Marquee Modifications and Performance Enhancement 3 - AVS Forum
Forum Jump: 
Closed Thread
 
Thread Tools
post #1 of 465 Old 06-13-2004, 07:14 PM - Thread Starter
Senior Member
 
Jaehong Lee's Avatar
 
Join Date: Mar 2000
Location: Seoul , Korea
Posts: 345
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
I changed carbon resistor in neck board to ceramic resistor as shown here

http://www.kavforum.com/Community/cl...e=&SearchName=

2*33ohm, 10k, 100K for each neck blard. I used Ohmmite OX type.

Results is better than expected.

razor sharp focusing.

virtually no background noise in Picture. very transparent and vivid color.

better contrast. punchier picture.


I stongly recommend change carbon resistor to ceramic resistor in HVPS too.
Jaehong Lee is offline  
Sponsored Links
Advertisement
 
post #2 of 465 Old 06-13-2004, 10:04 PM
Senior Member
 
mcpherv's Avatar
 
Join Date: Aug 2003
Location: London, Ontario, Canada
Posts: 498
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
(note this post was edited)

DISCLAIMER

Guys, just a warning here. If you want to try to apply any of the ideas contained within this thread, you do so at your own (and your projector's) risk. Also, this is all a work in progress, and therefore, this thread will not GUARANTEE you significant performance increases (other than the priorly established and tested performance increases). Also, please do not expect to obtain the level of performance that the professionals have been able to obtain up until this point.

DISCLAIMER


Table of Contents:

POST 1 (this one):

Preface and relevant links.

GENERALLY APPLICABLE CONCEPTS, mostly capacitor modifications at the present time.

POST 2:

A bit of background information, and VIM MODS.

POST 3:

Modifications to the Focus board, followed by the Control Board, followed by the Necks boards.



PREFACE

Well guys, here is the new thread regarding Marquee modifications. The main scope of this thread is going to be the further electronic improvements possible in the Marquee chassis. As is evidenced by the testimonials of many members on this forum, performance improvements via electronic changes to the Marquee can improve RBG performance in a huge way. To date, we have had individuals share many ideas related to CRT performance enhancements, including carbon component resistor changes, HVPS renewal, colour filtering addition, VIM to Neck board cable swapping (and the associated connectors), and several general maintenance issues. However, as good as all of this information is, there is little information regarding the specifics as to the further modifications necessary to transform a Marquee into the video powerhouse several individuals have displayed it has the capability to be.

IMPORTANT
With that said, I wish to state that, since several individuals who read and contribute to the forum are involved in the modification of the Marquee for a living, we will take a disapproving stance towards direct requests for proprietary information. Let us please adopt the policy that these individuals will share what they wish to. Please respect this. Note that some inquires into whether or not information can be shared will be tolerated, but respect the answer the individual provides in response.

I wish to extend a HUGE Thanks to all of those who have shared their experience and knowledge in presenting the existing modifications and for displaying the fact that further improvement is indeed possible.

I want to take some space and talk about some of the work that has been done in the past on the forum regarding modifications. Note that this is not a complete list, and if you have an idea as to a thread that deserves a special mention here, please PM me and I will add it in.

The most comprehensive thread to date on Marquee modifications is the infamous Marquee Maintenance thread, started by Mike Parker:
http://www.avsforum.com/avs-vb/showt...ee+maintenance
Here you will find every update your Marquee needs to bring it up to the quality and stability of a new machine, including the performance enhancements the Electrohome engineers made to the Chassis later in production.

Possible Explanation for the seemingly excessive bandwidth requirements for video here, with a calculator to calculate the adjusted bandwidth needs (see white paper link to the side – note that I just realized where some of my rationalization for the component requirements originated from – this is very well written and is very pertinent to the reason higher bandwidth signal pathways are desirable in the Marquee):
http://www.autopatch.com/dlrportal/bandwidth.html


Jaehong Lee’s excellent Posts covering the relative degrees of performance enhancement in upgrading the carbon component resistors in the HVPS and on the Neck Boards, among other things. Well worth the read:
http://www.avsforum.com/avs-vb/showt...hreadid=401345
(2nd post is the post above this one in this thread)

GREAT thread on colour filtering crts, by adding coloured ink into the glycol (Thanks to B.Hegelstad for the great writeup, and dav99 for the idea to use coloured ink):
http://www.avsforum.com/avs-vb/showt...0&pagenumber=1


(note that the above thread appears to present a better way of doing this)Great thread on colour filtering the crts. Thanks go to Energeezer for starting this one, and to KennyG for pioneering the method of filtering discussed here:
http://archive2.avsforum.com/avs-vb/...r&pagenumber=1

Marquee (Sound) Noise Reduction Thread:
http://www.avsforum.com/avs-vb/showt...hreadid=413501


Switchmode Power Supply Cap Swap-out (Lots of Marquee cap mod information here provided by Ken Hotte / KBK and supported by MP, Mark, and others)
http://archive.avsforum.com/avs-vb/s...threadid=91060

Jay Wilson’s First Post :D Topic is the ECP4000, but there is LOTS of good GENERALLY APPLICABLE IDEAS regarding capacitor information here:
http://archive.avsforum.com/avs-vb/s...&threadid=8272

Here is another general cap mod thread. It is one of the first times Ken mentions the usage of os-cons.
http://archive2.avsforum.com/avs-vb/...p&pagenumber=1

Barco 800 Cap Mods; good general capacitor thread:
http://archive2.avsforum.com/avs-vb/...0&pagenumber=1

Thread on relays (Reinhard posted here):
http://www.avsforum.com/avs-vb/showt...hreadid=394358





GENERALLY APPLICABLE CONCEPTS, mostly capacitor modifications at the present time.



Note: Please keep the discussion in this thread directly involved with the Marquee. I am posting this section to aid individuals in modifying all projectors (in general, and for the specific purpose of aiding in Marquee modifications despite the fact that I present generally applicable concepts). Please start a new thread if you wish to discuss other projectors.

Before we go further, I would like to discuss the theories presented in several important threads involved in laying the groundwork for both General Marquee Modifications and my own research into projector modifications in general. Please note that I have applied the capacitor concepts in these threads to my Sony 1031 with good results. I began researching capacitor modifications in December 2003, while most of the other research I have done began at the end of April 2004, only roughly 2 months before I created this thread. I have taken basic level physics (up to 1st year university), and therefore understand basic circuit design and function. I have also taken calculus up to 1st year university, and generally understand basic wave representations. Therefore, my suggestions are mostly the result of reading datasheets for the original components and attempting to find updated/graded, directly interchangeable parts. Much of my information regarding opamps was obtained via reading the www.diyaudio.com forums.
I would now like to discuss cap modifications in general terms, as I understand that many people looking here might not own a Marquee, and might be looking for suggestions as to how to improve the performance of their current projector.
Firstly, read the general capacitor threads I have listed above, several times, or until you have a good feeling about your knowledge level. Also, please don’t undertake these modifications on a large scale on your own unless you are prepared to do some trouble-shooting. I know a lot about failure modes on the 1031 in part because of this.
Secondly, find the schematics for your projector. This is important, as it will help you to find capacitors that might warrant replacing. Then, start by finding the schematics for the input board for your projector, and identify the entrances of the video signal you are using. Try and trace these lines, identifying the capacitors that “couple†the video lines (for newcomers, coupling means that these capacitors are connected on both sides to the signal line, not to ground or anything else). All the way to the crts (this will be across a couple of boards, finishing at the neckboards). Note the capacitance values of the capacitors, and replace them using good quality electrolytics to replace electrolytics (typically these types represent the higher value capacitors – note whether the originals are polar or non-polar and replace appropriately – polar will have + and – designations at the connection points) and film capacitors for the lower value capacitors of the same capacitance value and the same or higher voltage value. Make sure the replacements are 105 degree caps (find replacements at www.digikey.com and www.newark.com ). Last thing here – identify carbon composition (look at the list of component section of the manual and reference the appropriate number on the appropriate board) signal coupling resistors. Replace with metal film capacitors in all areas other than neck boards.
Thirdly, look for IC s that the signal enters along the path you traced looking for coupling caps. Then identify the decoupling (capacitors for power buffering/filtering – these go from the power line to ground) capacitors directly next to the IC, on BOTH power rails (identify the pins that are the power rails by checking which pins are labelled V+ and V- on the schematics). For those of you wondering, the positive sides of polar capacitors connect to the non-ground side of the connection (ie. The power rail) for a positive voltage rail, and vice versa. Here is the section from my VIM post covering these types of capacitors:

Based on some of the follow-up posts, I decided to add a section on power supply decoupling. Extrapolating on Bill’s comments (and remembering that all the datasheets I’ve seen have used multiple caps on the rails of their components while testing them), I will suggest that:
Firstly on the RGB signal opamps (and then anywhere else you feel like doing it on a power rail decoupling capacitor) lets add (note that changing the suggested values will affect performance, but this general guide will give you a basis, and is a good start):
1 x 100PF cap, mica type – these are used in military and medical electronics, and retain their capacitance value past the Ghz frequency point (at least according to the spec sheets on the SM types at www.digikey.com)
1 x 100NF cap (or you can just leave the original there, although, this isn’t the best thing to do from a performance standpoint, due to the relatively small effect this the cap additions are reported to have, for sake of time, unless someone really takes issue with this, just leave the original there)
1 x High value (~6uf – 100uf) range tantalum or oscon cap (oscons are at www.newark.com, as are tantalums).

