Mitsubishi PCB-CRT Modification to Remove Ringing

This document is the property of Craig Rounds, all rights reserved, and is being shared with The Home Theater SPoT. As such, it is unlawful to reproduce, in part or in full, distribute, or link to this document directly, without the express written consent of the author. Do not quote this document without citing the original thread on which this discussion can be found in the Mitsubishi Tweaks section of The Home Theater SPoT. Please direct all questions and answers to the Mits Tweaks in Progress section of the SPoT regarding the following discussion.
_________________________________________________________________________________
WARNING: toc
The following instructions are for informational purposes only. Doing the wrong thing could very easily render your TV unusable. I strongly recommend that you do not perform this modification procedure yourself. Do not play with anything inside your TV. Taking the front panel off your television will expose you to the potential of electrical shock, therefore use caution. This is for informational purposes only.

I must reiterate, the voltages found inside these TV sets are very high, up to 30,000 volts. Being hit with such a high voltage could be fatal or render very serious injury. Any actions taken by individuals are at their own risk. I am not advising that anyone attempt to do anything based on what they see here, I am only providing information about the function of the TV sets. Neither I, nor The Home Theater SPoT, can be held responsible for any damage or injury that may result from attempting this modification.

This modification may also affect your warrantee status. If components are damaged by following this procedure, Mitsubishi is not responsible. If damage or injury results from actions taken while applying this modification or as a result of this modification, no attempts should be made to hold Mitsubishi responsible.

Any actions taken by you or other individuals are at their own risk.
PCB-CRT Modification to Eliminate “Ringing”

Foreground:  toc
As in my last article, “Limitations of PerfectColor On the Mitsubishi xxx13 RPTV Explained”, I will be using as few words as possible to explain a very complex situation.  As a result, quite a few details will be completely left out in order to actually get to the point.  However, if you read this article you will almost certainly learn quite a bit about how an analog RPTV renders video and what you can do to improve the image quality of your Mitsubishi or Sony RPTV.
Also, I would like to thank Robert Vermazen for sharing his incredible knowledge on this subject with me.  Robert developed the theory behind this procedure while I carried out the modifications and took measurements.
Brief Summary:  toc

What follows is a rather lengthy discussion on some limitations of modern RPTVs. The causes and effects of these limitations are explored. A modification procedure is then outlined that can be applied to any television incorporating the TDA6120Q video output amplifier. Some of these televisions include Mitsubishi RPTVs from the v19-v25 as well as many Sony RPTVs. This paper closes with some simulated questions and answers in appendix A, as well as some other brief information in appendix B on the application of this modification to Sony televisions.

