I thought of doing this back in December. But decided that the month of January would probably be better, that way we can start 2003 out by giving our prized Marquees a good maintenance work over.

I think we should keep this open to anyone, and feel free to ask any question, or contribute any experience that you may have. There's always something new to learn about this projector.

I think the Marquee is very near its tenth birthday. It's the only commercial grade CRT projector (that I know of) that has maintained it's basic design for that period of time, and in its basic design, it continued to challenge and outperform a many CRT competitor, and did gain great status in the very demanding commercial and industrial arena. But over its ten years of being one of the best, there have been some minor changes made by the manufacturer, likewise there has also been some field experiences that has made it a much better performer as the years went on. So maybe it’s time we share what we know, so that all who has one can possible make it better.

And much thanks to Mark Haflich, who has loaned me one of his 9500’s so that I could sharpen my tools. It’s been fun. I’m now able to successfully repair most of the boards in a Marquee. There’s only been two boards that I’ve not been able to repair, that is the processor board and the HVPS. The processor board rarely fails, so it’s been hard to get my hands on a defective one, and the HVPS is on the bench as I type. And it’s only a matter of time, and I’ll place it among the rest of the repairable modules. If that coil in the potted material does not fail, I’ll have that power supply licked. I know exactly what’s inside, and I’m prepared to remanufacture a certain component if it’s a problem. I’m confident that I’ll be able to resurrect these power supplies.

I myself won’t get into tweaks, but let’s be open for discussion on them. I’m not doing this to promote business for myself, I’m willing to share a lot of what I know (not all), and most of what I’ll share you can do yourself, or have a competent technician do it for you.

To start, we’ll cover the neck boards. This is a very important module for performance in this projector. It’s the weakest link for performance on a Marquee, there has been several changes over the years to this board. And this was the biggest negative on the early 8000’s – We’ll look at and discuss how to improve the bandwidth and correct that high frequency roll-off problem (soft image/smearing) that plague the early Marquees.

We’ll also discuss the blanking problem (blanking out on bright scenes) and the fix.

And I’ll try to remember to post the power supply’s cleaning procedure, very important for ventilation.

That’s enough for now, I’ll get back to this later today

Hi Mike Parker

The short answer is YES. I will check this thread on every forum
visit for your follow ups. Thanks for sharing.

DH

Yes as well, please post all details of Marquee maintainance and improvement, they are an easy pj to pull apart, it would be nice to be able to do something with it when its apart.

Thanks, Mike. Sounds like a great idea.

Hi Mike

I'm waiting with a hot soldering iron in my hand ;)

When you say you wont get into tweaks, do that mean you don't do them or are not willing to share them? I remember you once said that you have some component updates, but you where not willing to share them, "for the moment". I was hoping the moment had come :)

Kind regards, Claus

Mike,

This is just a great idea, and considering you are not gaining anything monitarily, I applaude your commitment to the projector.

I'm sure everyone will appreciate what you're doing. You're very kind.

Ted

Mike,

that's very generous of you !
I hope in the near future to be able to profit from your vast experience and knowledge of the Marquee :)

Thanks

Mike I applaud your generosity... this is the type of thing that makes forums like this utterly addictive.

I just bought a Marquee 8500 and I know it could benefit from some good lovin :o)

I PM'd you last night with a particular concern but I will keep my eyes on this thread and hope that an FAQ evolves from it with contributions from all the Electrohome Wizards that frequent the forum.

Stephan
Vancouver, B.C.

Me too! I will watch and read with great anticipation. You mentioned some areas where my projector could use help. Soft image and so on.

I am a new member and the search function has let me down every time, so I would love to see all information whether old or new.

Thanks so much!

Need I really say anything :D

Mike, It would be great to have a list of capacitors to replace, with better specs or just similar specs for older marquees, if this makes sense PQ-wise.
Rob

We're not going to use the boards part numbers for this. So you'll have to visually inspect your boards for the following:

1, look for the numbers on IC 'U2' (near the mini RCA jack). On that IC you should see one of the following numbers: 'CLC409, CLC449, H1100.

2, Check transistors Q12 and Q19, see if they are round metal (TO-5) devices, or small square (SOIC) devices.

3, Check the opposite end of the card for four soldered on 'blue' capacitors.

4, Inspect the solder contacts of resistors R13 and R28, for burn at the point of soldering on the PC boards surface.

5, Inspect coils L8 and L9 for any indication of heat damage.

--------------------

(1) If IC 'U2' is a CLC409 it should be replaced with IC CLC449. Also if there's a H1100, it could also have a softening effect on the image, because it was later replaced with CLC449 (and some H1100's were also on the later boards 8500, 9500's).

(2) If transistors Q12 and Q19 are metal (TO-5) you have an early version card, and this board would more likely have the CLC409 or H1100 chip.

(3) The 'blue' caps are on the early version boards only, as they were a part of the upgrade for the brightness blanking problem that was also on the early version boards. Don't try this fix yourself, this is the protect circuit, and much care should be taking to prevent tube damage if the procedure is not done properly (that I won't provide). Also, much care must be taking when servicing this area of the board. This board SHOULD NOT be used for random modding, it has special interferance tweaks that were necessary for FCC compliance.

(4) If resistors R13 and R28 show burn marks at the solder contacts, replace capacitors C1 and C26. That burn is an indication of high hour use, and C1 and C26 would most likely test leaking. The value is 22mfd @ 100 vdc. You can replace with the same value, but use 105c (high temp) 160 Vdc caps.

(5) Coils L8 and L9 are 100 uh, and would more likely show indication of heat damage if the same is seen on resistors R13 and R28.