KEEP ALL LEADS AS SHORT AS POSSIBLE for maximum effect. For those reading this and wondering which way to connect the polarized, larger capacitors, connect the same polarity to the non ground side (voltage rail side) as the type of rail it is (ie. +ve rail, positive side of the polarized cap connected here). The other side should connect to the other side of the existing cap, which is ground (decoupling caps are caps from a line of some sort – power, signal, etc, that connects on the other side to ground – they are used to filter out noise, which is the deviation from the desired voltage value – higher value caps filter lower frequencies, and vice versa). The non-polarized types do not need to be compensated for as far as orientation is concerned.

Lastly, for the really ambitious, replace all of the electrolytic capacitors in the chassis with equal value, but 105 degree capacitors. Be sure to replace non-polars with non-polars, and polars with polars. Start with the video input board and neckboards, and then go to the focus controls and line/image placement circuitry.

I hope this helps everyone reading this. PM me if you need clarification on anything in this section.


Changes
- Removed rail to rail cap suggestion


http://reviews-zdnet.com.com/Labs/45...5021089-1.html

http://www.reed-electronics.com/ednm...ndustryid=2853

http://www.johansontechnology.com/te.../qne/index.php

http://members.aol.com/sbench102/caps.html

http://www.paktron.com/techarticles/...igh_speed.html

Capacitor fundamentals
http://www.murata.com/emc/knowhow/pd...-1/12to16e.pdf
mcpherv is offline  
post #3 of 465 Old 06-14-2004, 01:40 AM
Member
 
gerni g.'s Avatar
 
Join Date: Mar 2002
Location: Germany
Posts: 94
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
@jaehong,

looks like the two big e-caps are black gates, right?

is there a translation of that site?

best regards,

gernot

-crt or die!_
gerni g. is offline  
post #4 of 465 Old 06-14-2004, 02:54 AM
AVS Special Member
 
bblue's Avatar
 
Join Date: Nov 2001
Location: El Cajon, CA
Posts: 1,051
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Gernot,
to me those two caps look like the originals. They should be replaced with the 33mfd 160v 105 degree caps. Audio based performance caps will do no good there.

Jaehong,
I have doubts that ceramic resistors are a very good thing on that neckboard. First off they're a 10% tolerance, but more importantly they drift severely with heat. That is a very high temperature area.

I think I'd recommend a good 5% 1/2 watt carbon film for all four of those resistors. They're a lot more stable.

I noticed the ceramic drifting when I was testing the OX/OY resistors which I have installed in a couple of HVPS's. The area where the resistors are located doesn't have a lot of heat fluctuation so they appear to be ok.

Some of the newer HVPS's don't use carbon comp resistors at all, and in the place of the 560k and 1k resistors they use these rather long and skinny type of resistor which almost look like low wattage wirewound. They're something with caps on the end and a ceramic body, but I can't tell without ruining one whether they are wirewound underneath or some sort of a scanned film on the form. They're definitely not ceramic composition.

--Bill
bblue is offline  
post #5 of 465 Old 06-14-2004, 09:51 PM - Thread Starter
Senior Member
 
Jaehong Lee's Avatar
 
Join Date: Mar 2000
Location: Seoul , Korea
Posts: 345
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Bill you are right.

but carbon film produce aweful noise to neighboring circuits.

Buy 4- 5 times more ceramic resistor , select closest value one .

It surely clear up and enhance PQ.
Jaehong Lee is offline  
post #6 of 465 Old 06-15-2004, 01:00 AM
AVS Special Member
 
mp20748's Avatar
 
Join Date: Nov 2000
Location: Maryland
Posts: 6,339
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 13
Quote:
Originally posted by Jaehong Lee
Bill you are right.

but carbon film produce aweful noise to neighboring circuits.


I've never heard that one. You might be thinking of carbon composite resistors, which are very noisy and temperature sensitive. Carbon film is by far a much better resistor, it is not considered or known to be a noisy resistor. Ceramic is better, but it depends on the application. Carbon Film is also known as "flame Proof"...

When I mod the neck boards I've replaced the 33 ohm carbon composite resistors to 33 ohm carbon film. The two 33 ohm resistors on the neck board are the coupling for the video feed to the CRT. The 10K, 100k and 22meg (flameproof) that are in the G2 supply, I have not yet changed. Based on what you have been posting on this, I'll put the 100k and 10k on the list as well. There's no need to replace the 22meg, because it is flameproof, and that is the same as carbon film. From my understanding, carbon film is the economical low noise replacement for carbon composite. Cabon composite was also being replaced with carbon film and a special ceramic. Ceramic is supposed to be better, but that may not be the case for every circuit type. Myself I'm changing out the other two resistors (10k, 100k) on the neck board to carbon film as well. I'm with Bill here, I would not put ceramics on the neck board, especially since I know that carbon film will pretty much offer the same upgrade (noise) as the ceramics. The concern here is the circuit that it is being used in, though it should be safe to use the ceramics, we're not sure of how well they'll hold up to the high temperature that can be present on the neck boards, and how they will effect the color calibration, if they are heat sensitive. Carbon film are both low noise and stable in that hot environment. This is something that we should look into further. Also, it is very important to substitute resistors watt for watt. Especially for the circuits that we have discussed, any change in wattage of any of the carbon resistors could change the resistance load factor for both power supply G2 and neck boards.

I think you might have discovered something here with the replacement of these resistors. I did not think to go further and replace the 10k and 100k myself. I've wondered about the G2 supply, but never got around to it.

Anyway, respect the wattage first. And Bill is right again here (well, I agree on this myself). The use of special audio caps in these video circuits offer no benefit at all. the replacements simply MUST be high temps (105c).

Great discovery, but there's plenty room on that board for even better performance...

It's all about the performance... Got Marquee!

 

High Performance Marquee Video chain modifications.  Now available!

mp20748 is offline  
post #7 of 465 Old 06-15-2004, 01:19 PM
AVS Special Member
 
bblue's Avatar
 
Join Date: Nov 2001
Location: El Cajon, CA
Posts: 1,051
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
any change in wattage of any of the carbon resistors could change the resistance load factor for both power supply G2 and neck boards.
Mike,
Could you expand on this a bit? I've always understood that upward changes of wattage were acceptable at any time (subject to physical constraints).

--Bill
bblue is offline  
post #8 of 465 Old 06-15-2004, 03:24 PM
AVS Special Member
 
Tim in Phoenix's Avatar
 
Join Date: Mar 2000
Location: phoenix
Posts: 5,064
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 12
Say Bill

There seem to be a few resistors on every board that are designated as Safety items that are meant to act as fuses too, you want to refer to the Marquee schematics and not go to higher wattage or other types for those.

--------------------
Tim at E-Tech ooo ehometech@earthlink.net ......your Marquee Pro Shop!
Tim in Phoenix is offline  
post #9 of 465 Old 06-15-2004, 05:24 PM
AVS Special Member
 
bblue's Avatar
 
Join Date: Nov 2001
Location: El Cajon, CA
Posts: 1,051
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Thanks Tim.
Yes, I'm quite aware of those, but these particular resistors are not in that category and neither are the G2 resistors in the HVPS. Mike's comment seemed to be general in nature, and I've never heard of it before.

--Bill
bblue is offline  
post #10 of 465 Old 06-15-2004, 05:58 PM - Thread Starter
Senior Member
 
Jaehong Lee's Avatar
 
Join Date: Mar 2000
Location: Seoul , Korea
Posts: 345
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Thanks mike.

I also have a lot of experience of designning and bulding tube ausio amp and D/A, A/D converter for my hobby live recording.

Welcome to visit http://home.mic.go.kr/~jhlee/audio.htm

From my experience, carbion film is also not so desirable.
Anyway those who modified resistors in HVPS and neck board in here reported that significant improvement in PQ, especially cleaning up background noise.

Please test yourself. I am sure you will like .

BTW thanks again for your wonderful article about Marquee 9500.

I followed your instruction and it helped a lot . ^^
Jaehong Lee is offline  
post #11 of 465 Old 06-15-2004, 10:57 PM
AVS Special Member
 
mp20748's Avatar
 
Join Date: Nov 2000
Location: Maryland
Posts: 6,339
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 13
Quote:
Originally posted by Jaehong Lee
Thanks mike.

Please test yourself. I am sure you will like .

BTW thanks again for your wonderful article about Marquee 9500.

I followed your instruction and it helped a lot . ^^
Yes, it's on my to do list. This mod makes perfect sense, it's just that I'll need to do more research on the resistors. From my last reading, carbon film was highly recommended over carbon composite. Well, anything would be highly recommended over carbon composite. It has crossed my mind several times to look into the G2 supply for noise, but it too far down on my list of things to do next. Now it's on the top of the list.