Background: toc

I’m sure that most of you are already aware of “ringing” and the fact that all Mitsubishi RPTVs from the v19-v25 have excessive “sharpness”.  “Ringing” is a term used to describe an artificial “halo,” or “ghost image” that surrounds edges on video images.  You can easily examine this phenomenon yourself by simply displaying the AVIA “sharpness” or “inverted cross-hatch” patterns.  In these patterns, to either side of all black vertical lines you can observe this ghost image.  This image is not part of the DVD but is imposed by the circuitry in your television (and possibly DVD player).  These ghost images are the unwanted byproduct of “edge enhancements,” which are used to make the apparent detail of an image greater.
Bjoern Roy put together an excellent article describing the effects of edge enhancements (EE) called Ultimate Guide to ‘Edge Enhancement’.  While his article focuses on EE imposed during the DVD mastering process, the effects he describes are identical to those created by the Mitsubishi EE “system” incorporated into the more modern chassis.  I encourage you to read his article as it sites numerous real world examples of how EE affects picture quality.  I will point out, however, that it seems Hollywood has improved dramatically since his article was written.  There have been some very fine DVDs released lately that do not suffer from EE including “The Two Towers”, “Tomb Raider II The Cradle of Life”, ”The Matrix III Revolutions”, and many others.
These EE, whether they are imposed by the video mastering process or the display device, are very real but are also very much avoidable.  These “features” are currently being imposed on nearly all RPTVs on the market today.  The Mitsubishi and Sony RPTV in fact incorporates two different types of EE.
The first type is software controlled and includes three different kinds of “enhancements” including VSM, CTI, and LTI.  Most of you are already familiar with VSM (velocity scan modulation).  It is adjustable on most sets through the service menu and also happens to be a very malicious form of EE.  CTI (Chrominance Transient Improvement) and LTI (Luminance Transient Improvement) are two other kinds of software EE. LTI is quite malicious but CTI is only somewhat noticeable and only affects chrominance.  The only way to adjust these functions on Mitsubishi RPTVs is via direct communication through the I²C buss to the control EPROM.  Both CTI and LTI can be adjusted in the service menu on most Sony RPTVs.
The second type of EE is hardware imposed.  It is by far the strongest single form of EE used by Mitsubishi and Sony.  All Mitsubishi RPTVs from the v19 through the v25 (as well as many modern Sony’s) use three TDA6120Q video output amplifiers, one for each gun.  These amplifiers are tuned in such a way in order to amplify the high end of the video spectrum.  In effect, it is much like cranking the treble up on your stereo to induce excessive highs. In its application here, we are dealing with what is called “overshoot” or pre-emphasis. The idea behind pre-emphasis is to attempt to extend the video bandwidth further down the signal chain.
Equipment Profile:  toc
The three TDA6120Q video output amplifiers used by Mitsubishi (along with Sony) are made by Philips Semiconductors.  One of these chips can be found on each PCB-CRT of your affected PRTV.  Below is the block diagram for the chip (Figure 1).
Figure 1 Block diagram for the TDA6120Q video output amplifier
Notice that between pin 1 and pin 3 there is a simple RC network circled in red.  This network controls the “pre-emphasis” of the amplifier and affects the frequency response of the chip. This circuit functions in a similar manner as a crossover does in a speaker system.  For those of you not familiar with crossover systems, in simple terms, a crossover utilizes resistors and capacitors to adjust for frequency response.  You can change the way a crossover behaves by altering the values of the resistors or the capacitor(s).
The idea here is to modify the “crossover” so it allows the frequency response of the amplifier to behave naturally without imposing any extraneous influence on the signal (in this case ringing resulting from overshoot).  This is a little complicated because the RC network on Mitsubishi RPTVs, on each of the three PCB-CRTs, on each chassis, is different. As an example:

on the v21, R1 on the blue gun is likely different from R1 on the red gun, and is likely different from R1 on any of the guns in the v23.

In order to simplify this and to come up with a modification applicable to all affected Mitsubishi RPTVs, Robert Vermazen proposed that a variable trimmer capacitor be substituted on each of the three PCB-CRTs.  This allows for each amplifier to be “custom” tuned reducing the excessively high frequency response.  A trimmer capacitor (Sprague Goodman part number GCL3000) is available from Digi-Key by searching for the Digi-Key part number SG1036-ND.  This capacitor should be applicable to all Mitsubishi models incorporating the TDA6120Q video output amplifier.  This capacitor fits perfectly into the PCB-CRT once the original C1 capacitor shown in the block diagram is removed. The SG1036-ND will not fit on the PCB-CRTs of any Sony models.
Preliminary and Prep Work:  toc
Before you begin the modification there are a few things that must be done to ensure the best possible calibration.  First, all edge enhancements that can be adjusted on your chassis should be completely disabled.  This includes disabling VSM on all models, and turning off both CTI and LTI on chassis with known I²C applicable mods.  After this, optics should be thoroughly cleaned, followed by both electronic and mechanical focus and of course fine convergence. If you have not already done so, it may also be worth considering a “ground up” geometry prior to the modification.
You must also have the following equipment and supplies on hand:

1) Service Manual for the chassis

2) Three GCL3000 trimmer capacitors

3) PCB temperature controlled soldering iron

4) Solder appropriate for PCBs

5) Desoldering Braid

6) Phillips head screw driver

7) Nonconductive flat head jewelers style screw driver

8) Test Patterns: Multiburst, Sharpness, 50 IRE gray window, 30 IRE gray window

Optional equipment includes:

1) Oscilloscope

2) 1080i test patterns

3) Solder suction

4) Extra flux

5) Assortment of colored “sharpie” markers

6) Extra GLC3000 trimmer capacitor (recommended)
To begin, display a 50 IRE gray window on your TV. Observe the left vertical border between the black field and gray window. You will notice that some of the black from the field “overshoots” into the gray window. The overshoot will extend into the window approximately 3.5 mm.

