View previous topic :: View next topic |
Author |
Message |
mp20748
Joined: 12 Sep 2006 Posts: 5681 Location: Maryland
TV/Projector: 9500LC Ultra / Super 02 and 03 VIM
|
Link Posted: Mon Feb 23, 2015 2:32 pm Post subject: Marquee DIY Focus Board Mod |
|
|
For the absolute first time, I was able to make the focus better on the FGM module. In that past and for years I've tried everything, to seeing a very minor improvement, to for the post part changing out a lot of parts and not noticing anything at all. But the wonders of the scope has helped me identify and area that really made some significant changes on my setup.
The area that needed the attention was the addition of inductors on the main rails (as shown in picture). The inductors would need to be large or high current, of you could cause the focus to be worse. In the picture you can see the value of the main players on my board as well the part number. They are 27UH inductors that were inserted in series on the 15 volt and 24 volt rails. This is also very similar to what the Barco 909 does, by also using very large inductors right on the Focus Module. The G90 does the same, but there's is on the LVPS for that section.
Some of the things I immediately noticed after making the changes are:
- The famous convergence banding becomes obvious in the image on some scenes. Something I've only notice during some test patterns before this.
- The test pattern grid is super small, or at least the horizontal lines are. The vertical lines on my setup is showing the same size line as the horizontal, but it is within a larger line. I've also noticed my static stig is off, because the line within the line is showing to be more to the left. Not visible before.
- There is more mosquito noise on the screen.
- After making the changes, the convergence needs to be redone, and I'm sure the focus would also need adjusting, and I'll get to that later.
Please note that I've posted this as a DIY. Meaning it's not something I'm offering or have any intent on doing.
If doing this, I would leave the same stock caps in place, unless you'll want to upgrade the caps for the sake of it. I got this idea some time ago when thinking about making the focus better on the Marquee. Looking at the circuit and also looking at the Barco 909 and Sony G90 (both are just basic amplifier stages), they both had inductor filtered supply to their final amps in the focus stage. The scope allowed me to better understand why that stage would need to have the high frequency crap stripped and not going into the final stage. A single horizontal or vertical line will have a bloom to it if not stripped of noise. While a properly filtered supply produces a tight and more precise line.
|
|
Back to top |
|
|
Nashou66
Joined: 12 Jan 2007 Posts: 16171 Location: West Seneca NY
|
|
Back to top |
|
|
mp20748
Joined: 12 Sep 2006 Posts: 5681 Location: Maryland
TV/Projector: 9500LC Ultra / Super 02 and 03 VIM
|
Link Posted: Mon Feb 23, 2015 4:03 pm Post subject: |
|
|
Nashou66 wrote: | Yep did that in the Marquee thread .
Really helps and I went to 1000uf caps for the power rails and the focus amps.
Nashou |
And I've tried inductors in the same place, many times before and it's also in my Maintenance Thread. But this thread is suggesting something different, and that's why I said "The inductors would need to be large or high current, of you could cause the focus to be worse"
So the point in this thread is that you would need to use large or high current and of the right value, or it won't do anything or make things worse.
Look at a large ones on the 909
|
|
Back to top |
|
|
Nashou66
Joined: 12 Jan 2007 Posts: 16171 Location: West Seneca NY
|
Link Posted: Mon Feb 23, 2015 4:25 pm Post subject: |
|
|
Yep mine are 27uh as well. At first Tried the lower current 270uh but I saw no difference. Then I tried a pin wheel type( looks like a resistor) and it blew the Mosfet that switches the main cap in or out for the higher band. I never got that working again.
And it was by accident that I ended up with the 27uH as I thought I ordered the 270uH in the higher current but I ordered the wrong part. I used it anyhow and thats when I saw the improvement. I also tried a 22uH and that works well too.
Athanasios
_________________ Don't blame your underwear for your crooked ass~ unknown Greek philosopher
"Republicans believe every day is the Fourth of July, but the Democrats believe every day is April 15." --- President Reagan
One Smart Dog!!!
Marquee High Performance Bellows now shipping!!
Marquee Modifications and Performance Enhancement
Marquee C-element and Bellow removal
|
|
Back to top |
|
|
mp20748
Joined: 12 Sep 2006 Posts: 5681 Location: Maryland
TV/Projector: 9500LC Ultra / Super 02 and 03 VIM
|
Link Posted: Mon Feb 23, 2015 4:53 pm Post subject: |
|
|
Put a scope probe right on the rail at the final, and look at what happens when you change them around.