I'm glad that you've got something from the "maintenance" thread. It was really fun starting it and participating in making it grow. But there are some others who really put their weight in on it as well, and for that we owe a big thanks to Sly, bblue, Cary and some others, who have been fantastic with the thread. We're going to resurrect it sometime this summer, I've already sent PM's to the rest of the geeks to get ready (this includes you to). It's time we take the Marquee to the next level, and by then hopefully I'll be able to announce the HVPS rebuild. For now I'm looking forward to a vacation, I need to get away from these doctors and attorneys. I'm really exausted, and looking for a trip to New York next month, but when I get back, we're gonna get back on track.

I'm excited about the mods to the Marquee, and you have not seen anything until you've experience the VIM/Neck board mods that I do. I'll have to mod a set for you when I get time. But the thing I'm more excited about is the MP-5 transcoder. That I could no longer put on hold. I had to find time to move it along. Now if the final quote would only get here on time, I'm ready to introduce the MP-5. It will probably be the only transcoder made with correct gamma.

Bill,
the G2 supply has these resistors in series. and the ones before the G2 itself is a 100K to a 10k, with a 22meg going to ground. Based on the value of the 22meg, it must be a discharge resistor. But be changing wattage of a resistor in a series of resistors, should change the circuits load impedance. Maybe I'm wrong, but from experience, I've had bad luck with using too large a resistor in high voltage series circuits.

It's all about the performance... Got Marquee!

 

High Performance Marquee Video chain modifications.  Now available!

mp20748 is offline  
post #12 of 465 Old 06-16-2004, 09:24 AM
Senior Member
 
mcpherv's Avatar
 
Join Date: Aug 2003
Location: London, Ontario, Canada
Posts: 498
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
DISCLAIMER

Guys, just a warning here. If you want to try to apply any of the ideas contained within this thread, you do so at your own (and your projector's) risk. Also, this is all a work in progress, and therefore, this thread will not GUARANTEE you significant performance increases (other than the priorly established and tested performance increases). Also, please do not expect to obtain the level of performance that the professionals have been able to obtain up until this point.

DISCLAIMER




POST 2:

A bit of background information, and VIM MODS.



Note: There is a list of changes made during each revision to this post at the end.


Everyone,
I've been looking at the Marquee schematics for the last month in preparation of hopefully getting one of these projectors at the end of the summer. Therefore, I have some ideas for mods that I havn't yet been able to try myself, but that I would like to share for anyone that wishes to try them. Please note that I am not a tech nor an EE, but I have spent a lot of time reading KBK's, Chris Steven's, and Mike's posts in the present forum (ie. the maintenance thread - thank you to all who have shared their expertise) and all the ECP modding threads located in the archives. I have also done some research on the diyaudio forums on opamp replacement. I have also spent countless hours over the last few months looking at datasheets for opamps, analog switches, capacitors and resistors, and have a decent grasp on what I believe to be the main issues we are trying to fight here. Finally, I am a pre-med student, so I have taken both physics and calculus in the past, and have a general idea about circuits, and wave representations, which have given me a base upon which to build my knowledge.

I would say that with respect to the video signal, we are trying to maximize bandwidth while minimizing voltage noise, and distortion, which leads to a slight error in the colour represented on the image on screen (as the amount of red, green and blue in a pixel is dependant upon the exact level of the flat peak of the square wave sent to the tube - any variation across this peak would lead to a section of the pixel being represented slightly differently at different points, as the electron beam sweeps across the tube face, varying in intensity in response to the voltage input level, which would lead to a slightly different colour being seen by our eye, as we cannot focus on a section of a pixel from 10 feet away, be only see and average light intensity coming from the general area). Furthermore, since this noise is transient, and therefore not always at a constant level, when representing the same pixel in the next frame,(ie. same pixel representing the same part of a shirt, or someone's skin, etc - not necessarily in exactly the same place as before) the noise could possibly cause it to look a slightly different colour when it is supposed to be the same, or it might be the same colour when it is supposed to be slightly different. I think this would explain the "shimmering" affect that video noise causes. This would generally explains why Jaehong found such a change when he cleaned up the G2 rail (although I know it is not affecting the video signal itself in this case) - the variations in the brightness level cause a similar shimmer effect in a fairly similar fashion.

The other result of this voltage noise, is that the closer you go to black, which is 0volts, the closer you go to the value of the peaks of the voltage noise. Therefore, the noise would be as large a value as the actual video signal, causing loss of detail in the resulting image, as our eyes see by contrast. This would explain the increased black detail in all of Mike's mods, from the MP1, to the VIM and neckboards.

Note that the above is basically saying that we want to increase the signal to noise ratio (S/N ratio) in order to increase the quality of the video reproduction at all (and especially the low) IRE values.

The need for bandwidth beyond the fundamental frequency of the video data is at least partially explained here: http://www.autopatch.com/pdfs/techno...t%20enough.pdf . This white paper discusses the need for higher bandwidth being needed to allow the higher order harmonic frequencies to pass through the electronics in order to faithfully represent the square waves of the video signal, and is worth the read as long as you ignore the blatant self promoting they do.

The last few paragraphs are how I've rationalized the importance of the components I've been looking at replacing - if anyone can see a major flaw in the way I've stated this, please say so - I am in the process of learning, and I do not know everything about this.

Once again, I HAVE NOT YET PERFORMED THESE (as I don’t own a marquee yet), these are simply the result of reading the manuals descriptions of the function of the Marquee, and analyzing the schematics, and looking at datasheets for both the existing components, and the ones I am putting forward as possible substitutes. Go here: for most of the datasheets for the original components: http://www.datasheetarchive.com/ .

1. VIM
What we know: 470nf coupling Caps, power supply buffering (ie. the tantalums and polypropylene caps), video signal opamps (ie. clc449s), mini-coax connectors
What I am proposing:


1.


2 x Voltage feedback, quad opamps (U8 & U9) – MC34084 and TL084 Replaced with 2 x OPA4134 (or OPA4132 - they are basically the same opamp - find these amps at www.digikey.com) (http://focus.ti.com/docs/prod/folder...t/opa4134.html for specs)
These opamps are involved in the processing of both the gain setting, and the contrast setting for the projector. Gain would have a similar effect to G2, I'd imagine, and would have similar results to Jaehong's resistor changes. Contrast is directly multiplied by the video signal by U27 on older boards and U14 on the newer boards (for the red channel). Therefore, any variations in the contrast line will be directly applied to the video signal. The original opamps are horribly noisy and distortive, therefore I'd imagine that this change would give a great positive result. Voltage feedback opamps are, as far as I can tell from the audio stuff I've read (http://www.diyaudio.com/forums), far less finicky than current sensing. Therefore, a direct replacement such as I am suggesting should be fine (maybe a bit more power supply buffering would be a good idea if problems arise). The OPA4134s are roughly 1/4 as noisy as the MC34084s, and less than a 1/2 as noisy as the TL084s and are also far less distortive. Most of the specs are similar, ie. bandwidth, signal range, and they allow for high enough voltage rails to easily handle the split +-14.5v rail that the marquee uses to run the originals. The OPA4134s have a higher slew rate than the TL084 type, but a slightly less slew rate than the MC34084 types. This is simply the maximum slope that the edges of a square wave can have after
being output by the amp. Therefore, this:
1. Shouldn't have much of an adverse affect even if we were dealing with square waves, and
2. Shouldn't have much affect at all on this circuit, as the contrast should be a DC
voltage, I'd imagine, once set by the projector.


2.


Note: Per Bill Blue’s suggestion, I have added another section to this part, which you should do first (ie. do the next section directly under this note first)

i.
Firstly, check to see if you have AD835s or AD834s as your multipliers (U14, U20, U26 for AD835s and U27, U28, U30 for the AD834s).

a)

If you have AD835s:
Connect a 100 ohm, surface mount, thin film resistor (like these: http://www.newark.com/product-detail...121/17047.html ) in series between pin 10 and resistor 97 (and then another between pin 7 and resistor R131, and the final between pin 1 and R167). Then add a (try 400nf, then 40nf, then 4nf, and check the difference- use the best – this creates a filter at 4KHz, 40Khz and 400KHz respectively) capacitor to ground (ie. Decoupling) between each pairs of resistors. Then add a 100pf, mica cap across the chosen cap.
CLARIFICATION: The intent is to stick a resistor between the pin#s I've given for the MC34084s (or replacement) and the resistors I mention. Now that I've had a chance to look at it, you could probably shift the current resistors over, so that they are on one pad only, then stick the extra resistor on the other pad, and solder a capacitor in between (so we have MC34084 pin --> pad 1 --> additional resistor --> (capacitor and solder connection between resistors) --> Original resistor --> pad2.


b)
If you have AD834s:
Connect a 100 ohm, surface mount, thin film resistor (like these: http://www.newark.com/product-detail...121/17047.html ) in series between pin 10 and resistor 294 (and then another between pin 7 and resistor R295, and the final between pin 1 and R296). Then add a (try 400nf, then 40nf, then 4nf, and check the difference- use the best – this creates a filter at 4KHz, 40Khz and 400KHz respectively) capacitor to ground (ie. Decoupling) between each pairs of resistors. Then add a 100pf, mica cap across the chosen cap.
CLARIFICATION: The intent is to stick a resistor between the pin#s I've given for the MC34084s (or replacement) and the resistors I mention. Now that I've had a chance to look at it, you could probably shift the current resistors over, so that they are on one pad only, then stick the extra resistor on the other pad, and solder a capacitor in between (so we have MC34084 pin --> pad 1 --> additional resistor --> (capacitor and solder connection between resistors) --> Original resistor --> pad2.