PCB-CRT Modification Procedure: toc

Under normal operation the video output amplifiers run at around 200VDC.  Even when the television is turned off this voltage is retained for quite some time and can remain dangerous.  The first step of the modification procedure is to discharge this high voltage section.  The easiest way to do this is to simply unplug the TV for an hour or more.  This should drain the 200V section of the power supply.
The second step is to open your service manual and to look at the schematics for the PCB-CRTs.  Use the schematics to determine the part number of the C1 capacitor in your chassis, then use the PCB Layout Diagrams to locate the placement of the capacitor on the PCB-CRT.  These boards are not very complex and these two steps are not very difficult at all.  Start by finding the TDA6120Q chip and then look at the RC network running across pins one and three.
Remove the front access panel on your TV.  Work on only one CRT at a time.  Start with the green gun, then do red, then do blue.  This way you will not confuse which wiring harnesses go to which gun. Also, if you somehow get confused you will have the other two guns as a reference as to how everything is laid out.
The third step is to unplug all wiring harnesses from their brackets on the PCB-CRT (Figure 2). Wire Harnesses may also be behind the CRT assembly that are not visible in figure 2. Figure 3 shows the locations of these harnesses on the v23 after the PCB-CRT was removed. Figure 3 also shows two wires that should not be removed because they are soldered to the board (in this case the orange and gray wires).  I think on all chassis, that all of the harnesses are keyed and sized differently to prevent the accidental mix-ups of plugs when reinstalling the harnesses.  However, it would be a good idea to label each harness before it is removed to ensure it winds up in the same place when you put it back.  My favorite way to do this is to have on hand an assortment of colored “sharpies” and to draw a unique colored line going across the wiring harness and the wiring bracket before it is removed.  Also, as indicated, figures 2 and 3 are photos of the green CRT on the v23.  If you have a different chassis you can expect to find different layouts for the wiring harnesses and for many of the wires to be different colors.  Also, the wires that are permanently soldered to the CRT may be in slightly different places and be different colors.  The location of the C1 capacitor on the CRT-PCB may also be different.
Figure 2 Photo of green CRT and CRT-PCB Assembly
On some chassis, there may be a small amount of glue between the base of the CRT tube and the white CRT socket on the PCB-CRT.  If your chassis has this glue, carefully remove it by lightly scraping with a sharp object or by slitting the glue with a razor blade along the seam of the CRT and socket junction.  Be careful not to damage anything and don’t apply too much pressure.  Work very gingerly and the glue should come off without too much of a fight.  There are some chassis, including the v23, that do not have this glue so you can ignore this step.
Loosen the Philips head screw (Figure 2) holding the black support bracket to the CRT.  This support bracket is fixed to the CRT-PCB and is clamped to the CRT by means of this screw to ensure the PCB does not fall off the back of the CRT.  You do not need to remove this screw.  Only loosen it until you can use your finger to “open,” or loosen, the clamp to allow the PCB-CRT to slide off.
Now, carefully and gingerly, apply force downward to the white CRT socket in order to slowly slide the CRT-PCB from the back of the CRT.  Once you have the PCB-CRT removed, visually examine it and locate the C1 capacitor.
Figure 3 Photo of top view of CRT-PCB after it has been removed from the back of the CRT.  Notice the Amp attached to the heat sink and where the GCL3000 trimmer capacitor will be located.  Also notice that the thick orange and gray wires are permanently fixed to the board.  The gray wire actually goes into the white CRT socket.
At this point the PCB is free from the CRT and only the permanently soldered wires connect the CRT to the chassis.  You may be able to get some more slack by tracing these wires back into the chassis and unhooking them from any wire ties they go through.  Now, working through the front access panel, with the board hanging by these remaining wires, remove the original capacitor and replace it with the GCL3000 trimmer cap.  I must assume you, or the person performing this modification, is experienced with PCB repair and is competent in the use of PCB soldering irons and solder removal techniques.  I will however point out that this job is quite simple and it is not necessary to use a solder suction.  Simply using a bit of the braided copper solder removal stuff works just fine.  If you are skilled, it should be pretty easy to remove the original capacitor without damaging it.  This way, if for any reason you were unsatisfied with the modification, the capacitor could be returned to the board.
Also, before you solder your new trimmers caps to the boards, you should consider whether they should be mounted above (as shown on this example of a 55613 in figure 3) or below the boards.  Make this decision based on the need to adjust the capacitors with the TV turned on and with the PCB-CRTs mounted to the CRTs.  Remember, this is a 200VDC section and you must be able to adjust these trimmers without making contact with any other part of the TV.
Reinstall the PCB-CRT onto the CRT by following the removal procedure in reverse.  Here are some notes and cautions:
1)      When you slide the PCB-CRT back onto the CRT, even though it’s keyed, it is possible to bend the pins on the base of the CRT.  Be careful not to do this.  If this does happen, use small electrician’s pliers to straighten out the pin(s).  This is difficult, however, and will likely result in several additional hours spent on this modification.