For further maintenance of this board, make sure that the solder contacts on resistors R13 and R28 are resoldered. Also use denatured alcohol and a regular tooth brush and saturate the board with the alcohol, and then lightly brush the component side of the board with the brush. This cleaning will remove any oxidation and other crap from the board.

Let the board dry for at least one half hour before using. the alcohol should dry in a few minutes, but it may still be wet in some crevaces.

And make sure to not consume the alcohol :) !

Count me in too!!!!

This is going to be another really great thread.

i will be following and seeing what i can do

thanks

Wendy

Hello Mike

I have completed the electrolytic capacitor replacement, Resoldered
the resistors and put a CLC449 chip in place of the CLC409 on all 3
neck boards. I still have 3 CLC449 chips for a future job on the VIM.
Now a friend is wanting to do these mods on his M8500 too. So I
need to order at least 6 more CLC449s. Digikey lists them as obsolete
and not in stock. Allied and others do not even list them. Would you
know where some CLC449s may be obtained? National's site has a
cross reference indicating the LMH6207 should replace the CLC449.
But Digikey and others do not list those either.?? I would be willing
try the LMH6207 in spite of the lower bandwidth spec of 750Mhz. vs
the CLC449 bandwidth of 1.1Ghz. if I could only locate some.
Any info would be appreciated.

Thanks

DH

Hello All

A correction for my previous post. It is National Semiconductor
op amp part # LMH6702MA instead of LMH6207MA. Pin for Pin
Replacement for the CLC409AJE and CLC449AJE op amps.

DH

DH,
try this site:

http://www.future-active.com/results.asp?ddCountry=USA&ddCurrency=USD&txtSearch=CLC449AJE

I'm not sure if you'll be able to order small quantities, but give them a try. I've not tried the new LMH chip, they're new, being the replacements for the CLC449 chip. A sub company manufactured the CLC line for National, somehow Nation is no longer connected with them.

There are three or more CLC449 or 409 or H1100's on the VIM module. We'll get to that board later.

I'll place an order for the new chip. National had made the claim (before it was distributed) that it would be a better chip.

If anyone was willing to buy a few of any IC's at once and had trouble finding them, try 'Smith and Associates'. The are chip locators and have found many parts for me I could not find anywhere else. Many things obsolete here are still widely used in China and Russia.

Thanks Mike for starting this thread as I have always wished the 8000/8500/9500 would get the same mod attention the ECP has enjoyed.

Originally posted by mp20748
DH,
try this site:

http://www.future-active.com/results.asp?ddCountry=USA&ddCurrency=USD&txtSearch=CLC449AJE

I'm not sure if you'll be able to order small quantities, but give them a try. I've not tried the new LMH chip, they're new, being the replacements for the CLC449 chip. A sub company manufactured the CLC line for National, somehow Nation is no longer connected with them.




Nope. I went to the link and tried to order and the smallest quantity here is 95 pieces. You also have to call to get pricing. I remember some months back that Mike and I were talking about this, and we found a replacement that had the same bandwidth, but I think slew rate was different. Mike, do you have a memory of this? Is this what was different? It's been about 6-8 months ago and I just can't remember what all we discussed.

Analog Devices is crossing it to:

http://www.analog.com/Analog_Root/productPage/productHome/0,,AD8009,00.html

With double slew rate, but not the same bandwidth.

From Texas Instrument perhaps we can use this 1.2GHz, 4200V/us type:
http://focus.ti.com/docs/prod/productfolder.jhtml?genericPartNumber=OPA685
It looks like the best one for the job.

Unfortunately this 2GHz, 9000V/us is a dual type so not pin for pin compatible!
http://focus.ti.com/docs/prod/productfolder.jhtml?genericPartNumber=THS3202

If anyone is interested I have scanned an old service manual for the Marquee 9000 projector.

Kind regards, Claus

Guys,
the National Semiconductors true replacement would be "LMH6705" but it has not been released yet, so they are recommending the "LMH6702" as the direct replacement. But I'll not buy that...

Jay,
we were looking at the Max (Texas Semiconductors) replacement. I had the number but cannot find it now.

I still have a few CLC449's in my parts bin (not for sale). Give me about a week and I'll verify a properly functioning replacement.

In the mean time if you have the H1100 it should do fine, but should have a slight HF roll-off somewhere in the upper range/band of frequencies.

Mike I have some of the 449's (not enough to go around), but I want to be sure that what we are talking here (and in the furture with this thread) will also work fine with no problems on an 8000 model?

I ask because the 8100 parts board I am looking at has the metal Q12 (can't find a Q19) signifying an 'early' board, yet it does not have any blue smt caps. Are all the boards you will be discussing compatible?

------------------------------------

Just a suggestion, but perhaps people wait until near the end of this project and compile a partslist. Then one kind person act as 'the gatherer' and order in the parts since many will have minimum amount requirements. Work out a total on them and sell off kits to do all these mods.

Mike, I think I found what we were looking at...it was the Maxim (formerly Dallas Semiconductor) of the famed ECP Dallas chip.

Here is the spec sheet on a MAX4224 which they say is the replacement for the CLC449.

http://www.maxim-ic.com/quick_view2.cfm/qv_pk/1718

Anyone care to comment?

Tester,
you have a very early board. Q19 should be to the right of Q12 (one inch apart). And yes they'll work fine. I got this info from one of Christie Digitals engineers years ago, and I've been using them in the 8000 boards ever since. if the board you have has the metal (TO-5) transistors, and does not have the blue caps, it's probably of the very first ventage. Does your boards have the two white FCC ground connectors?