ii.
Across each of the three resistors: #60, #72, #78, place a (start with 160pf, then try 32pf, then try 16pf) capacitor. This should, according to the active filter equations given in this opamp guide:
http://focus.ti.com/lit/an/slod006b/slod006b.pdf (use the bookmarks to get to it)
Create a first order active filer (using the exiting opamp, the resistor, and the capacitor) set for filtering frequencies above (0.5, 2.5, and 5.0 respectively) Mhz. This should help to get rid some more of the opamp and DAC noise in the contrast line. You can get caps of this value pretty much anywhere, including digikey and newark. Use whichever of the values looks best, and don’t get too hung up on the exact capacitance of the cap, as long as it is close to what you want for the frequencies I listed above. I don’t know for sure what the amp typically outputs, although, I think it is dc, and therefore, of the above options, I’d say the 0.5 Mhz/160pf option looks the best.
Note that, based on Bill Blue’s suggestion to add a low pass filter in a location further down the line, I will be investigating exactly what values of components I need to do this, and I will post back later.



3.

3 x single voltage feedback opamps (U12 & U18 & U24) - TL071s replaced with OPA132 or OPA134 - again, find these for sale at www.digikey.com (http://focus.ti.com/docs/prod/folders/print/opa134.html)
These opamps help reset the black level of the singal during back porch, and their output is connected to the -ve input terminal on the video signal opamp U11, U17, and U23. The originals are just as noisy and distortive as the above ones (as they are just singly packaged amps of the same variety), and should be changed as well. The noise they input into the -ve terminal is getting at least partially into the video signal as a result, I'd think. This one I am less sure about having a large effect, but I'm sure it couldn't hurt.
Note that I am aware of the OPA627 opamp that many people in the audio forums like to use for low noise operation. However, I am unsure whether the increase in bandwidth (8Mhz to ~60Mhz) would cause a significant amount of higher frequency noise to enter the image, effectively nulling the improvement, and likely making things worse. Also, these opamps cost around 30 US dollars a piece, making them very pricey components.


4.
NOTE THIS IS AN ON-GOING RESEARCH AREA FOR ME. I recently found out that these switches are collections of Fet transistors, and I want to do some more research into this. I will post my findings – I believe this modification to be vitally important in making strides in the performance of the Marquee projector.

3 x jerry-rigged analog switch placed onto the connections connected to pins 14 and 11 on the existing analog switch, with the signal presently on pin 12 routed to the digital input in the chosen switch (pin 1 on the vishay DG613, pin 1 on the DG541 or DG542, pin 2 on the Pericom). Lift pins 14 and 11 from the existing switch, and mount the added switch beside it (or something like that, space permitting - use your imagination).There are some issues with this one that I would like to discuss. The SD-5401 analog switch that is in there is a piece of junk. This is evidenced by the fact that:
1. Jaehong said as much in the maintenance thread (pg. 13, you stated that the on-screen image became much more detailed when the switch was bypassed)
2. They don't list the bandwidth or distortions characteristics in the 2 versions of the
datasheet that I have.... This can't be a good sign. This is supported by the quoted 9500LC bandwidth of 120Mhz - much lower than anything in the signal path that we know the spec for.

The other, major issue I have run into is the fact that during every retrace, the switch is turned off. This is a period of 1.6 microseconds. Video is blanked for 1.8 microseconds during this same (roughly, I'm not sure of the exact time relationship - note these are the shortest intervals mentioned in the specs part of the 8110/9500 manual) period of time. This leaves 200ns for both turn on and turn off, as far as I can tell. However, the SD-5401 has a turn-on of .7ns and off time of about 10ns. This is much faster than I have been able to find in a replacement. I also cannot find a good pin compatible replacement either. From the Marquee info sheet, the fastest pixel the 9500 can handle is 3ns. Therefore, the on time for the existing switch is much faster than a single pixel. I have not been able to find a high bandwidth, low distortion analog switch that is that fast. However, the difference between the blanking and retrace times (remember the 200ns) and the fact that even a 40ns on time is only 12 or so pixels (and thus could be
possibly compensated for by front porch) gives me hope.

I have been thinking of using either or these three switches for the purpose of this mod:
http://www.saronix.com/products/swit...ductID=PI5A126 (Pericom PI5A126)
http://www.vishay.com/docs/70055/70055.pdf (Vishay DG541 or DG542)
http://www.vishay.com/docs/70057/70057.pdf (Vishay DG613)
The first, the Pericom PI5A126, has a bandwidth of 325 MHZ, and turn on of 14ns and turn off of 4.5ns. The distortion characteristic is not great, at 0.2%. It is still a considerable upgrade over the existing one. This switch operates from a 5v supply, which could be taken from a 5v source on the board (ie. the signal opamps - clc449s and such- are fed +-5V, taking 5v from the +ve side, and then placing a larger cap here to help buffering could work I think). The switch is a double switch. I think you could short all the pins but the digital input, in and out channels, and power in, to ground and then connect it to a ground on the board. You can buy this one at www.digikey.com
The second, the Vishay DG541/542, has an unlisted distortion characteristic (but it is stated that the distortion is negligible at the end of the datasheet) and a bandwidth of 500MHZ. This is much higher than the Pericom. However, the on and off times are longer, at roughly 45ns. The off time shouldn't hurt anything, as there is 1.4 microseconds allowed the switch for turn off. The on-time is equivalent to 15 pixels, which could be compensated for if necessary with porch settings. However, this might not be needed due to the leeway in blanking and retrace times. I am unsure. This switch is a quad switch, and is therefore larger than the Pericom, and might be harder to find space on the board for. This switch likes to run at a +15/-3 split voltage, but we can only supply a 14.5v supply and a -5 volt supply to the other side, using the power input from one of the voltage feedback amps on the board (or if there is a closer trace or something, it could be used) for the 14.5 volt, and one of the current feedback amps can supply the -5 volt. Similar method of shorting the excess pins should be used, I'd think. You can buy this one at www.newark.com
The third one, the Vishay DG613, is perhaps the best option if it works. It has an on-time of 12ns, and a 500 MHz bandwidth. However, the switch requires 4volts at its digital terminal to turn it on, whereas the original needs only 2volts. Therefore, I don’t know if the circuit uses anything higher than the 2volts it needs. Might be worth a try to see, as it is the best option. This is another quad switch, and requires the 14.5 volt rail for power like its brother. You can get this one at www.newark.com as well. Note that you must connect the 14.5 volt rail to pin 12 as well as pin 13. The -5volt rail should be connected to pin 4. If this causes the chip to not open and close properly (the value of the voltage here affects the threshold voltage for the gate), connect 5volts to pin 12. This should allow it to work properly. DECOUPLE BOTH SUPPLY rails well. Here is an application note about these switches, explaining their function and some of the issues revolving around on-resistance vs supply voltage, etc. :
http://www.vishay.com/docs/70605/70605.pdf
Finally, if you look at the Sanyo semiconductor store, you can buy these in packs of 10 (for 11.60 USD):

(datasheet)
http://lib1.store.vip.sc5.yahoo.com/.../SPM3215-e.pdf

(store)
http://www.sanyo.com/semiconductors/...tm#transistors

These have a fastish turn on time (between 10 and 100ns according to the datasheet), are switches (not Fets), that don't require an extra power supply, and have a low insertion loss at 1Ghz, but have a max control voltage of 5volts (I think we are currently using 7, so we'll have to adjust this is we want to use them), we don't know the current handling capability, or signal voltage level (probably around 5 volts, though, so we should be safe with this one), or the input/output capacitances, and are roughly 2mm long, so again will be hard to solder.

The other thought I had, if all three of these switches were too slow to be used, is to try a decoupling capacitor on the digital input in order to try and remove the opening and closing signal during retrace at R193, opposite side from the nearby XOR gate U32. I think I'd try 100Nf first, then work my way up a bit higher if that didn't do the job. I don't think this would cause any major problems, because Jaehong completely bypassed this switch and didn't suffer any problems that he reported other than the fact that he couldn't get into the internal setup menu. Note that doing this will, from all indications, cause problems.
Final word, I would suggest using the soic versions rather than the DIP versions due to space constraints and parasitic capacitance increase in DIPs. Or try the DIPs - they've used the DIP versions of the analog multipliers, so maybe its not that bad after all.



5.

