CurtPalme.com Home Theater Forum

[cmjohnson]

Register to remove this ad. It's free!

In this topic I'm going to log my experiments and results in taking the old "Frankenyoke" experiment to its logical conclusion, which is fully converting a Marquee to run with a full set of magnetics that are DESIGNED to deliver best results with Panasonic CRTs.

In briefest terms, here's the back story/history:

Marquees were originally made using Thomson CRTs and Thomson magnetics. Which are optimized for each other.

Every other projector manufacturer (well, all the ones we know well) chose Panasonic tubes, and Panasonic's chosen partner for magnetics (Deflection and focus yokes, including convergence yokes) was Kanto-Denshi, KD for short. From now on, when referring to Kanto-Denshi magnetics, I'll write KD instead.

Thomson magnetics are optimized for Thomson tubes.

Thomson tubes are substantially different from Panasonic tubes in at least one critical way: They were designed for
a 40 KV anode supply voltage. As compared to Panasonics at 34.9KV.

The difference is that when you run higher anode voltage, you need less current to achieve a designated brightness level from the tube. You also generate more x-rays.


It also doesn't require as strong a magnetic field when your beam current is lower.

The Thomson magnetics are optimized for the Thomson tubes and their higher anode voltage and lower beam current.


When Electrohome switched from Thomson CRTs to Panasonic CRTs, they elected NOT to redesign several boards in the Marquee in order to accomodate a switch to KD magnetics as well. They found that the Thomson magnetics still worked well enough
to meet the requirements of their customer base, so why bother? From a business perspective, it wasn't a bad decision.

But we home theater enthusiasts are in some ways more demanding. We want to be able to run higher contrast levels than you are likely to ever see in use in a flight simulator, and we want focus to stay sharp at high contrast levels. The Thomson yokes are not great at this, not with Pansonic tubes they're not.


Ultimately, the Marquee was only HALF adapted to make the best use of the Panasonic type tubes.

The adaptations that were not made: Complete changeover to KD magnetics. Adapt driver boards to work optimally with KD magnetics.


That's what I'm out to do.

But unlike the older Frankenyoke experiments, this is a "cut from whole cloth" approach. The magnetics were designed by their respective manufacturers to be used in complete sets, so that's what I'm doing.

Frankenyokes have not realized their potential for the simple reason that the longer Thomson deflection yoke forces the KD focus yoke to be too far back on the tube neck for best results.

So these projects INCLUDE adapting to use not only the KD focus yokes, but the KD deflection yokes as well.


I am pursuing two separate adaptation projects more or less concurrently.

Project 1, I call this the Marquee Cine 9500 project, is to use the KD magnetics found in a Barco Cine 9 or 909 in a Marquee.
This one offers a fairly high level of complexity because the modifications needed to the deflection yoke are substantial.
The adaptations require to the focus yoke are fairly straightforward, that's just a cable swapout as far as yoke mods are concerned.

The electronic component changes required to make these magnetics fully compatible with the Marquee have not yet been calculated. We'll get to that.

Project 2 is much the same but it is based on using a complete set (almost, I probably can't use the electronic CPC yokes) of magnetics off a G90. I guess I can call the resulting testbed machine a Marquee G9500.


In some respects this will be simpler as the deflection yoke doesn't need to be modified at all, aside from installing Marquee specific cables on it. The wiring is comparable, wire for wire, wind for wind, thus making the deflection yoke adaptation a trivial task. (Electronic component changes will likely still be needed on the HDM and VDM.) This also applies to the convergence coil clipped to the back of the G90 deflection yoke.

In my next post I'll be posting comparisons of the three families of magnetics, Thomson (Stock Marquee), Barco 909, and G90 magnetics, with photos, notes on what's different enough to require surgical modification, and, of course, electrical
values for the coils, including DC resistance and inductance values.