I'm not sure of the capatibility of the boards, but do know that the later boards will work in any of the previous Marquees.

Jay,
the MAX4224 is a six or ten pin IC, while the CLC449 is a 8 pin chip.

Edit: the MAX4224 does come in 8 pins: MAX4224ESA-ND. Digikey does not have them in stock. I'll do a search for them and order some for evaluation.

Jay,

The Burr-Brown (TI) part really have MUCH better spec than the Maxim part. You can order 3 pcs. as samples from their homepage. I did that yesterday.
The TI part is better than the CLC449 on some parameters, but not on all.

Kind regards, Claus

PS. The protection double diode on the input is adding both leakage current and capacitance to the video signal. Have you tried to remove it?

My local Farnell has 6 pieces of NSC CLC449AJE. Price is US$15.94 each and if order more than 25 pieces it will drop to US$10.90 each. It will take 5 days to re-order.
Let me know if I can help.

CH Yeow

Originally posted by Claus_LM
Jay,

The Burr-Brown (TI) part really have MUCH better spec than the Maxim part. You can order 3 pcs. as samples from their homepage. I did that yesterday.
The TI part is better than the CLC449 on some parameters, but not on all.

Kind regards, Claus

PS. The protection double diode on the input is adding both leakage current and capacitance to the video signal. Have you tried to remove it?

Claus,
I have not tried the OPA685/AD8009 on the neck boards, but I've tried them both on the VIMs. If using on the VIMs the circuit would need redesign, because with the AD8009 I could not get any gain, and with the OPA685 I got too much. But the neck board would not have been my choice to TEST an IC in (tube damage). Though all three chips on paper should have the same gain performance, mainly because the specs are for 150 ohm load, and they are all double gain chips. However, the Marquee video chain is directly coupled after the input coupling cap to the CRT's. Which makes this design a little more complicated than the stand-alone chip in testing.

I've also wondered about the protection diodes, and find it not a good idea for such circuit (video), but never considered playing with that circuit, plus most chip manufacturers put the same diodes in the chip. There are no other form of clamping circuits in the Marquees video chain (agc, limiting, etc), that's what I really like about the design of the video chain, it's very dynamic and direct. The thing that I have also noticed is that on the input section of the VIM, there are two protect diodes on each of the RGB inputs, if you examine that circuit you'll see that great care was used to place those diodes there for protection, while some consideration was giving for maintaining the proper impedance at the BNC. in other words it's not just a basic circuit with protection diodes that makes the proper impedance at the BNC's secondary. so I think that inspite of the protect diodes, some attention has been giving for a truer 75 ohm impedance load at the BNC's on the VIM. Maybe that's something we should look at, maybe there's a better diode that could be used in that circuit without hinderance, because we know that it's doing some hinderance, but on the other hand, if we remove those diodes, what could happen if there was a spike on the input of the VIM or the neck boards - The Marquees have very high power neck boards, there are six Motorola high power RF transistors on each neck board that drives the tube...

Thanks Mike

I was wondering! Have you looked at the overall voltage level in the video chain through the projector? With your MP-1 (I'm still dreaming about it) you could amplify the video signal at the very start at the video chain, and then we could get a better S/N level through the projector and then lower the gain at the very end in the projector (eg. at the neck board).
Or perhaps Electrohome made a heavy gain at the VIM just to follow the standard input video specs. Maybe it was better to transmit a much higher voltage range and then lower the gain at the VIM. This should prevent the noise being amplified.

I have no idea on how close the video signal is running to the rails, but perhaps they are way off.

Have you tested WHY your MP-1 makes such a difference with a higher voltage swing? Is it because it is bringing the overall video voltage swing closer to max (without raising the noise floor) or what.
Have you tried adjusting the gain up at the MP-1 and then down again at the VIM?


What I'm really going to look at, is lowering the noise floor. With the internal test pictures (all white), low contrast and looking into the tubes, they are so full of picture noise.

I think this is going to be a very exciting thread :)

Kind regards, Claus

Claus,
you raise some very interesting points, I'm glad that you're on board.

I like your idea of lowering the noise in the video chain. I plan to rig up a Marquee just for this, and maybe we can get some of the other masters in the electronic field to join in. There are some very smart people on this forum, but most of them are so humble that they rarely contribute much in these discussions. But I've had the pleasure of speaking with some of them by email and/or on the phone. they're here on AVS, and I would like for them to join us in this. How bout it guys! - Tom Rosbach, jcmccorm to name a few... I'm not an engineer, just a two bit technician, and guy's like this have been very helpful to me in their observations.

Actually the MP-1 works well on a very low output signal. With the jumpers removed, it's in the unity gain state (0.7v in/ 0.7V out). Adding the jumpers puts it in the boost state, and that boost is only 0.7V in to a 0.76 volt out. So there's very little boost, and most people could not notice a difference with the jumpers in or removed. It's doing something special with the signal, maybe I should redesign the whole video chain with the MP-1 design - Hmm! I once had the Mp-1 running at a higher output, it made a Marquee look like a large TV monitor, it was very bright and clear, and that was before clipping. But I thought it would be a little too much on the tubes.

Since I've not tested the Max chip, I've placed some on order (thanks Jay). And for some reason I believe that the Max4224 will probably be the best cross. And maybe National Semiconductor will get that exact cross to production.

I didn't want to get into tweaks, but again, if it comes up let's deal with them with very open minds. let's just avoid the ridiculous...:)

And Claus, if it's just you, myself and a few others - Let's do this...

Thanks Mike, I though you where ignoring me ;)

I remember reading about the Marquee wanting 0.9V to perform best. It was either you or Tim. It should produce a more "bright" image, but the whole idea is to lower the control to give the same picture as before, I was just hoping the "amplified" noise would go down with it.