I believe that I have found a good replacement for the CLC449s. This amp: http://focus.ti.com/docs/prod/folder...t/ths3201.html
has a slew rate of 10500, and a large signal bandwidth roughly double that of the clc449 at a gain of 2 (which is the gain we’re using here). It could replace U11, U13, and U23. The high slew rate would imply that it might be a bit more finicky than the clc449 with respect to feedback resistors. Therefore, I would suggest replacing R83 and R92 (red channel – R117 and R118 green channel – R153 and R162 blue channel) with 750ohm resistors (I was thinking of these to try: http://www.newark.com/product-detail...121/17047.html ) and R93 (red channel – R127 for green, and R163 for blue) with a 3600 ohm resistor (If this 3.6K causes any problems, lower the resister value by a hundred or so ohms). This should maintain the amplification, while maintaining the black-level reset voltage roughly the same coming from U12, U18 and U24.
Please note that I am aware that the main specs for this opamp are obtained while using a split +- 7.5 volt rail. However, based on the graphical representation of the frequency response at the end of the datasheet, it looks like the halved 0.1 Db frequency response when going from 7.5 volts to 5 volts is due to a SLIGHTLY higher peak in the frequency response at ~180Mhz (therefore this is the +0.1db frequency response), JUST going over the +0.1 db increment in this range. Other than this, the frequency response is very similar to 7.5 volt operation, retaining it’s –0.1 db rating out to over 300 Mhz, which is much higher than the CLC449.

6.

Based on some of the follow-up posts, I decided to add a section on power supply decoupling. Extrapolating on Bill’s comments (and remembering that all the datasheets I’ve seen have used multiple caps on the rails of their components while testing them), I will suggest that:
Firstly on the RGB signal opamps (and then anywhere else you feel like doing it on a power rail decoupling capacitor) lets add (note that playing around with these values will cause some performance change):
1 x 100PF cap, mica type – these are used in military and medical electronics, and retain their capacitance value past the Ghz frequency point (at least according to the spec sheets on the SM types at www.digikey.com)
1 x 100NF cap (or you can just leave the original there, although, this isn’t the best thing to do from a performance standpoint, due to the relatively small effect this the cap additions are reported to have, for sake of time, unless someone really takes issue with this, just leave the original there)
1 x High value (~6uf – 100uf) range tantalum or oscon cap (oscons are at www.newark.com, as are tantalums).

KEEP ALL LEADS AS SHORT AS POSSIBLE for maximum effect. For those reading this and wondering which way to connect the polarized, larger capacitors, connect the same polarity to the non ground side (voltage rail side) as the type of rail it is (ie. +ve rail, positive side of the polarized cap connected here). The other side should connect to the other side of the existing cap, which is ground (decoupling caps are caps from a line of some sort – power, signal, etc, that connects on the other side to ground – they are used to filter out noise, which is the deviation from the desired voltage value – higher value caps filter lower frequencies, and vice versa). The non-polarized types do not need to be compensated for as far as orientation is concerned.
The signal opamp (and I’m throwing in the multipliers for contrast as well) power caps are:
All type of boards:
+ve 5v rails: C38, C54, C70
-ve 5v rails: C27, C44, C70
Newer boards with AD835 multipliers as U14, U20, U26:
+ve 5v rails: C29, C46, C62
-ve 5v rails: C39, C55, C71
Older Boards with the AD834 multipliers:
+ve 5v rails: C109, C111, C113 (note these three are originally 1uf – either replace this one and add the other three, or add all of the other 3)
C39, C46, C63 – normal 100NF caps
-ve 5v rails: C110, C112, C114 (note these three are originally 1uf – either replace this one and add the other three, or add all of the other 3)
C40, C48, C74 – normal 100NF caps

7.

Remove C32, C43, C59.
These are located directly next to the input BNC connectors, and are video signal decoupling, and are setup as a passive filter for the video signal. I will be suggesting a way to use the first opamp on the video signal opamp on the neck boards as a 1st order active filter and therefore do not think these are required, as I am fairly sure they can cause some signal phase distortion (although if anyone feels differently, please say so).


8.

U15 is another one of the noisy, distortive MC34074 amps. It has to do with the internal video signal, and probably should be replaced with another of the OPA4134 (www.digikey.com) type opamps, since the internal video signal is activated during blanking and retrace, and could affect perceived black level, if there is sufficient noise beyond the active video, which is supposed to be black. This is unlikely, though, as this section of the circuit outputs blacker than black (therefore a negative voltage) during the time intervals it is active. Therefore, I consider this one a very low priority.


9.
Note: Might not be much change in doing this section – if you want to experiment and try, please post your findings.

Signal Coupling and decoupling resistors.
These resistors are in the RGB signal path, and maybe should be swapped out with newer, better models (like these again?: http://www.newark.com/product-detail...121/17047.html - note originals are 1/8W) :
Red:
R91(35.7 ohms) R103(39.2 Ohms) R1000(100 ohms) R90(39K ohms)
Green
R125(same as 91) R137(like R103) R1001 R124
Blue
R161 R175 R1002 R160
Note that there are a couple more, but I’m going to get into those later, as that section of the circuit opens another can of worms for us.


10.

Brightness value coupling resistors: same reference for replacement parts.
R25, R32, R34 - all 200 Ohms Note that I think this one might be overkill. Brightness affects black level. I will get into more mods involving this setting when we hit the neck boards. Like above, 1/8W


11.

I am wondering if there is enough room on the board to change from mini-coax connectors to normal full size connectors? Obviously the cables running from the VIM to the neck boards would have to be changed (or just change the end connectors) as well. I havn’t done much of any research into this, but it was an idea due to the fact that the minis seemed to be causing problems in the maintenance thread.


12.
Note: The gain from doing this section is suspected to be little at the current time. If you do these mods, please be prepared for this, and post back your results. Also, be prepared to have an urge to relace all the removed components.

Based on a couple recent posts about optimizing for RGBHV, and remembering Mikes sync on green removal, I have thought up a bunch of other slash’n’burn type mods.

i. Removed due to datasheet perusal, leading to the conclusion that the realys are fine as is.
ii. Since we’ve destroyed our ability to use anything but RGB, lets remove R107, R141, and R182. This removes the projector’s ability to detect composite and S-video at the RGB inputs. If this mod in its present state causes problems due to an un-terminated connection, similar to having an un-terminated RGB connector on a video card, switcher, etc, then replace these resistors with a ~2Mohm resistor in place of each. If there are still problems after this, please undo the mod and post back so that I can remove it from my post.
THIS DESTROYS THE PROJECTORS ability to detect composite and S-video input into the RGB BNC connections and the secondary RGB input.
iii. Who wants their Marquee to sing? Not me. Therefore, remove the projector’s ability to produce sound by:
Removing R6 and R3. These should be directly after the audio-in.
Lifting Pins 2, 6, 11 and 15 on U37.
Lifting Pins 8 and 16 on U36.
The pin lifting should remove the un-necessary burdens of these two IC’s from the power supply, while the resistor removal will prevent the goddawful noise that this thing probably tries to pass for sound reproduction to ever reach your ears by breaking the signal line.

13.

Analog Multiplier discussion.
Here are the datasheets for the two multipliers:
http://www.analog.com/UploadedFiles/...087AD834_d.pdf
http://www.analog.com/UploadedFiles/...427AD835_b.pdf
I THINK boards with the AD834s might be better than the revisions with the AD835s due to noise and bandwidth limitations on the AD835. The fact is that the 835 is basically an AD834 with a built-in opamp. This built-in amp has a 250mhz -3db / FIFTEEN Mhz 0.1db (CLC449 has 500 and 200 for those two, respectively) and a 50nV/(Hz)^0.5 noise figure (which is very high – as a reference, the CLC449 is around 2.2 – note that these references are often exaggerated downwards, also the AD834s contribute 16nv/(hz)^0.5 , so the difference by spec is around 32 - which is again, a lot, especially when we are dealing with the very high frequencies found in video applications), and a 1000V/us large signal slew rate (compare to the clc449 ~2000). The AD834 is 500mhz -3db, and needs to be used with an external opamp, and several other components - which can obviously be upgraded (and I personally like options like this)- but will cause a bit of noise themselves - a clc449 was used on the newest revision of the Marquee boards containing these multipliers (therefore, these two components are not directly interchangeable). Take it for what its worth – I believe that it would be beneficial to have one a board with AD834s, but others have expressed opinions to the effect that they both have merits, and it is a bit of a matter of preference (I havn’t seen the difference personally, I can only go by the specs).


14.

Use a 100NF capacitor to connect pins 4 and 11 on the quad voltage feedback opamps U15, and one to connect pins 4 and 7 on each of the three opamps U12, U18, and U24. This should help quiet them down a bit if they were doing anything undesirable before. U8 has power on pins 4 and 13.


15.

Resistor mods/upgrades.
All of the following resistors are 5% tolerances, and should probably be changed out with these, which are 0.2% tolerances for the 1/8 W versions, and have only 25ppm/degree temperature drift:
http://www.newark.com/product-detail...121/17047.html

R90, R124, R160 - 39K these are a part of the input differentiator circuit, and should probably be as close to the desired ohmage as possible. 5% of 39K is a fair bit, weighing in at 1950ohms.

AD834 multiplier boards (03 revisions): multiplier power rail resistors:
R53, R79, R85, R86, R290, R285 - 10ohms

The next resistors are 5% tolerances, but are not available in the models I referenced above. Here is a possible replacement: http://dkc3.digikey.com/PDF/C042/0910-0911.pdf - look on page 2 for 1/8W, 1% models. These drift a bit more with temperature, but are still pretty good at +- 200 ppm.