Inductors are tricky, especially when trying to use in a power output stage. If the value is not right, it will introduce noise.
So far, I'm thinking along the lines that for best focus, you'll need a low noise amplifier stage after the DAC that controls it. So when scoping the final, you find the same HF noise from the LVPS being in the output to the focus coils. I no longer think the magic of focusing is in the electronic circuit. I'm thinking now it's the focus coil itself, but it would need a clean pulse in order to perform best. In fact, I'm now able to prove that.
And whenever you use an inductor in the supply to an low frequency amplifier, it has to be high current.
|
|
Back to top |
|
|
Nashou66
Joined: 12 Jan 2007 Posts: 16171 Location: West Seneca NY
|
Link Posted: Mon Feb 23, 2015 5:34 pm Post subject: |
|
|
The focus coils, astig coils and basically all the coils will all pick up noise. That Tube is an analog mess. That is why the Snap on
Ferrite chokes did so much on the astig wires right near the amp. It is all fed back to the circuits. Some circuits like the CVA use that feedback as part of the signal but use filters to ground to clean it up. At least thats how i see it on the CVA. One thing you can do that might help is add more Aquadag ground wires from the tube to the VNB's extra tabs. It will help remove more of that static mess to GND. I noticed on Barcos and I think the Ampro I have here they have a separate GND from the actual housing of the focus coil to GND. The Marquee only has it in the harness, and its not very large.
Athanasios
_________________ Don't blame your underwear for your crooked ass~ unknown Greek philosopher
"Republicans believe every day is the Fourth of July, but the Democrats believe every day is April 15." --- President Reagan
One Smart Dog!!!
Marquee High Performance Bellows now shipping!!
Marquee Modifications and Performance Enhancement
Marquee C-element and Bellow removal
|
|
Back to top |
|
|
Nashou66
Joined: 12 Jan 2007 Posts: 16171 Location: West Seneca NY
|
Link Posted: Mon Feb 23, 2015 10:23 pm Post subject: |
|
|
Another thing you can do is replace Resistors R163- 166, R263-266, R363-366 , 2 ohm 1/8 watt resistor are used and on some boards if you look underneath you will see some discoloration from heat. I would use same value but larger wattage and ones that have a good ppm value as not to drift with heat. Also lift them up off the PCB for airflow cooling.
Also change caps 115,215,315 to film caps from the mica I think they use there on older boards, the new boards may use film now.
After this is all done you'll have to let it warm up and then go to the RGB focus menu and adjust focus.
Another little tip, if you want a sharp image but a slight loss in resolvable resolution( I now it sounds contradictory) adjust focus so there is no flaring from out under the grid line. You'll notice you won't see the pixels that draw the vertical lines after but the flare will go away. From your seating position it will look sharper, the flare adds a layer of haze to the image from a distance and the actual resolution you loose is not noticeable.
Athanasios
_________________ Don't blame your underwear for your crooked ass~ unknown Greek philosopher
"Republicans believe every day is the Fourth of July, but the Democrats believe every day is April 15." --- President Reagan
One Smart Dog!!!
Marquee High Performance Bellows now shipping!!
Marquee Modifications and Performance Enhancement
Marquee C-element and Bellow removal
|
|
Back to top |
|
|
mp20748
Joined: 12 Sep 2006 Posts: 5681 Location: Maryland
TV/Projector: 9500LC Ultra / Super 02 and 03 VIM
|
Link Posted: Tue Feb 24, 2015 1:19 pm Post subject: |
|
|
I would be very careful in changing any of the other components. I've been trying a lot of things over the years to finally getting a better understanding why the focus could not be better, with the answer being on the better focusing G90 and Barco 909. They both have better focus coils, to include, they both have better filtered supply rails for the final amps.
Now and again, those inductors would need to be HIGH Current, or you'll create a noise problem. And by high current (see link below) they would be very large in size, and would have a DC impedance that's near zero (using the average meter).
You'll have to look at those focus coil driving stages much like and Power Amplifier in audio. It has to have near zero ohms on the power supply rails to the final stage. So by going with high current inductors, the final stage would see it as a piece of wire (zero ohm), and NOT create a problem (distortions) having it there. But having it there as I said before is to remove or strip the HF noise from getting into the focus yokes. This is the problem on the Marquee, because both G90 and 909 are both using basic drive amplifiers to drive the coils. The CLM on the Marquee is where the pulse is generated, and there is a DAC on the FGM or Focus board that manipulates the focus range for the E-Static Focus.