Of course I hope KBK, Tim or Chris would give away some goodies, but I doubt it.
One think I would like a comment on is regarding capacitors and their rated voltage. I once had a very clever college and he told me (said not many people new!) that capacitors only have their rated value very near their rated voltage.
To give an example. Bang & Olufsen is using 10V capacitors on supplies instead of 6.3V. He told me that this is nuts, because a 100uF/6.3V capacitor is near 100uF, but a 100uF/10uF is nowhere near 100uF at 5V! In all the capacitor mod thread I read, I see people changing to much higher voltage and temperature. The temperature is making sense, but perhaps the voltage is not! Of course there need to be "room" for ripple, but that is hopefully a very small amount. Actually that is what we are trying to avoid, but in the power-supply itself it can be big, but not in the decoupling of the power-supply around the board. I just don't know!

I don't even know what parts are on my 9500LC. It is still in its flight case with its melted C-element. I think I can be see the different mods looking directly into the lenses! Then all setup issues is removed from the puzzle.

Kind regards, Claus

Tim stated to me the Marquee likes 1v input to look the best. I was going to buy a signal amplifier from Extron but that would necessitate either buying a few hundred dollars in new cables or adding on adapters front and back which would probably degrade the signal. If there was some internal mod to boost this then I would like to see it! I got in some tubes from an 8000, but only one made intact it and was burned (slightly) contrary to the sellers inspection.
So Mike if it would help having a tube to experiment on for this thread that you can fry or make a discovery to help us all, I will ship it to you since I will not be using it. Don't know if a single tube would help you or not.

you have a very early board. Q19 should be to the right of Q12 (one inch apart). And yes they'll work fine. I got this info from one of Christie Digitals engineers years ago, and I've been using them in the 8000 boards ever since. if the board you have has the metal (TO-5) transistors, and does not have the blue caps, it's probably of the very first ventage. Does your boards have the two white FCC ground connectors?
-----------------
It has a white ground connector by the heatsink and then 2 white wires, one with a tube connector (right near the center of the board) and the other with a standard spade type connector right by the wire harness. But this is just one of my spare boards, not my actual one in the machine.. Have not gotten around to pulling them out yet so I have no idea of their vintage. It was just odd because this board does have on the cl449a IC.

p.s. found Q19. It is 8 pin smt. Was looking for the same wire-and-tube assembly like the other one.

Guiy's
maybe Tim will join us in this discussion, and I do agree with him that the Marquee prefers a higher voltage than 700mv on the BNC's. And maybe KBK and Chris will also chime in on the mods and tweaks. I only hope we stay away from discussion on power cords and bat wings and such :D

I have several tweaks and mods myself, but for the most part I would only want to discuss upgrades and certain fixes.

The goal with a properly tuned Marquee is to find the right signal level on the BNC's, that way you'll be able to keep the contrast at closer to mid point. because when you increase the contrast, you also bring up the noise floor. By increasing the contrast you bring up the noise that's in the projector. Claus mentioned this:

"you could amplify the video signal at the very start at the video chain, and then we could get a better S/N level through the projector and then lower the gain at the very end in the projector (eg. at the neck board)."

Ok, enough on that, now let's get back to the neck boards. here's a little test to check the boards for those who can't remove and inspect them.

Turn on the projector and let warm up for 30 minutes or so. After warm up, follow this procedure: Press the '#' key once and then...

Press 'UTIL' on the remote, press 1, and then press 6.

Select number '3' (47.71/60) and then scroll thru all three colors by pressing 'Color' and then 1 (red), 2 (green), 3 (blue). you should be seeing all three color grids being displayed on the screen, now move very close to the screen. If the projector has a reasonable setup, all three color grids should be tight in the middle of the screen with no flare or smear to the right of the grid.

Now, go back '#' (Util, 1, 6) and select 5 (89.30 / 70) that should put you in the high band, scroll thru (color - 1,2,3) all three color grids should display, now see if the grids smear to the right (in the center only).

Do the same as above, but select '6' (126.84 / 60) as the final test.

post back your results...

I don't have the exact same frequencies, but have no smearing side to side. The green does have some smearing downward in the central area.

Mike,

You PM inbox is full. Could you please make room for a mail?

Best regards, Claus

Tester,
the smearing is more noticeable in the upper frequencies (high band, above 60 Khz). Did you use an internal frequency above 60 Khz?

If so, and no smear, your projector would more likely have either the CLC449 or H1100's. the CLC409's were no longer used in the Marquee neck boards after 1995 (my guess).

If your projector is pretty well aligned, you could also try running AVIA at 720x480P (native rate of the DVD) and look at the 'sharpness' pattern and the 'multiburst' patterns for detail. Compare the detail of these patterns using a second resolution of 1440x960P.

Any well aligned Marquee should be able to handle both of these resolutions quite well. The 1440x960 using the above test patterns in AVIA should let you know if there's a problem in the video chain. And this test would also include the cables.

Claus,
it's now clear!

Hi Mike. No smearing at all from my pj. Blue is a teeny bit underfocused compared to the other 2 at the highest frequency...at least this is where it was most apparent, but no streaking. Mine is a M8110, built Nov 1996 so based on your above comments, I probably have the later chips. I assume it is recommended to go ahead and replace with the CLC449 (or eq) anyway?

I did spend an extraordinary amount of time on stig and flare...like about 7 hours by the time I was happy with the results. Is it possible to get a clean grid with the lesser components if you spend enough time?