R82, R110, R144, R117, R176, R151 - 4.7ohms. These are the resistors on the supply rails to the first clc449/signal amps. I think it might be worth changing these out to 1ohm resistors (or less), like the electrohome engineers did with the blue gamma amp, as this would allow a bit more voltage to the amps, and hopefully improve performance. It might be too small a change to notice, though.

R95, R129, R165 - 1ohm. These are in the clamp circuit beside the first signal opamp.

R199, R190 - 1ohm. These are on the power rails of the blue gamma amp.

AD834 boards: R267, R268, R271, R272, R275, R276 - 4.7ohm. Maybe replace these with 1ohms as well?

AD835 boards: R45, R42, R53, R87, R111, R121 - 1ohm





Vic




-Note: I've edited for a couple typos, and for a couple time values I had to remember from the top of my head that turned out to be slightly off in the switches section. I was typing at work earlier.
-Note 2: Changed the function of the first two opamps from Contrast and Brightness to Contrast and Gain. I was slightly off before. The brightness comes directly off DAC U10 and heads to the neck boards, whereas the gain and contrasts come off the same (but different from the brightness)output off the DAC, and then are independantly created by these opamps (the CLM controls the master contrast, I mean that the DAC is the local originator).
-Changed blanking to backporch in desricption of single volt-feedback amps.
-Added a bit more into the switches section on possibly removing the turn-off signal during retrace if the replacement switches prove to be too slow.
- June 18: Major update on the VIM mods.
- Added small note to relay removal regarding grounding of relay switching signal (as I don't know if this can perhaps cause damage to the DAC that produces the signal if the signal is not going anywhere).
- Removed a lot of the Switch section, as I have found out that all is not quite how I perceived it here. I am currently researching this.
- Added note to section 12
- Changed value of noise of TL084s in my description.
-Swapped a couple of sections around – I am sorry if it causes any confusion for now, but it will be easier to find the associated information in the future. Specifically, the decoupling of power rails was moved to next to the signal opamp section, and the creation of the active low pass filter was moved next to the contrast setting opamp discussion. I updated the latter section with information on how to build an RC lowpass filter here – Thanks to Bill for the suggestion!
-Added notes to the sections that are now #s 2,3,4,5 and 6.
-Added further discussion of background theory, this time involving projector and components Bandwidth requirements, and S/N ratio.
-Added section discussing the merits of the two types of multiplier IC’s used on the VIMs.
- Altered my statement about using a 2nd order filter on the neck boards – for now this is a 1st order, as I took a look, and I am unsure that a 2nd order is feasible here for a couple of reasons.
-Update noise figure in the AD834 / AD835 discussion - I had meant to do this before posting that section, but I forgot.
-Added pf capacitor addition to #2
-Added note to #1
-Added some clarification to #2
-Removed section regarding relay removal.
-Added a bit more to switch section, VIM #4
-Added a bit more to switch section, VIM #4, again
-Added in a #15 for some resistor changes.
July 10
-Added more to switch section, #4
-Added a possible replacement for the smaller value resistors in section #15.
-Removed rail to rail cap addition suggestion


EDIT: I downloaded the spec sheets for the old components mostly from here:
http://www.datasheetarchive.com/
mcpherv is offline  
post #13 of 465 Old 06-16-2004, 09:53 AM
Member
 
Robert Zuch's Avatar
 
Join Date: Dec 2001
Location: Armonk, NY
Posts: 65
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally posted by bblue
Gernot,
to me those two caps look like the originals. They should be replaced with the 33mfd 160v 105 degree caps. Audio based performance caps will do no good there.
We have had success with changing both the resistors and the cap's. Cap's with improved spec's (ESR, etc.) can improve PQ.

On increasing the wattage of a resistor, that should not be a problem. The load factor (whether it be impedance or power factor) will only change if the values of the components in the circuit change.

It's great to see so many people working on keeping CRT projection the King of Kings!

Bob Z.
Reference Imaging
Robert Zuch is offline  
post #14 of 465 Old 06-16-2004, 02:49 PM
AVS Addicted Member
 
mark haflich's Avatar
 
Join Date: Dec 2000
Location: brookeville, maryland, usa
Posts: 19,510
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 229 Post(s)
Liked: 363
I agree with Robert. Changing the wattage of a resistor will have no load effect on the circuit. Changing cap values can be a problem re load but there can be no harm by substituting a same value but higher grade cap and such things as DA and ESR are evils with respect to caps whose reduction can only help even a video circuit. :)

VIC! Welcome to the CRT forum. An awesome post! I would be curious to hear from MP and BBLUE re you ideas. Thanks for sharing your ideas with us.

Mark Haflich
markhaflich@yahoo.com
call me at: 240 876 2536
mark haflich is online now  
post #15 of 465 Old 06-16-2004, 07:27 PM - Thread Starter
Senior Member
 
Jaehong Lee's Avatar
 
Join Date: Mar 2000
Location: Seoul , Korea
Posts: 345
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
VIc awesome research indeed !!!!

Gernot

I changed 22uF 160V/105degree capacitor, but not black gate.

I had cahnce to look into Accurate Imaging Technologies , A.I.T reference Ultimate 9X which is ultimate upgrade versiopn of Marquee 9500LC.

That machine used 100uF tantaluim capacitor in neck board.

one user in here repalced same tantalium capacitor ( Vishy wet tantalium) , but much darker image. so he replaced again with 100uF elctolytic one, which is oK.

BTW AIT ultimate 9X used a lot of wet tantalium capacitors in nearly every board.

2 users in here changed as AIT 9X did . results is minor improving but not so much. they spent 1200$ only for purchaing tantalium capacitors.

you can see naked AIT ultimate 9X at http://www.kavforum.com/Visual/Scree...e=ÃÖ¿øÅÂ
Jaehong Lee is offline  
post #16 of 465 Old 06-16-2004, 07:46 PM - Thread Starter
Senior Member
 
Jaehong Lee's Avatar
 
Join Date: Mar 2000
Location: Seoul , Korea
Posts: 345
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
I also upgraded VIM board as shown in here

http://www.kavforum.com/Community/cl...e=&SearchName=

Decouping OP amp IC power rail pins with 1uF/50V polyproplene cap. and changing to expensive wet tantalium capacitor as AIT ultimate 9X did.

Result is some benefit of blacker black.

i don't recommend changing wet tantalium capacitors.

It is too expensive with a little benefit.
Jaehong Lee is offline  
post #17 of 465 Old 06-16-2004, 08:21 PM
AVS Special Member
 
bblue's Avatar
 
Join Date: Nov 2001
Location: El Cajon, CA
Posts: 1,051
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Vic,
Thanks for an excellent and thought provoking post! You're attacking some areas in the VIM design that may well deserve some looking into, along with the standard analog upgrades.

The Opamp you proposed as a replacement for the CLC449 is also quite interesting and is one I hadn't stumbled across yet. It looks initially like a reasonable candidate, except an interesting omission in their spec sheet caught my attention. They repeatedly give specs and graphs for small signal non-inverting, and large signal inverting modes of operation. But nothing anything for large signal non-inverting. Otherwise it's got the wide .1 db down bandwidth and high slew that I would look for.

I need to re-read your post a few times and look at the references for it to all sink in.

--Bill
bblue is offline  
post #18 of 465 Old 06-16-2004, 08:32 PM
AVS Special Member
 
bblue's Avatar
 
Join Date: Nov 2001
Location: El Cajon, CA
Posts: 1,051
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Jaehong,
None of the pictures of the AIT ultimate seem to be coming up, despite browser refreshes.

I think you're largely wasting your time with large polypropelene caps on the rails. For effectiveness, you need physically small components with very short leads. Mylar or stacked film caps do a very good job across a wide frequency range, which is one of the goals here. Usually you will get the best performance with a combination of 2 or 3 caps of different capacities (and physically small) to cover the desired frequency ranges for bypass or reservoir functions.

If you have ever seen Mike's VIM and VNB board mods in action, you would appreciate this wisdom. You're headed the right direction but with the wrong type of parts.

Do you know what the difference is supposed to be between standard and wet tantalums?

--Bill
bblue is offline  
post #19 of 465 Old 06-16-2004, 08:37 PM
AVS Special Member
 
bblue's Avatar
 
Join Date: Nov 2001
Location: El Cajon, CA
Posts: 1,051
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Mike,
Quote:
and the ones before the G2 itself is a 100K to a 10k, with a 22meg going to ground. Based on the value of the 22meg, it must be a discharge resistor.
Yes, I'm sure it is. On the boards I have here it appears to always not be a carbon comp, though. Maybe a carbon film.

Anyway, it seems logical then to replace the 10k and 100k if for no other reason than they are comp!