[Tim in Phoenix]

Hmmm

Marquee was designed to display small text from cad/cam terminals, and we had no problem doing that. Hard core HT enthusiasts seem to be rather more demanding than most of the commercial users I ever dealt with.

[cmjohnson]

True.

Of all the 9" LC coupled projectors, while the Marquee arguably has the best video chain, it also is generally acknowledged
that the weak spot on the Marquee is focus, particularly at higher contrast levels.

I am attempting to address this issue with a complete swap-out of magnetics, using the same systems that are proven to yield
sharper results in other projectors but using the same tubes.

It stands to reason that if I can duplicate a G90 or Cine 9's CRT and magnetics configuration, the only thing keeping me from matching their focus performance after that is just a matter of driving the magnetics correctly. Which will require mods to, at the very least, the HDM and the focus/FGM board. Convergence may also need some mods as we're changing out those coils, too, but I think astig will be OK based on previous experiences with other KD focus yokes.

Now, granted, both the Cine 9 and the G90 have extra electronic coil systems the Marquee does not have. The Cine 9 adds a second
astig winding (6 pole, as I understand it) to the focus yoke, and the G90 has electronic CPC adjustment. But I don't think that these extra systems are so critical that I can't achieve improvements in focus importance without using them.

Attempting to add circuitry to the Marquee to run those extra coils is quite beyond my intent and it would certainly stretch my electronic design capacities farther than I've ever pushed before. So don't look for that.

[ronaldus]

Hi

you wrote

"It also doesn't require as strong a magnetic field when your beam current is lower.

The Thomson magnetics are optimized for the Thomson tubes and their higher anode voltage and lower beam current."

I don't know if that is correct.

I work on a particle accelerator I have learned that the magnetic force that you need to bend the particles (in this case electrons) depends on the energy (the kV's of the anode voltage) and not on the beam intensity (beam current).

Also if you consider the CRT as a lineair accelerator then at the kathode the energy is still low but the further you go towards the anode the beam energy is higher and when the electrons hit the phosfor they are at 34.9KeV. So the further you put the focus coil towards the phosfor the harder it is to focus.

Is the focus coil a dipole quadrupole or higher order magnet by the way?

Just a few remarks to discuss about. Would this make sense?

regards,

Ron.

[cmjohnson]

My observations are that the KD focus yokes have stronger permanent magnets in them as compared to the Thomson yokes, but I do not have the capability to do tests on the dynamic field strength of any focus yokes.

The magnetic configuration of the permanent magnets in the Thomson yokes is that there are six cylindrical magnets evenly spaced in a ring, around the perimeter of the yoke, and the ends of the magnets are glued to ring shaped plates at either end.

The KD yokes use permanent ceramic ring magnets separated by magnetically permeable spacers, I do not know the alloy the spacers are made of but I think they're just soft iron or close to it.

My previous experiments with KD focus yokes consistently showed two common problems: Sharpest focus was out of the range of focus adjustment, and the range of defocus, both high and low, were, if anything, too great, which is what I would expect of a
more efficient and more powerful magnetic system.

I actually think it may be necessary to current limit the focus yokes with current limiting dropping resistors so as to reduce the available focus range to something more reasonable, and by restricting the focus range, also improve the fineness of focus as well.
It would be annoying if a focus setting of 49 was too low but 50 was too high.

[cmjohnson]

Time for some measurements.

Starting with the Thomson deflection yoke:

Horizontal windings (there are two, the HDM connects them in series or in parallel as needed according to H. scan rate)
measure to have an inductance of 0.28 milliHenries each. They have a DC resistance of about 1.1 ohms. (My meter is not
calibrated. Some uncertainty exists in DC resistance measurements.)

Vertical winding: DC resistance of 15 ohms. Inductance of 8.8 milliHenries.

The Sony G90 deflection yoke's readings: (It's very similar, it will not need any physical mods to it other than changing out the wiring harness.)

Horizontal windings read an inductance (each) of 85 microHenries. Not milli, MICRO. WAY lower inductance than the Thomson windings. Interesting. This implies they're made to run efficiently at high scan rates.