Jay,
I think you're in good shape on the neck boards. I have in the 9500 here two good neck boards, and the third has a very slight smear to the right. I can see it on the scope but it's hard to see in the image. I like using the test patterns in the high band because that's where the chips really slow up (bandwidth) switching from black to white on the resolution patterns.

When running the resolution pattern (multiburst) in AVIA, the pattern should be equal in brightness over whole pattern with a good card. If the card is uncapable of handling the resolution, you'll see a brighter area on the right of the screen, and less (or no) black lines. A properly functioning card should show that pattern with the white lines and the black lines in place, with equal brightness.

And yes you can get a pretty clean grid with lower quality components, but the important thing to look for is true black lines on the far right of that pattern. Black lines on the left is not a problem (low frequency), on the right is the high frequency, and it shows the speed of the system.

The CLC449 is only necessary if operating in the high band somewhere around 70Khz (1440X960/72) and up. But if yiu have the H1100 they should be fine below that range.

Thanks Mike. 1440x960 is about the minimum I have been runniong lately. I was using 1280x720, but when I spent so much time redoing stig & flare, I found I could go higher. I set the unit up with a 1920x1080 signal. I'll spend some time with the AVIA test patterns this weekend when I bring the disc home from the office. Then I'll let you know the outcome of that.

Originally posted by Jay Wilson
.... I set the unit up with a 1920x1080 signal. I'll spend some time with the AVIA test patterns this weekend when I bring the disc home from the office. Then I'll let you know the outcome of that.

Jay,
that's a very powerful nugget on Marquee setup (this is secret stuff). Set it up above the resolution of use, copy down to your desired resolution - touch-up. The Marquees will have much better color tracking when setup in the high band, plus you'll tame the dots better. Kudos!

And if you really want it to sing, start your initial setup at 31 Khz, and then go above your user res...

Hmmmmmm.

Looks like a reset will be in order this weekend. i'd nver heard (or don't remebering) of starting low, then going high. I always start high and go low.

Well, I guess you earned your keep for another week :D

Don't suppose this MP1 has anything to do with the aesoeme picture I get huh? :D

Hello Mike

All 3 neck boards had severe streaking. From light areas into dark
and dark into light. On both low and high bands. After replacing the
CLC409s with CLC449s and the two electrolytic caps all streaking
was gone. I reflowed the solder on the big resistors even though I
had very little discoloration from heat. The video image was also
cleaned up a little, less noise in the picture.

That is a worthwhile refurbish and upgrade. I am now ready for
your next chapter in the Marquee maintenance thread.

Another question. I have 2 more CLC449 op amps to do 3 neck
boards (friends PJ). One neck board has a H1100 while the other
have CLC409s. I will replace the CLC409s. Which CRT would to be
best to get the neck board with the H1100 op amp? Blue?

Thanks again for sharing this information.

DH

What a great thread! One note of caution on Video Amp replacement.

Specs only tell part of the story, and sometimes even direct replacements aren't. So, if you don't have the test equipment, like a really fast oscilloscope, stick with replacements that the experts like Mike have tested and proven. Current feedback op-amps like the CLC (Comlinear) parts, and the Burr-Brown and Analog Devices parts, are all optimized with somewhat different feedback components. The (-) inputs are very sensitive to capacitance to ground, and you can easily turn these amps into oscillators.

Mike's advice to stay away from experimentation on the neck boards is sage indeed.

Regards,

Now, that we've had a chance to look at the neck board, and was able to examine it physically, or using test patterns - let's keep going...

If all was fine from your observations and/or corrections if necessary, let's move on to the VIM. The same IC's on the neck board (U2) could be found on the VIM. You should look for the same chips (CLC409, H1100, CLC449). The same upgrade applies here on the VIM. The only difference would be that the CLC409 may not cause smearing if on the VIM, you would more likely have a problem getting a very clean resolution pattern, or the image would suffer from a weakness in detail, either way, it could have a noticeable effect on the image. If the neck boards tested ok, or have been corrected, you'll now need to perform a visual check for that IC (CLC409) on the VIM, it's much easier to get to than the chips on the neck boards.

Next, you'll want to resolder the mini RCA jack's on the PC board. And clean both sides of the VIM with denatured alcohol and a tooth brush.

After this, we'll get into cleaning the power supplies and the rear vent cover. I hope to post some pictures on this (make sure you have a can of compressed air handy).

OK.. in my 8000 I know for sure the board I did the resistor mod for the anamorphic squeeze was from an 8500 so there may be others..

Now I have pulled out the VIM board. I also have a spare VIM from an 8100 I just bought. The one I pulled out looks completely different than the spare. 1st, the one in the machine is FAR cleaner.
But I will list some of the other differences I can see offhand. The one I am using has H1100 chips, the spare has cl449a chips. The spare has a socketed U41 SYNC2.0 chip while I can find no U41 on the board I am using. The board I am using has 3 OMRON 1474Y1 (RL4,6,9) and the spare none. Anyways.. there are many differences.
I am using 50-2005-03P ISS
spare is 50-2035-02P ISS.2
Now is it worth cleaning off all the pet urine off this spare I bought and using it, or just take the 449's off of it and put on my current board in place of the 1100's? Do any of the other parts listed (or not listed.. there are quite a few) make a big difference? These boards look not much alike except for the 6 blue boxes and the large green cap.

Tester,
your spare board is a later version. It has a wider bandwidth video chain. It also has the blue gamma chip (U22), that might not be on your other board, because some of the early 8000 VIMs did not have the gamma circuit. There are several other technical differences, and they were mostly added for higher scan rate use (super computer presentation), and may not have any benefit for HT. You should be fine with both, but if you like the high band go with the CLC449's.