--Bill
bblue is offline  
post #20 of 465 Old 06-16-2004, 08:38 PM
Senior Member
 
mcpherv's Avatar
 
Join Date: Aug 2003
Location: London, Ontario, Canada
Posts: 498
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
(note that this post was edited)


DISCLAIMER

Guys, just a warning here. If you want to try to apply any of the ideas contained within this thread, you do so at your own (and your projector's) risk. Also, this is all a work in progress, and therefore, this thread will not GUARANTEE you significant performance increases (other than the priorly established and tested performance increases). Also, please do not expect to obtain the level of performance that the professionals have been able to obtain up until this point.

DISCLAIMER




POST 3:

Modifications to the Focus board, followed by the Control Board, followed by the Necks boards.



NOTE: There is a list of changes that have occurred during revisions to this post at the bottom.

2. FOCUS Board


1.

Capacitor changes (all of these will affect the focusing ability of your projector, as the existing caps are aged electrolytics – note that this part is more maintenance than modification):
Focus Coupling Caps:

C102, C202, C302
C103, C203, C303 All 6 are 22uf, 50V (replace with same value Panasonic FC series –www.digikey.com caries them) These caps should be fairly important, I’d imagine, as they couple the dynamic focus output.

C120, C220, C320 These are 2.2uf, 250V, non-polar. All three couple another section of the dynamic focus. Take a look at www.digikey.com for possibilities for replacements – replace with same specs, (but 105 degree) non-polar electrolytic capacitors.

Power decoupling caps:

C108, C208, C308 These are decoupling for the –24volt rail supplying one half of the dynamic focus circuit. All are 470uf, 35V (replace with same value Panasonic FC series – www.digikey.com caries them – increase the voltage rating)

C107, C207, C307 These are decoupling for the +24volt rail supplying one half of the dynamic focus circuit. All are 470uf, 35V (replace with same value Panasonic FC series – www.digikey.com caries them – increase the voltage rating)

C5, C10 These caps are +- 15V rail decoupling. They are 470uf, 35v. (replace with same value Panasonic FC series – www.digikey.com caries them – increase the voltage rating)


2.

OPAMPS 101, 201 and 301 are all MC34074 Opamps, which are more of the noisy, distortive opamps we dealt with on the VIM. These ones are involved with processing the static focus portion of the projector’s focusing ability. However, these ones are DIP versions of the amps, which poses a bit of a problem for us, as Texas Instruments have discontinued the DIP versions of their OPA4134s and the OPA4132s. So, for anyone wanting to try this change, you will have to use the SOIC (surface mount versions) and use a socket adapter (like the ones on the right side of this page – SOIC to DIP: http://dkc3.digikey.com/PDF/T042/0281-0282.pdf - note you need 14 pin versions here for the quad opmaps). Note that using these socket adapters is not ideal, and it likely will cause loss of performance compared with using the discontinued DIP OPA4134s. Hopefully this change will still be better than using the original amps – please post your results if you try this modification.



3. CONTROL BOARD (CLM)

1.


OPAMPS U73 (TL084 original), U76 (MC34084), and U82(MC34084) all create focus waveforms, and are more of our favourite noisy, distortive amps. Since these ones are creating waveforms, for now, we will refrain from filtering their outputs. Therefore, these should be replaced by OPA4134, one at a time, starting with U73, and checking changes in the focusing ability of the projector after each. This is one area where the lower slew rate of the OPA (lower than the MC34084, not the TL084) might have a negative effect, but the other much improved qualities in the replacement (noise and distortion figures) hopefully will prove to be more important. After this, place a 100NF capacitor from pin 4 to pin 11 on all three of these (even if you havn’t replaced them).

2.

There are a lot of decoupling capacitors on this board in order to prevent the digital circuitry from polluting the power rails. I have read conflicting opinions as to the benefit of adding additional capacitance (KBK stated this was a huge improvement, whereas Mike Parker and Mark Haflich have gone on record to say that it is not a major issue) on this board, however, I believe that it is worth noting here. Therefore, borrowing from the VIM section, this is what we are going to do to each of the capacitors I am going to list (note that I recommend for those of you on a budget to get 47uf, radial type Tantalum capacitors from ebay – they usually go for a decent price in bulk there):


1 x 100PF cap, mica type – these are used in military and medical electronics, and retain their capacitance value past the Ghz frequency point (at least according to the spec sheets on the SM types at www.digikey.com)

1 x 100NF cap (or you can just leave the original there, although, this isn’t the best thing to do from a performance standpoint, due to the relatively small effect this the cap additions are reported to have, for sake of time, unless someone really takes issue with this, just leave the original there)

1 x High value (~6uf – 100uf) range tantalum or oscon cap (oscons are at www.newark.com, as are tantalums).

Note that KBK only directly swapped out the existing capacitors with tantalums when he did it – he didn’t have the other values in there as well. Hopefully, this will improve the effectiveness of this mod. Here are the capacitor numbers (note there are 107 of them, not including the electrolytic listed separately):

+5 volt rail: X1, X2, X4, 5, 10, 11, 19, 21, 24, 25, 26, 27A, 27B, 28A, 28B, 29, 30, 33, 35, 36, 38, 40, 41, 42, 44, 45, 46, 47, 48, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 74, 75, 77A, 80, 81, 83, 88, 89, 91, 94, 95, 96, 97, 98, 99, 100A, 100B, 100C, 101, 102, 104, 106A, 106B, 107, 108, 109, 110A (Note all of these are designated as X#, not just the first couple , all are originally 100NF)

+5 volt rail electrolytics (replace with Panasonic FC series, www.digikey.com, keep capacitance value, raise the voltage)
C6, C7 - both 47uf, 16V

-5 volt rail: X77B, X110B (both 100NF)

-5 volt rail electrolytics: (replace with Panasonic FC series, www.digikey.com, keep capacitance value, raise the voltage)
C113 - 10uf, 16V

+15 volt rail: X23A, 37A 39A, 73A, 76A, 82A, 85A, 86A, 87A, 90A, 92A, 93A, 105A (all are 100NF, all are X#)

+15 volt rail electrolytics: (replace with Panasonic FC series, www.digikey.com, keep capacitance value, raise the voltage)
C2 - 10 uf, 35V

-15 volt rail: X23B, 37B, 39B, 73B, 76B, 82B, 85B, 86B, 87B, 90B, 92B, 93B, 105B
(all are 100NF, all are X#)

-15 volt rail electrolytics: (replace with Panasonic FC series, www.digikey.com, keep capacitance value, raise the voltage)
C1 - 10 uf, 35V


4. NECK BOARDS

For resistor changes, see here (very important, a highly effective mod):
(see first post in tis thread by Jaehong Lee)

1.

Brightness circuit: I believe to be absolutely critical to detail in low – light scenes, as this sets the cutoff point of your crts – therefore instabilities here will cause a slight “bounce†in background brightness, which will cause a loss of detail in dark scenes, as this bounce will bring the “black†close to the level of imgae detail. Therefore, we are going to try to stabilize this circuit. Note that this circuit outputs DC, I believe.

i.

DO NOT TOUCH U1. THIS IS PART OF THE PROTECT CIRCUIT.
U4 is a DIP version of a MC34082 opamp. The OPA2132 (http://focus.ti.com/docs/prod/folder...t/opa2132.html , available at www.digikey.com) is still available in DIP packaging, and therefore we are going to swap this opamp out (I’m sure I don’t have to remind you of the reasons at this point). Add a 100NF capacitor from pin 8 to pin 4.
DO NOT TOUCH U1. THIS IS PART OF THE PROTECT CIRCUIT.

ii.

Add a 100pf, mica cap across both C46 and C21.

iii.

Replace C2 with a better quality film cap.

PLEASE DON'T DO THIS ONE UNTIL SOMEONE ELSE HAS
Look at U1 (and see above note in section i.). Then find R36. Now, on the side of R36 that is AWAY from U1, connect first a (first a 10NF, then a 1NF – creating 40KHz, and 400KHz low pass filters respectively) capacitor as a decoupling capacitor. Evaluate each performance, and keep the best. Then add a 100pf, mica cap across the cap you decided upon.
PLEASE DON'T DO THIS ONE UNTIL SOMEONE ELSE HAS

iv.

Add a 100pf mica cap across C2.



2.

I know I previously stated that we would create a 2nd order low-pass active filter around OPAMP U2 here, but for now, for a couple of reasons, we will try a 1st order. Therefore, replace R100 with a 10ohm, 1/8 watt thin-film (like these http://www.newark.com/product-detail...121/17047.html ) and, on the OPAMP side of the resistor, connect a 32 PF (or there about) capacitor of good quality (like the MICA capacitors I have been suggesting for usage in decoupling) as a decoupling capacitor. This filter is designed with a 500Mhz corner frequency.

3.

I believe that I have found a good replacement for the CLC449s. This amp: http://focus.ti.com/docs/prod/folder...t/ths3201.html
has a slew rate of 10500, and a large signal bandwidth roughly double that of the clc449 at a gain of 2 (which is the gain we’re using here). It could replace U2. The high slew rate would imply that it might be a bit more finicky than the clc449 with respect to feedback resistors. Therefore, I would suggest replacing (http://www.newark.com/product-detail...121/17047.html ) R12 with a 750 ohm resistor, R22 to 300 ohms, and R90 to 4.87 Kohms. This should keep all the resistors in this area close to the original ratios, in order to maintain the voltage dividers.