DC resistanced of H windings: So low I'm not able to read it with any degree of confidence. My meter zeros out at .5 ohms when you short the probes together. The windings read about .7 ohms so the difference is .2 ohms.

There's not a lot of wire on the H windings. That implies they may take a fair amount of current to drive. This is going to be a
matter of concern. WILL the Marquee HDM be able to deliver enough current? Won't know until I try it!

G90 vertical windings: Inductance of 12 milliHenries, DC resistance of 16.9 ohms.


Barco 909/Cine9 deflection coil measurements:

H. winding:
0.3 ohms, note, this is before separating the two windings as needed for Marquee usage. Assume 1.5 ohms per wind.
130 uH

V. Winding:
5 ohms
4 mH



Now for the convergence coils:

Thomson: 0.1 ohms DC for both windings, 25 microHenries
G90: 0.1 ohms DC for both windings, 28 microHenries
909/Cine9: 0.2 ohms DC per winding, 40 microHenries


These numbers are close enough that I'm confident that no mods to convergence drive circuits will be needed if you choose to use G90 deflection/convergence coil assemblies in a Marquee. Neither should it be any problem to use the 909/Cine9 convergence coils.

[gjaky]

As I know the G90 does not use the distributed H-DEF coils like the Marquee so how are the windings factory wired parallel or series?

[cmjohnson]

The G90 yoke does in fact use two separate and identical H windings, each with its own pair of pins.
If the G90 does not do the series/parallel switching like a Marquee, then they would presumably be wired
one way or the other by the G90 circuitry.

It's the deflection yoke for the Barco 909/Cine 9 that doesn't have a pair of individually terminated windings.

But it does have two windings. They are connected (soldered) at the midpoint. They can be separated, I'll get into the details for that later. That deflection yoke conversion is fairly involved.

At this time, the G90 deflection and convergence yoke assembly looks like it'll be absolutely EASY to make run in a Marquee.
The 909 deflection and convergence yoke assembly is considerably more involved but I will have to adapt a full set of each,
as the whole point of this experiment is to test FULL complete sets. I don't want to try mixing and matching magnetics made for different applications until I've tested compete sets first.

It MIGHT be that the ultimate combination could be, for example, the 909 focus yoke with the G90 deflection yoke.

But that kind of experimentation has to wait until I've got a fully modded Marquee optimized for and running a G90 magnetics stack, and another one optimized for and running a 909 stack.

Yes, I've got two projectors that I'm dedicating to this experiment. I actually currently have six Marquee chassis, and
in just a few more days all six will be 9500LCs with LUG tubes. So I am not lacking for equipment to do these experiments.

[gjaky]

As I see the G90 does not switch the coils, they are tied permanently parallel.

[cmjohnson]

That's interesting, and fortunately it also makes adaptation for usage in a Marquee very easy.
Wire for wire, G90 deflection yokes are equivalent to Thomson deflection yokes, and the overall
range of values is generally close enough that I don't see any show stoppers yet.

I actually think that the deflection yoke adaptations will be the easy part. Adapting the focus yokes
will almost certainly require circuit modifications. If nothing else, capacitor values in the drive loops
will have to be readjusted to maintain system resonance.

I'm going to try to post some photos today, and record some data on the focus yokes as well.

[cmjohnson]

Some useful data from the Marquee service manual. Mostly this is for my own quick reference as I revisit this topic.

The turnover point between series and parallel yoke winding switching is 60 KHz. If the scan rate is below 60 KHz, the deflection yoke windings are connected in series. At 60 KHz and above, the yoke windings are connected in parallel.

1080p-60 requires a scan rate above 60 KHz so the parallel winding rewiring job on the Barco deflection yokes is mandatory.




Focus system information:

The focus board, like the HDM, uses band switching capacitors to adjust the resonant frequency of the focus circuit to match
the scan rate currently selected.