Now for the Zones, it too needs the inductors (as seen in my shot). And much of the same applies for the Zones as does for the Static.
Now there is another area that needs attention, and according to the scope, it is not on the FGM or Focus board. I'll run it down later and post here. It deals directly with the condition of the pulse... later on this one
If you're running at the lower scan rates you may not notice a difference, but for full 1080P and above, it is a substantial improvement, but let me also say I've made a few changes on the FGM (some seen in the shot) that helps with the problem from the CLM. I'll also get back on that as well.
http://www.ebay.com/itm/27-uH-Inductor-NOS-New-Old-Stock-QTY-5-ea-H3-/190854205390?pt=LH_DefaultDomain_0&hash=item2c6fcc0fce
http://www.ebay.com/itm/Vishay-Dale-IH-15-27UH-High-Current-Inductor-/170160393621?pt=LH_DefaultDomain_0&hash=item279e598d95
|
|
Back to top |
|
|
dvh99
Joined: 25 Dec 2009 Posts: 2158 Location: nederland
|
|
Back to top |
|
|
mp20748
Joined: 12 Sep 2006 Posts: 5681 Location: Maryland
TV/Projector: 9500LC Ultra / Super 02 and 03 VIM
|
Link Posted: Tue Feb 24, 2015 2:01 pm Post subject: |
|
|
dvh99 wrote: | http://nl.farnell.com/bourns/2306-v-rc/inductor-27uh-15-6a-15-radial/dp/1929758
would these 2 do on both locations mike?
they are new and cheaper then the ebay products.
is +- 15A really necessary? |
They look to be a higher ohm rating, based on the amount of windings and size of the wire used.
The reason for the higher Amp rating, is where you'll get the lower ohm rating you need for the amplifiers.
|
|
Back to top |
|
|
Nashou66
Joined: 12 Jan 2007 Posts: 16171 Location: West Seneca NY
|
Link Posted: Tue Feb 24, 2015 2:15 pm Post subject: |
|
|
Mike the resistors I said to replace are part of the grounding for the Feedback. They get really really hot and that heat usually increases resistance. From recent diagnostics of my CVA which has the exact same resistor array to ground for the feedback . I found mine on the CVA did not measure the 2 ohms they should have been but into the kohms. So I think having a more heat resistant resistor there will keep the focus value you choose stable. Or in other words it wont drift over time from the resistors becoming heat damaged. I noticed that over some time I need to re adjust the focus. I then looked at the board and saw the darker traces under the board. Used a laser heat meter and sure enough those 4 resistors and the trace where 10c higher in temp than the surrounding board.
Also the trace that goes to some of the amps on the heat sink are also discolored. Changing those to more heat resistant ones should also help in drifting focus but I have not done that yet.
Athanasios
_________________ Don't blame your underwear for your crooked ass~ unknown Greek philosopher
"Republicans believe every day is the Fourth of July, but the Democrats believe every day is April 15." --- President Reagan
One Smart Dog!!!
Marquee High Performance Bellows now shipping!!
Marquee Modifications and Performance Enhancement
Marquee C-element and Bellow removal
|
|
Back to top |
|
|
Nashou66
Joined: 12 Jan 2007 Posts: 16171 Location: West Seneca NY
|
|
Back to top |
|
|
CSX
Joined: 11 Feb 2015 Posts: 142 Location: Ohio
|
Link Posted: Tue Feb 24, 2015 4:59 pm Post subject: |
|
|
Would providing air or other forms of cooling to the board provide more stability in conjunction with the addition of large inductors?
|
|
Back to top |
|
|
mp20748
Joined: 12 Sep 2006 Posts: 5681 Location: Maryland
TV/Projector: 9500LC Ultra / Super 02 and 03 VIM
|
Link Posted: Wed Feb 25, 2015 12:12 pm Post subject: |
|
|
CSX wrote: | Would providing air or other forms of cooling to the board provide more stability in conjunction with the addition of large inductors? |
Possibly, but a more stable temperature environment is better, than a fan cooled heatsink, because you don't want an open door in the room causing a different blown on temperature on the heat sinks. That's the main reason the VERT, STIG, CONVERGENCE boards are mounted on the rear for ambient cooling opposed to blown on air.