Your spare is a better board...

It would be cool if someone could make a DVI input board for the Marquees, then I could go straight from my HTPC to the projector with a cable...

Mike P is this doable ?

Stephan

Guy's
somehow I did not receive the samples (CLC449 replacements). I wanted to test them and post back the results. I'll purchase the chips and we'll continue with the maintenance. I want to make sure the replacement chips will work before we go to the VIM (video input module).

And before this maintenance is over, I plan to announce my success with repairing the HVPS.

Stephan,
I can check on that, but I'm sure DVI does not break out to analog. If it does, I'm down - let's go for it...
Stephan,

Hi Guys

It is for sure possible, but I don't think it will improve things. If we are talking "normal" DVI we could use a Silicon Image part or perhaps a TI part like this one: http://focus.ti.com/docs/prod/folders/print/tfp403.html
It is possible to get a part to transform it analog RGB, but I cannot remember where I found it last. It will be more or less a glue-less solution combined with Mikes knowledge about low-noise power supply it could be good.
I think the problem is the bandwidth and cable length.

Another much better solution is the new upcoming Inova chip-set. They can transmit up to 50 meters using twisted pair cable.
Take a look here:
http://www.inova-semiconductors.com/products/gig_ddl_link.htm

They can interface glue-less to a Silicon Image DVI receiver on the input.

Kind regards, Claus

Stephen,

Already available....Geffen makes a DVI to SXGA (analog) convertor in the "Extend-it" product line. Retail: $400. If you want one, we're dealers for Geffen if you can't find it locally (and you probably can't). Shoot me a PM if you're interested.

Mike, I haven't gotten my samples either. When/if I do, I'll post. If you haven't gotten yours, I'll send you mine.

I guess I was thinking more along the line of somehow tying in the computer video cards DVI output ( or out like on the centerstage DVI ) directly into the digital video part of the marquee, and bypass the analogue to digital portion of the projector ?

I guess I dont even know if a Marquee's video chain is ever really condsidered digital... so I am may be babbling.

Stephan

Its not - stays analog right to the neck bds. It's everything else that digitaized.

Guy's,
the MAX4224ESA (Dallas Semiconductor) chips arrived on Friday. I've already installed them on a neck board, tested and viewed. So far this chip seems to be a perfect substitute. Could not even notice any difference in the gray scale.

This chip seems to be a good replacement for the now obsolete CLC449.

That completes the neck board, now we'll pick back up on the VIM...

Cool! Great news (Because I am already setup to order from them :) ) I hope that means mine are close by as well.

On to the VIM!

Thanks Mike

I placed an order for 6 of these from Digikey. I will use 3 of them
on neck boards dated 24/93, 26/93, and 07/94. The latter has a
H1100. All have the blue cap circuit upgrade. Streaking on these 3
boards is not bad, only a little on the low band, more on the high
band. I will post results after replacing.

Thanks again

DH

DH,
you're on the right track, but also replace the two electrolytics (C1, C26). They are 22Mfd@100Vdc.

Replace the caps without removing the PC board from the heatsink. That can be done from both sides of the heatsink. removing the PC board from the heatsink can damage the leads on the RF power transistors that are screwed to the heatsink.

Once we're finish with the VIM, you should not have any smearing at all, and if you do it would most like be coming from the source on the BNC's.

There are several other things that can cause some streaking, if not on the source. But they would require a more technical approach, and I'll not post that on the forum, the repair is very involved. And I can do that repair if needed.

Donald, how did you get from Digikey? 100 minimum count and backordered to March 3rd (I think that was the date). ... You can now order directly from Maxim/Dallas on their website for cheaper.

Hi Tester007

I checked. My order is still pending. I must have missed a
keystroke somewhere. I just thought I had finished the order
with Digikey. Thanks for the tip about Maxim/Dallas. I will
cancel the Digikey order and go for cheaper.

DH

Mike,

I have a 98 or 99 (can't remember right now) EH 8500 Ultra. Will these or any other mods for the 8500 benifit me or have they been done already by the manufacturer since it is a newer model? Does it still need the anamorphic squeeze mod done to it?

Thanks,
Craig

Craig,
you should be good. Unless you have an early version board in that ultra.
Read back over the thread and do the test for smearing.

You may need the anomorphic sqeeze if it has not already been done.

MP,
Can you describe what the anamorphic squeeze mod is. I have read a little bit about it on the forum, but I don't know the details. I can't get the search function to work here. It just hangs.

What does it improve and how much do you charge? I'm assuming because the PJ was designed for computer graphics and such, that we need this mod to improve the PQ of movies.

I have a 1995 Marquee 8000. I'm going to save this thread when your fiinished and do as much as I can myself when I have time.

What is your estimate of the average time it would take to do these mods? A full day, or more?

By the way, thank you for sharing your time and talent with us.

Brad

Originally posted by BLS
MP,
Can you describe what the anamorphic squeeze mod is. I have read a little bit about it on the forum, but I don't know the details. I can't get the search function to work here. It just hangs.

What does it improve and how much do you charge? I'm assuming because the PJ was designed for computer graphics and such, that we need this mod to improve the PQ of movies......


Brad

Brad,
that mod is commonly known as "vertical squeeze mod" it decreases the vertical size for 16:9 operation. I think you're right that the circuit was probably designed more for computer presentation.

The mod requires the replacement of 3 resistors on the vertical board (R704, R804 and R904). These resistors are 30K ohm, they should be replaced with 56K ohm resistors, and it's best to use metal film resistors @ 1 to 5% tolerance. I use 1% 1/4 watt from Digikey.com:

Part# 56.2KXB

Mike,
It sounds like the anamorphic mod and the other mods you have been sharing need the same skill level. Common sense, manual dexterity and a basic knowledge of electricity and safe work habits.