The capacitors of interest are identified on the FGM or FCM as C117, C118, C119, and C120 for the red channel,
and C217 thru C220 for the green channel, and C317 thru C320 for the blue channel.

Since they're identical I'll just refer to the red channel and any changes made here will be made to all three channels,
and that will be assumed from now on.

The values of those capacitors: All are 250 volt rated Vishay "box" film capacitors.

C117 is .033 µF and this capacitor is ALWAYS in the resonant circuit. It is not band switched.
C118 is .022 µF and this is band switched.
C119 is .068 µF and this is band switched.
C120 is 2.2 µF and this is band switched.

Incidentally, to get the µ symbol (micro), hold down the ALT button and type 0181.


The Thomson focus yoke's values are:

30 milliHenries (mH) for the static coil
45 microHenries (µH) for the dynamic coil


The service manual does not go into such detail as to specify exactly which capacitors are switched into the focus circuit
at a specific frequency range. While there's enough information to determine the switched operation of the deflection system,
the data on the focus system is a bit more sparse.

But no matter, we can figure it out.

Since the switching system ADDS capacitors, it stands to reason that the highest scan rates will cause no, or the
smallest, capacitor to be switched into the circuit.

So let's begin figuring this out using the smallest values, for the highest scan rates.

I call attention to the fact that the frequency switched capacitors apply ONLY to the DYNAMIC focus system and coil,
NOT to the static focus coil.

I also call attention to the fact that there are no capacitors in the STATIC focus system that serve the same resonance
adjusting function. We may need to add one. That should not be difficult. It would be added in series with R149, 249, and 349,
if necessary.

But we can get back to that later. Let's stay focused on (pun intended) that dynamic focus system.

The dynamic winding of the focus yoke is 45 µH and the smallest capacitor value in the loop is 0.033 µF.

The DC resistance of the dynamic winding is 0.6 ohms. This is relevant, and is part of the resonance formula.

So we now have the R, L, and C values. (Resistance, Inductance, and Capacitance)

R = 0.6 ohms
L = 45 µH
C = 0.033 µF

Applying the series resonance formula, this combination gives a resonant frequency of 130,604 Hz, which is a nice fit for
the top scan rate range.

Let's recalculate by adding the .022 µF cap to that circuit, giving a new cap value of .055 µF to the formula.

Result....new resonance frequency is now 101,166 Hz.

So going down all cap values that are possible in the original unmodified circuit, here's the rundown:

Before we get into this, understand that while the whole focus circuit is a loop, with all components in series for all practical
purpose, the band switching capacitors are in parallel with each other, which both simplifies the math and makes the process of
figuring the total value a simple matter of adding up their values. If they were actually connected in series to each other, then
that changes the formula dramatically but we don't have to worry about it.

C117 alone, .033, 130,604 Hz
C117 plus C118, .033 .022 (.055) 101,166 Hz
C117 plus C119, .033 .068 (.101) 74654 Hz
C117 plus C120 .033 2200 (2200.033) (Yes, that 2.2 µF cap is HUGE compared to the others.), 15995.6 Hz

C117 plus C118 plus C119, .033 .022 .068 (.123) 67649. Hz
C117 plus C118 plus C120 .033 .022 2200 (2200.055) 15993.5 Hz
C117 plus C119 plus C120 .033 .068 2200 (2200.101) 15877 Hz

C118 plus C118 plus C119 plus C120 .033 .022 .068 2200 (2.233) 15995.2 Hz
(There are a few minor value errors in the list above, please don't worry about it, they're right BELOW where the data has
been weeded out to separate the useful stuff from the useless stuf.)

As you can see, once the big C120 cap is switched in, the other caps only change the resonance by less than 1 Hz so we can safely assume that several of these combinations are not used at all.