I need to also post the best procedure for focusing the higher resolution (1080P) images. You will need a point and shoot camera (hopefully with a manual setting) and any of the later version SONY Blu Ray disk
|
|
Back to top |
|
|
Nashou66
Joined: 12 Jan 2007 Posts: 16171 Location: West Seneca NY
|
|
Back to top |
|
|
cmjohnson
Joined: 03 Apr 2006 Posts: 5180 Location: Buried under G90s
|
Link Posted: Mon Mar 02, 2015 3:03 am Post subject: |
|
|
Though I don't remember the values I calculated, I did figure out that the focus system could be optimized for a specific resolution (and framerate) which would in most cases be 1080P-60. As the focus system works as a resonant circuit, and with the inductive values of the focus coils being known, you can calculate the ideal capacitance values for the resonant amplifiers. As I recall, the stock focus coil values are 30 uH for the static coil and 5 uH for the dynamic coil, with the dynamic coil value being the one of interest.
So the question is then phrased as this simple electronics question: In an LC circuit where L is 5 uH and the resonant frequency is (insert value here for the scan frequency of a 1080p-60 signal , or another value if you prefer), what is the value of C?
I believe that optimizing part values for your chosen scan rate can only help the focus system to operate more efficiently and run cooler. Cooler running circuits should exhibit less drift due to the reduced temperature shift during warm-up.
|
|
Back to top |
|
|
barclay66
Joined: 27 Jun 2011 Posts: 1291 Location: Germany
TV/Projector: Marquee 9500 Ultra
|
Link Posted: Mon Mar 02, 2015 7:59 am Post subject: |
|
|
Hi,
With 5µH for the coil and a Horizontal scan rate of 67,43KHz (for 1080p60) the ideal capacitor for a self resonant L/C circuit would be 1.11 µF.
The next step would be determining which of the switched capacitors (C118-120, C218-220, C318-320) is used when displaying 1080p. It must be one of the three that are connected using the signals RES_0, RES_2 or RES_4 and the three corresponding MOSFET switchers for each color channel.
If the value of the stock capacitor is much different to the calculated value then there could be additional parts that need to be taken into account for the calculation.
The current values are:
RES_0 -> 22nF
RES_2 -> 68nF
RES_4 -> 2.2µF
Regards,
barclay66
|
|
Back to top |
|
|
Nashou66
Joined: 12 Jan 2007 Posts: 16171 Location: West Seneca NY
|
|
Back to top |
|
|
Nashou66
Joined: 12 Jan 2007 Posts: 16171 Location: West Seneca NY
|
Link Posted: Mon Mar 02, 2015 1:53 pm Post subject: |
|
|
Found it.
Not in our threads but discussions we had in them led Scott to make this thread.
http://www.curtpalme.com/forum/viewtopic.php?t=15724&highlight=cut+voltage
tse wrote: | The Marquee uses a resonant configuration in the output amplifier to lower the power dissipation. There is actually a patent dealing with this. The resonant circuit reduces the voltage needed by the amplifier, when compared to a linear amplifier, for a given amount of output. The amplifier switches capacitors in or out depending on the frequency which means one cap value is used for a range of frequencies. Certainly one specific frequency would have an optimum value of cap. If that value were known it might allow an increase in output current without increasing the power supply voltages used by the output amplifier.
Attached is a schematic showing the difference in output voltage needed by a linear amp compared to the resonant circuit amp. Both output currents are the same.
Scott |
tse wrote: | A range of output voltages with caps from 85 to 100nF with horizontal sweep frequency of 68kHz. The minimum peak voltage looks like 95nF (red trace) is lowest. This would allow a little more gain in the amplifier before clipping happens. Spice is cool. Of course the actual world will be a little different.
Scott |
Nashou
_________________ Don't blame your underwear for your crooked ass~ unknown Greek philosopher
"Republicans believe every day is the Fourth of July, but the Democrats believe every day is April 15." --- President Reagan
One Smart Dog!!!
Marquee High Performance Bellows now shipping!!
Marquee Modifications and Performance Enhancement
Marquee C-element and Bellow removal
|
|
Back to top |
|
|
Nashou66
Joined: 12 Jan 2007 Posts: 16171 Location: West Seneca NY
|
|
Back to top |
|
|
|
|