2) When you tighten the Phillips head screw that clamps the black support bracket to the CRT use very light force. This screw should not be very tight and too much force could cause the CRT to shatter. That would be bad.

3)      If glue was holding the CRT-PCB to the CRT, you do not need to replace it.  The PCB-CRT should stay in place just fine without it.
Now, repeat this procedure for the remaining two CRTs.  Once you have completed the modification to all three guns I would suggest you take a short break before you turn the TV on; have a coffee or something.  Then go back and look at your work again.  Double check and make absolutely certain that all wiring connectors are put back in their proper places and that everything is back and installed properly.
Calibration of the GCL3000 Trimmer Capacitor:  toc
Leave the front access panel off and turn on the TV.  Allow the television to reach full operational temperature.  If you have access to a 1080i pattern source, connect it to your television.  If not, a 480p source is almost as good and will suffice.  Display a 50 IRE gray window and be sure your TV is set to “Standard” format in the user menu.
If you have an oscilloscope, now is the time to use it.  Following are two procedures, one for setting the trimmer without an oscilloscope and the other for setting the trimmer with an oscilloscope.  I will point out that using an oscilloscope is much faster.  You simply look at the scope and you know when you have the trimmer set to its best value.  However, when I first performed this modification, I did not use my oscilloscope, but later confirmed that I had indeed set the trimmer almost perfectly by eye, without the scope.  So I don’t think it’s an essential tool for this modification to be successful.
Either way, please read both methods, as I have highlighted distinct conceptual points in each that can be applied to the general procedure.
Without a Scope:  toc

What needs to be done next is to adjust the capacitor until this overshoot is at its dimmest possible level.  Make sure the room is nearly dark.  Enter the xx59 convergence service menu. In this service mode, pressing “1” on the remote will enable the red CRT only, pressing “2” is green only, and pressing “3” is blue only.  Press “2” on the remote to mute the red and blue guns.  Use your nonconductive jeweler’s screw driver to turn the trimmer capacitor.  Observe how the band of overshoot along the left boarder of the 50 IRE gray window can be made much brighter or much dimmer as you turn the capacitor.  Adjust the capacitor until you reach a point where the overshoot does not get any dimmer and that you start to see “undershoot”. Undershoot is evident by a thicker bar approximately 7 mm  wide that will start to emanate from the left border of the window as it superimposes itself on top of the overshoot.  This undershoot is similar in appearance to the overshoot, only it is not as bright or well defined. It is also harder to detect.  Adjust the trimmer so that you just start to detect undershoot.
Once you are finished tuning the green trimmer, you will of course need to tune the trimmer capacitor for the red and blue guns in the same manner.
The problem with the 50 IRE window is that the undershoot starts to mask the overshoot as corrections are made. After the 50 IRE windows, it’s helpful to display a 30 IRE window for fine tuning.  At 50 IRE it’s easiest to see the undershoot develop, but it can be hard to see how much overshoot you have left. For this reason, I follow by looking at all three guns individually with the 30 IRE window because it highlights the overshoot more than the undershoot. Then it’s easy to see what’s left of the overshoot without having to compete with the undershoot. In the end, what seems to provide the best adjustment is to set red and green so that you start to get some undershoot in the 50 IRE window while leaving some overshoot behind in the 30 IRE window. Blue is different. If you set blue like this your high frequency multiburst patterns will take on a blue tone. I set the blue overshoot in the 30 IRE window so that I can just barely make it out… it’s almost gone. I don’t worry too much about the undershoot of the blue CRT in the 50 IRE window, I pretty much get rid of all the blue overshoot in the 30 IRE window.
One unfortunate thing, however, is that the capacitance in the trimmer is so minute that the screw driver you are using is actually adding capacitance to the trimmer whenever it touches it.  Thus when you remove the screw driver the overshoot will change.  You’ll probably need several iterations of adjustments each time you remove the screw driver to get it right.