So let's weed out to a more likely list:

C117 alone, .033, 130,604 Hz
C117 plus C118, .033 .022 (.055) 101,166 Hz
C117 plus C119, .033 .068 (.101) 74654 Hz
C117 plus C118 plus C119, .033 .022 .068 (.123) 67649 Hz
C117 plus C120 .033 2.200 (2.233) 15877 Hz



OK, that's pretty reasonable. Actually I think that there's one option in this list that isn't actually selectable, but right
now I don't know which one that would be.

What's interesting to me is the big gap between 16 KHz and 67 Khz. Perhaps the cap value choices could be improved even
when using stock magnetics? I think so!


So now we have a rather clear picture of the stock focus system.

Let's look at it when we change the value of the dynamic focus winding to that of a different KD focus yoke I intend to try to use.

Choosing the one from a Barco 909, which is a KF-3212, I find that its dynamic winding value is, apparently. 42.5 µH
which is really not far at all from the 45 µH reading of the Thomson focus coil, which is very encouraging.
It's close enough that MAYBE no electronic changes are needed to the dynamic focus circuit.

But let's find out.


So, assuming I've got the right winding, and I think I do, then here's how this changes the game with regard to resonant frequencies:

New inductor value = .42.5 µH

C117 alone, .033, 134,391 Hz
C117 plus C118, .033 .022 (.055) 104098 Hz
C117 plus C119, .033 .068 (.101) 76814.4 Hz
C117 plus C118 plus C119, .033 .022 .068 (.123) 69610.2 Hz
C117 plus C120 .033 2200 (2.233) 16337.3

These values are really close! Well within 5 percent. I'd say that we don't need to make any cap value changes to the FGM
in order to use the KF-3212 focus yokes, we just need to put them in the right place on the tube neck, which requires
using the shorter Barco deflection yoke assembly.

But what about the static focus winding?

I'll visit that next time. I've done enough for one sitting!

[cmjohnson]

As I suspected, I'll need to do a little more work than simply change over some wiring in the Barco deflection yokes to make them run. My test article would not permit the projector to completely power up, and I believe it's due to the lower impedance and resistance values in the Barco deflection yoke.

For test purposes, I'll probably have to add some resistors in series to achieve power-up. Once I've established the parameters that are needed to solve the power problem, then I can work on figuring out how to integrate such required changes into the yoke,
and maybe the HDM as well.

[cmjohnson]

I just want to mention that at this specific moment, the best performing focus yoke I've tried on a Marquee is the KF-3203 type. The KF-3212 (Barco 909/Cine 9 type) has not yet achieved satisfactory results in my own tests up to this moment.

I have not even yet logged the electrical parameters of the 3203 for comparison's sake, but as a focus yoke, it works, even with a stock Thomson deflection yoke.

[cmjohnson]

I made some useful observations this evening.

I have acquired some magnetics from VDC that were to be tossed out, a varied box of focus and deflection coils, including some
that are absolutely unique, one-off products made for experimental purposes.

One of them is an extended Thomson focus yoke, with a longer shell and something different inside. Not sure what. I'll investigate what that does vs. a regular Thomson focus yoke at a later time, but since there's only one of them, even it it turns out to be the sharpest yoke ever made I can't really make the best use of just ONE of them.

Another item is a very small (relatively speaking) deflection yoke, already wired for a Marquee. I believe it was made for some Sony product originally. No idea which one.

Since it's so short, I realized that it would be ideal for evaluating various focus yokes to determine where they work best on the tube neck. So I set up that experiment.

First, it runs fine. But it does not have the strength to get a full raster. A maximized raster with it is maybe 2/3rds of the available CRT face by width. Nonetheless I can use it for these tests.

Using only this short deflection yoke, which I now call my reference deflection yoke, and NO convergence yoke, since it's of no use in this experiment, and I need the free tube neck space, I tried out four different candidate focus yokes, all completely disconnected from any electrical signals, and positioned each one on the tube neck and marked where the back end of the clamps ended up on the tube when they were positioned for sharpest natural focus. I also measured the gap between their front faces and the back end of the reference deflection yoke.