My screen size is 47.25" X 84"

With my 8000 on almost all aspect ratios the horizontal size is 100 (Maximum) in the size menu and I am able to adjust the vertical size to match the correct aspect ratio.

Do I assume correctly that I don't need the mod if I'm able to adjust the vertical to where it supposed to be?

BLS,
so far, I'm sure the vertical mod is the only mod I've mentioned on this thread (I don't intend to get into mods). We've been discussing maintenance and upgrades. And most of the upgrades (chips) come from my notes, and we've not really got into much maintenance. The "vertical squeeze mod" was not commonly used back then, it was used only on one install that I could think of, and that was a 16:9 screen where the projector was used for a High def source. It seems to much needed for HT use though...

As long as they improve picture quality and reliability, you're the man! I'll call them whatever you want me to.

MP, it's all new territory to me. This is my first PJ. I have been dreaming about it for years. Iv'e got tons of patience. So if you only want to help with maintenance and upgrades, I'll be grateful for that. It will be more than I started with and that's why I'm here. Maybe someday I can offer something useful on the forum.

Brad

Brad,
I hope I didn't make you feel out of place, because everyone that post has something useful to contribute on the forum, you're only asking questions. And if no one asked questions on the forum, there would be no purpose for the threads.

My goal is to stay focused on "upgrades" and "maintenance", and this thread is certainly open for discussion on mods and such, but MY intent is to stay the course, and I'll respect any comments between others on mods. And there are some very valid mods out there for the Marquee. It just that I'll not go into discussion on the ones that I know of.

For those who are into modding, I would highly recommend KBK. I've seen his workmanship, it's class "A" also because I'm against the average Joe with a Radio Shack soldering iron, attempting to transform one of these precious products into something it's not capable of doing. And it's been good hearing from some who would like to have their neck boards upgraded rather than try the procedure themselves. And the reports of the difference that's been made once the upgrades have been performed. keep in mind, I've not suggested something that the manufacturer did not do, I've suggested what the manufacturer did, but only without replacing the entire board.

I'm waiting to get back out into the shop, and when I do, we'll finish with both maintenance and upgrades, but I hope to post pictures on cleaning the dust from the various parts of the projector, and a few other things.

I'm against the average Joe with a Radio Shack soldering iron, attempting to transform one of these precious products into something it's not capable of doing. And it's been good hearing from some who would like to have their neck boards upgraded rather than try the procedure themselves.

I am one of the average Joe's who did the neckboard fixes himself. I needed to get rid of the horrible streaking cheaply as possible. I went 2 for 3. All I can say is unlike branches of the goverment, 'two out of three' is bad. One of the boards never worked properly after the fix, and I even had some help from someone with good soldering skills. Luckily, I was able to get a new board for free from the same place I got the PJ.

Anyway, I just wanted to backup what Mike said and be a voice of caution to any DIYers. The improvements are great, but so are the risks. If I had the $, would totally have someone else do this kind of stuff for me in the future.

I'm sure a lot of you folks would agree. There is nothing like the feeling of fixing something yourself. I'm 42 years old, so I have made a mistake here and there over the years. I agree it's risky, but during those years I have developed a sense of when I'm out of my league. Then I get to admire the skills of someone else.

I may start these procedures and find out i can't do it. I'm willing to try though. I've been into electronics one way or another most of my life.

Nathan, what do you think went wrong?

Brad

The board would work for a while, then start flickering, and then stop all together. I am not much of an expert, but it sounds like a heat issue to me, maybe a poor solder joint expanding? I came across the new board pretty soon after that so I didn't bother trying to fix it.

I am sure most of these things can be done with only moderate skill, but now I am a little hesitant. Especially since the thing is finally mounted, working, looking great, and I would rather watch movies than mess with it for a while. :D

Im all confused about most what your all talking about. My friend and I bought 2 EHome Marquee 8500's but the dept we bought them from at work says that they were going dim. we shined a flashlight into the tubes showing that the tube wear wasn't all that bad. We used that webpage as a guide that shows pics of various conditions of tube wear.

We would like to do everything your talking about here but have no clue what your even talking about. Power grid voltages, replacing all sorts of chips, getting electricuted. All sounds fun but we need someone to do this for us. Is there anyone qualified around Denver CO to do this?

Stylinlp,

Those Marquees sound like they are suffering the dreaded heater overvoltage problem. Down on the motherboard near the back of one of the outside CRT's you will see a small white 2 pin plug labelled P14 - using a multimeter set to DC voltage, check it. It should not exceed 6.45v - if it does, chances are high that the tubes are trash (but re-buildable through VDC). The LVPS needs to be upgraded to prevent this from happening again.

I have been saving the tips, fixes and so on for when I take mine down from the ceiling.

Can a do it yourselfer perform the LVPS upgrade?

Can you give any detail here?

Brad

MP,

I know that you want to avoid discussion of mods, but please clear one thing up. You described the vertical squeeze mod as:

"The mod requires the replacement of 3 resistors on the vertical board (R704, R804 and R904). These resistors are 30K ohm, they should be replaced with 56K ohm resistors."

On my EH 9501 LC, the person who sold the unit to me was adamant that a different resistor swap was needed. His assertion is that resistors 706, 806, and 906 should be changed to 490 ohms (leaving 704, 804, and 904 unchanged.) Is this correct?