Exit the service menu. The next step is to set sharpness in the user menu of the TV. Since you don’t have a scope, use a sharpness test pattern to set the ramped multiburst part of the pattern so all frequencies have “equal brightness” (as explained on the AVIA DVD). After the modification, expect that sharpness will need to be set MUCH higher. On a v23 the proper sharpness setting will be right around 47.
Now examine the multiburst test pattern. Do high frequency portions of the pattern have any color preference (does the multiburst seem red, green, or blue)? If so, that color’s trimmer capacitance is still set too high. Adjust the trimmer on the CRT in question until the multiburst pattern again appears to be the same “color” gray throughout the pattern. This method will locate the optimum trimmer setting and produce neutral gray multiburst patterns and good brightness until just under 12 MHz (labeled 5 MHz in 480i test pattern).

With an Oscilloscope:  toc
As stated, with a scope the trimmer cap adjustment is much easier.  Display a 50 IRE gray window.  Connect the ground of the oscilloscope to the grounding test point of the green PCB-CRT.  Connect the oscilloscope probe to test point six (TP6) of the green CRT.  Adjust the trimmer capacitor to best remove as much of the overshoot as possible.  You can use the small screw driver used to adjust probe compensation on your oscilloscope to adjust the GCL3000 trimmer capacitor.  Figures 4 and 5 are photos of the oscilloscope screen. Figure 4 shows the overshot circled in yellow while figure 5 shows proper trimmer capacitor adjustment.
Repeat the procedure for the red and blue guns.  The ground should always remain connected to the ground test point on the green CRT.
Figure 4 50 IRE window with overshoot                         Figure 5 50 IRE window without overshoot
Next, display a multiburst test pattern. Reconnect the oscilloscope probe to TP6 on the green gun.  Adjust sharpness in the TV user menu so the peak-to-peak voltage is equal for all frequencies on the green gun. Figure 6 shows properly adjusted sharpness (in this case on the v23 set to level 47).  Figure 7 is a photo of the oscilloscope screen displaying sharpness set too low.  In figure 7, sharpness is set at level 22 which was the appropriate setting before the modification.
Figure 6 Properly adjusted sharpness                             Figure 7 Sharpness set too low

Highest band 10MHz Highest band 10MHz
Now reconnect the probe to TP6 of the red PCB-CRT. Adjust the trimmer capacitor so that the multiburst pattern is displayed on the oscilloscope properly with equal peak-to-peak voltages for all frequencies. Try and get the red multiburst to look the same as the green multiburst as shown on the oscilloscope’s display. Finally reconnect the probe to the blue gun and adjust the blue trimmercap to properly display the multiburst on the oscilloscope. Look at the multiburst pattern on the TV screen. Its gray should match perfectly through out the entire test pattern. If there is color in high frequency portions of the multiburst pattern the trimmer capacitance is still set too high on the gun imposing the color. Adjust the trimmer on the CRT in question until the multiburst pattern again appears to be the same “color” gray as the rest of the gray in the test pattern.

Observations: toc

Figure 8 is a photo of the AVIA sharpness test pattern. This photo was taken after the PCB-CRT modification was completed and properly adjusted. Here, sharpness is set to level 47. As can be seen from the photo there still exits some faint ringing. However, at this point its effects are negligible. Before, with sharpness set at level 22 in the user menu the pattern actually displayed more ringing than it does now at level 47.
Figure 8 AVIA sharpness test pattern being displayed after modification and adjustment on 55613. Sharpness in user menu is set to level 47.