The four yoke types tested are the stock Thomson yoke, the G90 yoke, the Barco 909 KF-3212 yoke, and the Barco KF-3203 yoke which came from some other model Barco projector, not sure which one but it appears to be similar to the 3212 except it lacks the 6-pole coils that are unique to the KF-3212.

I found that there is a very wide range of optimal yoke placements. They vary by as much as almost 19mm in placement relative to the deflection yoke.

The stock Thomson yoke requires a spacing of 38.8 millimeters from the back of the reference deflection yoke. I remind you, that reference deflection yoke is MUCH shorter than the standard deflection yoke AND there is no convergence coil on the tube neck, either. So this is not a number YOU can make any use of, other than to see how the different yokes require different tube neck positioning.

The G90 yoke wants to be closer to the deflection yoke than any other, with a spacing of only 20.3 millimeters.
The 909 KF-3212 yoke comes in at 36.9 millimeters spacing for best focus.
The KF-3203 yoke requires 26.5 millimeters spacing.

In every case I marked the position of the rear of the yoke clamp on the tube neck with a sharpie so I could have visual references to go by when trying them out with the standard deflection and convergence yoke assembly.

I must note that I'm assuming that all these yokes are at the magnetic strength they're supposed to be at, which I can not
prove in any case. I do know that I used a Thomson deflection yoke that was calibrated not too long ago by VDC. So I trust that one to be about right.

I then removed the reference deflection yoke and mounted the standard deflection and convergence yokes, and rechecked ever one of the four focus yokes to see if they will achieve optimal focus.

The G90 did not. Nor did I expect it to.
The stock yoke did, no surprises there.
The 3212 was close, not quite there but really close.
Surprisingly, the 3203, despite showing that it should have needed to be closer to the deflection yoke in previous testing,
seemed to come into perfect focus before touching the back of the convergence yoke.

I surmise that there is a difference in the optimal position for some focus yokes when they are placed close to other active
magnetics. This actually seems quite logical.

So, pleasant surprise of the evening, from the standpoint of static magnetic performance, the KF-3203 type yoke appears
at this time to be the best candidate if you are planning to use a stock deflection and convergence yoke system.

I have not yet measured the inductance of the 3203 yet, but I'll be doing that shortly and then working the formulas to determine what, if any, capacitors need to be changed on the focus board to drive it optimally.


One thing to point out, though, is that it may be that the optimal place for a focus yoke is not where it is sharpest, but instead, is just close to that spot, requiring a small amount of drive current through the yoke for the purpose of making the magnetic field conditions stable. I remember that Tse told me that the Marquee's focus system is operating at optimal drive levels when static focus is at 60, not 60. So that could affect where we really want to put the focus yoke.

[cmjohnson]

Just got done doing trials with a G90 deflection yoke wired for a Marquee.

In a word, it is AWESOME. Super efficient, I have H and V sized backed down to zero and the raster is still too wide for the tube face by a little. I'll have to adjust the size coils to get it down.

This is a GOOD thing, because it means that it's efficient, and will allow the HDM to have to do less work, generate less heat, be more reliable, and be more stable.

I can tell now that as long as I can tame the width and put it where I want it, I'll want to run these deflection yokes just to make the HDM run cooler and last a very long time.

The convergence coils that are part of it, though, may be too much (or too little) for the convergence system to handle. I will look into that...in the next exciting episode!

[cmjohnson]

Convergence coils on the G90 deflection yoke assembly work FINE in the Marquee with no mods necessary other than changing over the wiring harness. Electrically they are nearly the same thing, with DC resistance and inductance values being so closely matched as not to matter.

However, that doesn't mean that you want to use them as they are! Reason: They're swinging HUGE rasters widths even with the size controls backed off to zero. The raster is still beyond the limits of the CRT face.