Please note that I am aware that the main specs for this opamp are obtained while using a split +- 7.5 volt rail. However, based on the graphical representation of the frequency response at the end of the datasheet, it looks like the halved 0.1 Db frequency response when going from 7.5 volts to 5 volts is due to a SLIGHTLY higher peak in the frequency response at ~180Mhz (therefore this is the +0.1db frequency response), JUST going over the +0.1 db increment in this range. Other than this, the frequency response is very similar to 7.5 volt operation, retaining it’s –0.1 db rating out to over 300 Mhz, which is much higher than the CLC449.


4.

Decouple U2 like we have been doing to the others:

1 x 100PF cap, mica type – these are used in military and medical electronics, and retain their capacitance value past the Ghz frequency point (at least according to the spec sheets on the SM types at www.digikey.com)
1 x 100NF cap (or you can just leave the original there, although, this isn’t the best thing to do from a performance standpoint, due to the relatively small effect this the cap additions are reported to have, for sake of time, unless someone really takes issue with this, just leave the original there)
1 x High value (~6uf – 100uf) range tantalum or oscon cap (oscons are at www.newark.com, as are tantalums).


5. (copied here from pg. 12)
The output stages of the neckboards are fed 85 volts. It is used to feed the output buffering stages for the video signal, and I'd expect that having that rail as clean as possible would be a good thing. If you look at pg 1 of the neckboard schematics, you can see that each neck board filters the +-85 rails with a 100uh inductor, and a 22uf in parrallel with a 100nf cap (+ve: L8, C1, C52 ; -ve: L9, C26, C56). I'd expect that beefing this up, combined with adding some filtering to the LVPS on these rails would be a very good idea. The other thing is that the large, (top of the pg 2, middle section) 400ohm 15w non- inductable resistors on those boards also are in the path to feeding the output stages, and we might seriously benefit from replacing them, if they are at all similar in quality to the carbon comps in the G2 rails and the final output stages as well. Looking at it, there is a 820uh inductor in parrallel with a 100ohm resistor directly after the large resistors (L3, R84 for +ve, L1, R85 for -ve), so the electrohome engineers must have thought of this as well. There is also another place where the 85 volt rails are applied in parrallel with the large resistors, which is decoupled with 10Nf caps C52 and C58 (+ve and -ve). These would be good to beef up as well, I'd imagine. Next to this, there is a third parrallel section, which shows one 100Nf decoupling cap on each side as well. These are C60 and C59 (+ve, -ve).

6.




- Added links to Jaehong Lee's Posts about resistor changes to neck boards.
- Added pf capacitor recommendation to Neck #1, iii.
- Edited out the 4Khz low pass on the output of neckboard U1, as I am unsure that low is adviseable
-Added a section neck iv.
-Completed neck #3
-Added note to CLM #1
-Added note to Neck #1 part iii
-Added another note to Neck #1 part iii
-Added a Neck #5
July 11
-Miscellaneous small changes, removed rail to rail cap addition suggestion
mcpherv is offline  
post #21 of 465 Old 06-16-2004, 08:41 PM - Thread Starter
Senior Member
 
Jaehong Lee's Avatar
 
Join Date: Mar 2000
Location: Seoul , Korea
Posts: 345
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Bill

standard tantalium cap has limit in value and voltage.
wet tantalium cap from Vishy has large value such as 100uf/50V, 560V/25V
Jaehong Lee is offline  
post #22 of 465 Old 06-16-2004, 08:57 PM - Thread Starter
Senior Member
 
Jaehong Lee's Avatar
 
Join Date: Mar 2000
Location: Seoul , Korea
Posts: 345
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
VIC

I added only 1uF/50V polypropelene cap which is very small, to OP amp IC power rail pins.

I changed wet tantalium caps at the bottom of VIM board as shown in

http://www.kavforum.com/Community/cl...e=&SearchName=
Jaehong Lee is offline  
post #23 of 465 Old 06-16-2004, 09:15 PM
Senior Member
 
mcpherv's Avatar
 
Join Date: Aug 2003
Location: London, Ontario, Canada
Posts: 498
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Jaehong, I was under the impression that others in Korea had tried upgrading the individual power rails with wet tantalums. I know KBK put in 47 uf tantalums wherever he found a decoupling cap on all the boards in his Marquee a long time ago, and raved about the changes. He mentioned that the control board gave the best results while doing this.
I was planning on using either normal tants or the oscons I just posted for the purpose of decoupling the individual power rails.
Vic
mcpherv is offline  
post #24 of 465 Old 06-16-2004, 10:35 PM
AVS Addicted Member
 
mark haflich's Avatar
 
Join Date: Dec 2000
Location: brookeville, maryland, usa
Posts: 19,510
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 229 Post(s)
Liked: 363
The noise on the control board is lowered by about 2db with the KBK cap substitutions. A whole lot of caps were replaced by KBK on my control board.

The biggest improvement to my machine was the MP neck board mods and the VIM mods. I consider these boards to be works in progress based on my readings here. Vic there are other improvements to the boards possible other than the ones you are considering. THe MP mods address some things you have not yet considered. Some of these may be proprietary and are better addressed by my friend MP rather than me.

Mark Haflich
markhaflich@yahoo.com
call me at: 240 876 2536
mark haflich is online now  
post #25 of 465 Old 06-16-2004, 11:04 PM
Member
 
Robert Zuch's Avatar
 
Join Date: Dec 2001
Location: Armonk, NY
Posts: 65
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally posted by Jaehong Lee
I had cahnce to look into Accurate Imaging Technologies , A.I.T reference Ultimate 9X which is ultimate upgrade versiopn of Marquee 9500LC.
Well, I have to disagree, although I'm biased... We are very familiar with the Ultimate 9X and it's design is very similar to our original projector (Version 1 of the Cinepro 9x Elite). We are now up to Version 5 of our upgrades, and are getting the best PQ ever!

Jaehong, your ideas and information are very much appreciated. But please try to avoid posting pictures/info obtained from "reverse engineering" someone else's design. This is against forum guidelines for good reasons; if a company or individual wants to share the results of their R&D that is great but let it be their decision to do so. For example MP shares many of his ideas but keeps his designs proprietary. We try to do the same. Thanks to everyone who shares ORIGINAL ideas/designs and evaluations of the various projectors available.

mcpherv, awesome post! I'm sure you spent countless hours of R&D; thanks for sharing!

Bob Z.
Reference Imaging
Robert Zuch is offline  
post #26 of 465 Old 06-17-2004, 01:03 PM
Senior Member
 
mcpherv's Avatar
 
Join Date: Aug 2003
Location: London, Ontario, Canada
Posts: 498
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Thanks everyone for the feedback and for the positive responses so far.

Jaehong - Thanks for following up on those resistor changes.

Marc - Thanks for the information regarding KBKs cap mods, the specifics are very much appreciated. I do have a couple more mods coming for the VIM, although I've already posted the major ones I've been thinking of and researching into.

Robert - Thanks to you to for posting your feedback as a professional in this area. I have indeed spent many hours on R&D.

I've gotten started on the outline of the next major post. I still have mods for the VIM (although as I stated, these are mostly finished for now), Neckboards, Control board, and Focus board. I havn't done much in the way of research into the deflection boards and convergence board, as I feel that Mike has already done a pretty thorough job on them. Most of these revolve around the same various species of amps I've recommended changing on the VIM, and a few cap mods, and resistor mods. Thanks,
Vic
mcpherv is offline  
post #27 of 465 Old 06-18-2004, 01:01 PM
Senior Member
 
mcpherv's Avatar
 
Join Date: Aug 2003
Location: London, Ontario, Canada
Posts: 498
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
BUMP for major update to my first mod post covering more VIM mods.
The others are coming, I just need to get in the authorizing frame of mind, as there are a lot.
Vic
mcpherv is offline  
post #28 of 465 Old 06-18-2004, 03:10 PM
AVS Addicted Member
 
mark haflich's Avatar
 
Join Date: Dec 2000
Location: brookeville, maryland, usa
Posts: 19,510
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 229 Post(s)
Liked: 363
I am getting confused. MP's mods are to the VIM and to the neck boards.

Mark Haflich
markhaflich@yahoo.com
call me at: 240 876 2536
mark haflich is online now  
post #29 of 465 Old 06-18-2004, 03:14 PM
Senior Member
 
mcpherv's Avatar
 
Join Date: Aug 2003
Location: London, Ontario, Canada
Posts: 498
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Yes, I just updated the post you read and responded to the other day (the big one), with more mods to the VIM, as I wanted to keep them together. The rest of the ones I am going to cover are going to be for the neck boards, focus board, and CLM.
Vic

PS Sorry for the confusion
mcpherv is offline  
post #30 of 465 Old 06-18-2004, 03:57 PM
AVS Special Member
 
bblue's Avatar
 
Join Date: Nov 2001
Location: El Cajon, CA
Posts: 1,051
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Good grief, Vic! I haven't finished on the first version yet! :-)

Thanks for the update. Will get into it when I get done with some work in my theatre.

--Bill
bblue is offline  
Closed Thread CRT Projectors

User Tag List

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off