You’re done. You may feel the need to do some fine tweaking in the future but if you correctly followed the above procedure you should already be there. Put in a well mastered DVD such as one of the titles mentioned in the background portion of this paper. You should immediately notice unbelievable detail. The clarity will floor you and your mouth will drop. You will be able to find hidden details in the video images that were previously masked by the overshoot but are now apparent. Congratulations!

_________________________________________________________________________________

Appendix A toc

Follow Up Discussion and Simulated Q&A:

1) Q: Ever since I have performed this modification, at times my image is very “sharp” and has unbelievable clarity. Other times, especially when watching HDTV, such as the Tonight Show, the image seems soft and unfocused. Often one camera angle looks very well defined and another will seem a little blurry. Is this normal?
A: Yes, this is normal. What you have done is essentially removed all edge enhancements from your video chain. This will affect your perception of the image much the same way as when you first turned the sharpness control on your TV down. These adjustments will take a couple of weeks to get used to. Furthermore, with this modification you have essentially raised the detail level your display is capable of rendering to near reference quality standards. When programming is aired, often some error in focus will be acceptable to the producer. This is especially true for shows that are live or that are taped and aired the same day. Producers don’t intend for their viewers to be watching the program on such a high quality display. In effect they are counting on your TV to add EE.
2) Q: Should I perform this modification myself?

A: Only consider this mod if you have prior experience working with PCB repair or modification. Soldering a component to a circuit board is far more complicated than soldering two wires together. This type of work is like art. You need very skilled hands to know just how much heat, time, and solder to use. Don’t make your multi-thousand dollar TV your first experiment. It takes practice to learn how much heat is right and how much is too hot. Very quickly you can ruin the traces on a PCB with too much heat. It will happen before you know it. That being said, if you are experienced with PCB repair this job is about as easy as it gets. Just take off three components and replace them and the hardware work is done.

3) Q: I performed the modification and I am unsatisfied with the picture. What should I do?

A: First make sure you have given yourself enough time to get used to the new picture.  If you are still unsatisfied, you can do several things.  First, try enabling or turning up VSM, LTI, CTI or any combination of the above (I would suggest trying LTI at level one or two first and using VSM as a last resort).  You may find that you actually prefer some EE.  Your other option is to add some overshoot back into the picture by raising the capacitance of the trimmers.  With the GCL3000 trimmer capacitor you can actually add more overshoot to the picture than it originally had if you were so inclined.  If you’re unsatisfied with those two options reinstall the original capacitors back on the boards or purchase new capacitors with the same capacitance as the originals.
4) Q: OK, I have installed the capacitors on the PCB-CRT’s. However, when I adjust the capacitors, the overshoot does not get any thinner. In fact, no matter where I set the trimmer capacitor, the overshoot stays the same width.
A:  This is absolutely normal. The overshoot is not supposed to change width. As you adjust the trimmer capacitor, the overshoot will get brighter or dimmer with respect to capacitance. Furthermore, the undershoot will also respond in a similar manner with respect to capacitance. The overshoot and undershoot are inversely proportional. In other words, as the overshoot gets dimmer, the undershoot will get brighter. If you are calibrating the trimmer with an oscilloscope this is obvious. However, if you are making adjustments by eye, with the gray windows on the TV screen, this can be more difficult to realize. When adjusting by eye, the goal is to reduce the brightness of the overshoot as much as possible without making the undershoot too bright.
5) Q: When I try to use the multiburst or sharpness test patterns it doesn’t look right at all.  High frequencies are blurry and run together.  My multiburst may even look as though thicker bars are imposed underneath the thin lines of the high frequency portions.  When I hook up my oscilloscope to view the multiburst pattern all the high frequencies are squashed and no matter how I adjust the sharpness their peak-to-peak voltage hardly changes.