So while they're electrically compatible, we're going to have to back off the drive on the HDM or pad down the yokes with curent limiting resistors.

If attempting to use a G90 deflection yoke in a Marquee, do remove the black plastic circular front cover that clips onto the face of the yoke. By doing so it'll allow the yoke to sit a bit closer to the front of the CRT and reduce raster size slightly. You'll already have too much raster size and will want to trim it down by adjusting the size coils on the HDM.

Word of caution: Use a proper plastic hex coil adjustment tool. Do NOT attempt to use an allen wrench. Not only will you break
the slug, it'll totally alter the results as it is a highly magnetic object in a system that depends on magnetic field strength in order to set the width of the raster. Update: The width controls on the HDM don't have the control range needed to tame the wildly efficient G90 deflection coil so I'm going to be looking at ways to current limit the system.

The vertical coil will also need to be tamed. I think this should be done with resistors at the deflection yoke rather than attempting to mod the VDM. Strapping resistors across the vertical coils to the center tap on the yoke terminal board is a simple matter.

As for values, there are already a pair of 560 ohm, 1/4 watt resistors on the back of the terminal board. I think that a lower value is going to be needed. I'll try 220 ohm, 1/2 watt as a starting point.

[cmjohnson]

To add to the fun, I'll have to take a close look at the impedance constraints on the HDM as I have found, by extensive testing and reading the service manual, that there is a current sensor on the HDM for the deflection coils but it ONLY applies to the green circuit, so when I connect the rewired G90 yoke to red or blue, it powers up but when I connect it to the green channel, it doesn't power up due to either too little or too much current being drawn. Since the width of the raster is already excessive, adding a series resistor in the loop, or changing to a larger value one, is probably exactly what I need to do.

Anyway, this has been a lot of fun so far and I'm learning a lot. I'll get the G90 deflection yokes sorted out so I can run three in the projector at once and once that's done then I'll find out what focus yokes are most suitable for adaptation to a Marquee.

[cmjohnson]

Turns out my clamp-on digital ammeter is good for reading frequencies up to 999,999 Hz and up to something like 400 amps AC or DC. So I can easily measure deflection yoke current.

The G90 deflection yokes want to draw 610 mA while stock deflection yokes want to draw about 150 mA. (On any lead of the coils.)

Clearly I'm going to have to limit the current draw on the G90 yokes. Looking into ways to do that. I hope to do it by reducing the drive level to the HDM deflection amps. Maybe even with a trim pot.

[cmjohnson]

I know it's a bit early to be mixing and matching but I did want to do a quick test, and found that a properly positioned KF-3203 yoke behind a G90 deflection yoke delivers completely acceptable results. I'm not saying perfect, not yet, because I had to push the edge focus zones to close to their lower limit, so some tweaks are needed, but I was able to get corner to corner sharpness.

Interestingly, a gap between the deflection and focus yokes is required for best results.


I strongly suggest you try this if you are not running stock magnetics: Disconnect the focus yoke and move it forward and back on the tube neck until you find the point of sharpest focus. That's test 1.

Test 2 is to reconnect the focus yoke, set focus to 50, and then slide the focus yoke back and forth until it is again at the point of sharpest focus.

For both tests, see if you can focus the edges and corners.


I will consider a focus yoke to be 100 percent successful if central focus can be achieved between 40 and 60, and edge and corner focus can be achieved between 20 and 80.

[cmjohnson]

G90 focus yokes are kind of confusing. Even more so if you have the complete wiring harness from a G90 (which is the work of a sadist, I am sure. I have never seen such a complex wiring harness in a single piece of consumer electronics, EVER!) because the harness cabling isn't labelled!

The G90 focus yoke has two connectors and 13 wires. One is ground, that's easy. One connector has four active pins, and I don't even know if that's for focus or if that's one of the two astig windings. (The yoke contains both 4 pole and 6 pole astig windings.)

I don't want to open up the yoke magnetic assembly to find out.

I'll post more about them soon.