments.

I am REALLY starting to regret selling off my best o-scope! It'd be handy right now!

cmjohnson · Joined: 03 Apr 2006 Posts: 5180 Location: Buried under G90s

I've had a series of conversations with various people including Tim, and the various service manuals for various projectors, and what I now believe is that the Marquee line of projectors has the WRONG yokes for its own tube application

Marquees were originally equipped with Thomson tubes that were designed to run at a 40KV anode voltage, and early Marquees equipped with those tubes had 40KV HVPS units.

Thomson provided their own yokes for use with their own tubes, and those tubes and yokes are designed to work with each other
and designed to work when the tube is running on a 40KV anode supply.

Believe it or not, the higher anode supply is part of the difference. Because with a higher acceleration potential, beam current required for a specific level of brightness is lowered.

It stands to reason that the Marquee was fully configured to make the best of Thomson CRTs, but at some point, I don't know
exactly when, the switch was made to Panasonic (MEC, Matsushita Electric Corporation) CRTs.

This required a few changes. The neck cards were redesigned for MEC tubes, and the HVPS was changed to 34KV, and a little later down the road, to 34.9 KV for a noticeable improvement in brightness and apparently an improvement in tube life as well.

But, very significantly, the engineers at Electrohome apparently considered the Thomson magnetics to be sufficient for use with
the MEC tubes. By making this choice they avoided having to redesign the HDM, VDM, and focus board, or request custom wired
deflection yokes from another vendor.

Notably, EVERY other projector manufacturer that adopted MEC tubes selected complete yoke sets made by Kanto-Denshi,
and it is my belief that K-D was the de facto, if not actual, partner with Panasonic for magnetics for Panasonic tubes.

I can not be sure of this but it is my belief that if yoiu were to have started up a projector company in the late 90s and approached Panasonic for tubes and magnetics, Panasonic would have provided, or at least recommended, Kanto-Denshi magnetics for use with their tubes.

So, not concerning ourselves with any Sony products at the moment, we can consider that the rest of the projection CRT market
was either Thomson or MEC/K-D. Thomson's tubes and magnetics were designed as a complete system to be used together
for best performance, and MEC/K-D's tubes and magnetics were also designed as a complete system to be used together.

In both cases, mixing and matching parts between systems has NOT worked out well. While I saw the potential for better focus performance in certain K-D focus yokes in a Marquee application about ten years ago, it was only very recently that I realized that I only had one piece of the puzzle.

Some actual experimentation with more than just the focus yoke has become necessary.

Cutting to the chase, look at the Kanto-Denshi yokes, both deflection and focus, as found in a Barco 909/Cine 9. Compare them to the Thomson yokes found in a Marquee on the SAME type of tube.

The K-D deflection yoke is SHORTER, the K-D focus yoke is LONGER. As compared to their Thomson counterparts.

We've noted that when attempting to use a K-D focus yoke, we see rotation in the raster.

That's because the K-D yoke is in the WRONG LOCATION on the tube neck.

The K-D deflection yoke is shorter so that the focus yoke can be placed correctly.

The Thomson deflection yoke is almost an inch too long to use with K-D deflection yokes (some models) and have it work correctly.

I now believe, and with a fairly high degree of confidence, that the true answer to getting best focus performance out of a Marquee is to adapt the ENTIRE Kanto-Denshi system to the Marquee.

But this, while it will take the image to the next level, will also take the art of modification to the next level as well.

For starters, the deflection yoke found on the 909/Cine 9 has to be totally rewired. There aren't even enough connections on the tag board for Marquee usage.

The vertical winding isn't a problem, and rewiring the convergence yoke isnt' a problem, either.

The problematic part is the horizontal deflection windings.

In a Thomson yoke, the windings are fully separated and independent. This is necessary as the HDM connects them in either series OR parallel according to the scan rate.

In the K-D deflection yoke, the windings are tied together at a common center point. THAT has to be separted and each winding brought out on its own pair of wires.

For starters, I suggest that you take a Thomson deflection yoke (and convergence yoke) and carefully pry off all three of the connector strips because you will want to reuse them. You will be removing the tag board from the K-D deflection yoke and you
will then attach all three connector strips from the Thomson yoke to the K-D yoke, in approximately the same relative positions. This will require judicious use of some JBweld or other epoxy, maybe silicone.

You need to be careful about placement of the connector strips. I recommend not placing them until all the wires are removed
from the original tag boards, the tag boards are removed, and the new strips are placed where you are sure that the wires reach.

Be careful when desoldering the wires. They can be a bit fragile. And in some cases, litz wire is used, so one wire is several independent strands of wire that are individually insulated except at the point they're soldered.

I found the vertical winding rewiring job to be straightforward. The convergence rewiring isn't too bad, either, but you do want
to be really careful here so as to preserve good wire geometry and DON'T disturb the evenly wrapped wiring running around the
ferrite ring in four locations.

The rewiring job on the horizontal coils is where will earn your stripes. I found this to be rather challenging because you have
to sort out the various individual windings and be sure you have them connected correctly. I recommend doing a LOT of
inspection of the coils along the way so you really understand the way they're constructed.

Once you've done all the rewiring and the harnesses are soldered to the connector strips, and you have continuity through all
circuits, you're not even close to done yet.

Why?

Because the inductance values of the different kinds of yokes require the driving circuitry to be modified to match up to the
new values.

Oh, yeah...there are balancing resistors in the focus yokes, too. You need to take that into account as well. They function as current limiters for the various coils and honestly I don't yet know what the values of the equivalent resistors for the modified
coils should be. But it does appear that you may need to include some.

Interestingly, some deflection yokes taken from Marquees have these resistors, and others don't. I do not know what that's all about.

Going back to the driver circuitry modifications, mostly what you will be doing is changing capacitor values so that the loop resonance (LCR circuit) of the system is brought back into the right frequency ranges.

I'm not getting into the details of that in this post.

If you don't know how to calculate the resonance of an LCR circuit, and if you don't even know what this means, don't try this yourself. Let those of us who understand it figure out the values and parts changes needed and we will then give you detailed instructions on how to make the changes that will be required.

If LCR and reactance and resonance are things you do understand, then you can test yourself by going into the Marquee service manual, seeing how the band switching works, and using the capacitor values on the HDM, VDM, or focus board as you C values, attempt to calculate the inductance range of the coils that are connected to that circuit. Or, alternatively, read the yoke inductance values, get the capacitor values, and calculate the resonance of the circuit.

If you're in the ballpark with your calculations, then you will be able to calculate the NEW desired values.

But a word of warning: Even if you do this, even if you buy the parts, even if it all looks good on paper and to the calculator,
I am NOT certain if the loads that will be drawn by the various coils will be within the capacity of the driving circuits to handle.

I don't know if the HDM, VDM, or focus driver boards are powerful enough.

I'm HOPING they are.