A: You have ruined your TV, go out and buy another one (just kidding hehehe…) Seriously though, you probably have your “FORMAT” selected in the user menu of the TV set to something other than “Standard”. Your TV format must be set to “Standard” and your DVD player or pattern generator must be putting out a “standard” signal to properly make adjustments in viewing the patterns.
Whenever the video signal is modified by the doubler, the signal gets “trashed”. In Figure 9, notice that in standard mode the frequency bars of the multiburst test pattern have clearly defined edges along their tops and bottoms. Now look at the right multiburst in Figure 9. Here the TV is set to “Wide Expand”. The jagged irregularities in the multiburst pattern are the result of setting the TV to a mode other than standard and are an effect of the doubler. This response represents a high level of distortion which makes it impossible to set sharpness or the trimmer cap.
Figure 9 1920x1080i Multiburst: Left photo “Standard”, Right photo “Wide Expand”
6) Q: Aside from the obvious risks of damaging components while working on them or electrocuting myself, are there any long term consequences from this modification that could over time damage my TV?

A: Honestly, this is still unknown. However, at this point I have performed the modification many times and several other technicians have also taken on this procedure. To date my own 55613 has been operational for one year with no problems. Also, no problems have been reported by any other technicians.  I don’t think that anything in this mod could strain any part of the TV or make anything wear out faster.  In fact, by lowering the capacitance of C1 and properly tuning the amp, there is even less load on the amp and its job is now easier.  One concern raised by a technician that may be valid is:
“I do have one concern though… Trimmer caps are notorious for becoming intermittent in poor environments. Having one on the “CRT dust magnet” could work against you in the long term… Am I worried about nothing?”
It may not even be a concern.  However, if problems did arise you could always retune the capacitors, or if necessary replace them.
7) Q: So why do Mitsubishi and other companies incorporate these EE into their sets?  I know you went over this in the beginning but it still seems strange.

A: We have been discussing this for quite some time. Here are some excerpts you may find interesting. Mind you, they are just educated guesses and may not be accurate.
Quoting myself:
There are many reasons that come to mind as to why Mitsubishi would indeed have intended to incorporate overshoot into the sets. For one, to the novice viewer, a set on a showroom floor will appear sharper, clearer, and more focused with this overshoot. I think the argument could easily be made that Mitsubishi wants this overshoot for similar reasons as to why they incorporate red push and EE into their sets. Also, when considering the fact that electronic and mechanical focus as well as convergence is never spot on out of the box (and on most sets will never be adjusted) this overshoot probably helps to “improve” the image by increasing the perceived definition of the picture. So in my opinion Mitsubishi does indeed want this overshoot.

Another possibility exists… regarding the predefined value for the capacitor (C1) in place of a trimmer cap. It is very possible that there exists enough deviation in the circuits that Mitsubishi is unable to choose a capacitor that provides the correct capacitance for every RC network on every TV…
What I mean in the second paragraph is that it may be impossible to install a capacitor that would provide correct tuning for the amp. In other words, there may exist enough deviation between the circuits on each TV that Mitsubishi is unable to find a capacitance that will work correctly on each individual TV. So rather that risking installing a capacitor with too little capacitance (undershoot and blurry picture) they always install one that they know will overshoot to ensure the picture looks “sharp”.
This is a quote from a Sony service bulletin posted by SonyCrusader referring to TVs using the same TDA6120Q Video output amplifiers:
Subject
Faint ghost along right side of image.

Cause
The ghost line is the result of the performance limit of the video amplifier ICs used on the C-boards. These particular ICs have extended video bandwidth. The extended bandwidth offers superior picture qualities not obtainable in other video amp ICs; the ringing or ghost image is the trade-off for having these other superior picture qualities.

Resolution
This is not a design or manufacturing defect, but rather a performance limitation trade-off. There should not be any attempts to eliminate this symptom by trying to repair the set; nothing can be done to eliminate it.

Most viewers do not perceive the symptom during typical TV viewing (movies, TV programs, etc.). The symptom is most noticeable when viewing test patterns included on adjustment DVD’s.
SonyCrusader then goes on to suggest:
It’s pretty easy to read between the lines, especially knowing the facts. The speculation that this preshoot / overshoot “ringing” is just used as another “sharpening” illusion seems likely.
Keep in mind that Mitsubishi is in business to make money, not to make TVs. Joe-Six-Pack is not as critical or picky as a videophile, who judges the image quality by a different criteria. Thus, if Mitsubishi wants to make the most money by selling the highest volume of televisions possible, they should from a financial perspective, adjust the televisions to have the best looking picture to the widest range of potential customers. Unfortunately for the videophile, this is not the same as calibrating the television to have the most accurate picture. I think it’s important to make the point that this is not “wrong” or bad policy. After all, Mitsubishi needs to sell as many TVs as possible. It would be unreasonable to expect them to tune their video amplifiers in such a way that would be less appealing to their largest customer base. After all, to remain competitive, Mitsubishi needs their RPTVs to look as good to Joe on the show room flour as competing models from other brands.
8) Q: O.K. I want this mod but I can’t do it myself. Who can I get to do it for me?

A: This modification is well outside the realm of what is included in any normal calibration. Not all calibrators will want to perform this mod. If you want to use a calibrator, you will need to find one who is also a field technician. Chuck Williams, Mr Bob, and myself (craigr) have all either performed this mod already or have expressed interest in trying this procedure out. Robert Vermazen may also be willing to perform this procedure. There are also probably other SPoT calibrators who can do this. If possible you may want to find out if one of them will be in your area. You may even be able to take a copy of this procedure to a television repair shop and hire a technician there to do it.
_________________________________________________________________________________
Appendix B toc

Application to Affected Sony Models:

As mentioned earlier this modification should be applicable to any television using the TDA6120Q video output amplifier. This includes many of the modern Sony chasses. At this point I don’t think anyone has attempted this mod on a Sony yet. However it has been a topic of much discussion amongst calibrators. SonyCrusader has done quite a bit of research in this area. If you are interested in trying this out on a Sony he would be the person to consult.
At present he has pointed out two obstacles. The first is that unlike the Mitsubishi which uses good old fashioned standard lead capacitors, Sony is currently using surface mount (SM) components. SM components require more skill to work with than standard lead components. Also, SM trimmer capacitors have less range, making it more difficult to procure a capacitor with enough variance to compensate the amplifier. Second, Sony also uses a more complex RC network on the blue PCB-CRT. Robert Vermazen has theorized that changing the same C1 capacitor would likely work just as well as on the “normal” RC networks. However, at this point this is unproven so buyer beware. Figure 9 is a photo taken by SonyCrusader showing the back of a Sony PCB-CRT. It should give you an idea of what you have to work with.

Figure 10 Part of Sony PCB-CRT

_________________________________________________________________________________

Appendix C toc

Sprague Goodman Trimmer Capacitor:
Sprague Goodman Part Number – GCL3000
Digi-Key Part Number – SG1036-ND
Data Sheet
Specifications:
Capacitance: 0.9-30 pF

Digi-Key Price $0.85 each
Figure 11 GCL3000 Trimmer Capacitor
_________________________________________________________________________________
Appendix D toc
Revision History
6/13/04

1) Added Appendix D

2) Added numbers to simulated Q&A

3) Added Q&A number 4

4) Added several comments regarding Sony TV’s throughout

5) Increased font size of “The Home Theater SPoT Thank You” on first page

6) Changed some wording throughout to help clarity
8/18/04

1) Changed some wording throughout to help clarity

2) Added description and Figure 9 in Q&A 5

3) Increased preceding figure numbers by one
9/4/04

1) Added link to craigr’s email address
10/19/04

1) removed reference to F9A0 due to “conflict of interests”

2) added information on cover page about craigr and CIR Engineering

3) linked Robert Vermazen directly to his business page

– Robert Vermazen’s profile on The Home Theater SPoT can still be found here

4) linked Mr Bob directly to his business page

– Mr Bob’s profile on The Home Theater SPoT can still be found here

5) capitalized “t” in The Home Theater SPoT throughout this document

6) changed my name in the Legal section of this document from my alias “craigr” to Craig Rounds

7) added hyperlinks in Appendix D
11/4/04

1) changed cover

2) fixed spacing in toc
11/7/04

1) updated article to reflect new developments including Q&A number 6
6/17/05

1) fixed broken links

2) updated contact citations

3) remastered Figure 1

4) eliminated “Print Friendly” MS Word download

5) fixed spelling and grammar to improve clarity