Ron

Ron,
I'm looking at the tech bulletin on this from the manufacturer

"VERTICAL SIZE MODIFICATION ON THE MARQUEE"

In this tech bulletin there is a reference to R706, R806, R906. And it states that 'IF' the existing resistor is 430ohm, replace with 470ohm, otherwise replace with 680ohm.

In other words, the resistance needs to increase and the existing resistors must first be removed. Some of the older boards used 430ohm resistors and they should be replaced with 470ohm resistors. Newer boards used 470ohm resistors and they are to be replaced with 680ohm resistors.

In this application the manufacturer recommended replacing the RAMP resistor. In the popular (R740,R804, R904) we're replacing the FEEDBACK resistor on the buffer amp.

the replacement of the RAMP resistor was the fix for PC boards:

50-2023-03P and 50-2023-03P these boards the tech sheet indicates that you replace R706, R806 and R906.

On PC board 50-002020-02P you replace R704, R804 and R904.

What I've found is that you can replace R704, R804 and R904 on all of the three boards. And the replacement of the RAMP resistors is not necessary. So the simplistic way is to replace R704, R804 and R904, mainly because the tech sheet has too many options to achieve the same thing.

------

Brad,
I would recommend that you have the LVPS done by someone who has sucessfully done it more than once. Many are simply adjusting the pot to achieve the proper voltage, but are not replacing the pot, and if you make the adjustment without replacing the pot, it'll eventually flake out again. and if that happens the tubes are at risk. Also, it's important to make sure the repair is done right so that you don't accidently allow too much voltage on the tubes filaments, hoping to the voltage set to the right level.

My opinion - have it done by someone who's very familiar with doing it.

I have a digital camera on order. When it arrives we'll pick back up on this thread.

We're still going to cover the VIM, and from there we'll look at the rest of the projector.

Then, you'll get a chance to see why this basic chassis design has not been changed in the past ten years.


You'll need the following for the maintenance:

- Denatured alcohol (about $3.00 from home Depot).

- Standard tooth brush.

- Standard bed sheet.

- Vacuum cleaner

While I getting parts for my dead HVPS I took out my neck boards and to my surprise they seem to be the new type
my PJ it a 1995 8000

to I took a snap of it>>>
http://www.avsforum.com/avs-vb/attachment.php?s=&attachmentid=7044

Next I took out the VIM
It had H1100 op-amps and a junk op-amp for red
I had cooked red a long time ago and just used what I had on hand.

here is the VIM ready for rework I had order the max4224
http://www.avsforum.com/avs-vb/attachment.php?s=&attachmentid=7047

After removing the old chips off of the VIM
http://www.avsforum.com/avs-vb/attachment.php?s=&attachmentid=7048

now the VIM with the New MAX4224 chips installed
http://www.avsforum.com/avs-vb/attachment.php?s=&attachmentid=7049

Sly, you're on the ball here. Thanks for the pictures. I'm on my second camera and I can't get the pictures right :(

OK,
we'll put off the cleaning until I get my hand on a camera that I can make sense out of.

But until then, we'll move on to the rest of the set.

So far I've received very good reports from the chip upgrade on the neck boards, and some have already replaced the chips on the VIM, with very good results. The CLC449's were used in the later version Marquees, they replaced the CLC409 and H1100. Both of these chips had a roll off problem and could cause smearing and slight distortion in the image when used on certain resolutions. The H1100 may not show a problem, and may not be a problem from testing, so the replacement of the H1100 is an upgrade option.

Next we'll look at the VIM (I was hoping to have the camera for this). EArly version 8000's did not have the blue gamma circuit on the VIM. It was included on all of the later models. That circuit is a simple op amp, and maybe we can add it to the boards that does not have the circuit, I'm sure someone could make that little circuit and make it available, it'll be quite easy to install.

I'll throw in a mod.
The first thing that we'll do on this board (after the chip upgrade) is a Sync Mod. This is the removal of the sync circuit from the green input (sync on green), this is very important especially if using component to RGB conversion, because that could cause TRI- level sync on the BNC's (composite sync on all three BNC's). The problem that this could create is that you could have sync on all five BNC's. That's not a problem for the Red and Blue, but by having this signal on both the 'H' and 'V' BNC's to include the green BNC can confuse the auto sync detect circuit that automatically detects what type of sync that's being feed to the projector. The auto detect circuit looks for sync on the following: separate sync on 'H' and 'V' BNC's, composite sync on green BNC, composite sync on 'H' BNC. This circuit worked fine before component sync conversion came along possibly placing sync on all five BNC's.

This is a simple fix, remove R152 (it should be located about 1" from the blue relays) on the VIM module. By removing that resistor, you remove the sync circuit from the green BNC. There is presently no present modern application for sync on green (that I know of), especially for HT use.

Let's look at this and then we'll move on to a few other things on this board.

Mike, if you can develop a circuit, I'd be happy to design and etch circuit boards, and even write instructions, for a drop-in module. Heck, help me source the chips, and I'll even build ready-to-install boards. I have printed-circuit-board software and etching services available.

Larry
www.fineelectricco.com

Mike does my PCB have the blue gamma circuit ??

http://www.avsforum.com/avs-vb/attachment.php?s=&attachmentid=7044

if it does I could draw it up in eagle.
it appeared that all 3 pcbs were the same
I did not keep track of which was which..
I just pulled all three out :rolleyes: I guess I should have marked them..

Sly, The gamma circuit would be on the VIM (Video input module) not on the neck boards.

The early version VIMs that did not have the blue gamma circuit, they had all three of the mini RCA plugs in line. The later version VIM (8000, 8110, 9000, 8500, 9500, etc) had the blue RCA plug slightly higher on the board and a little off to the right. The gamma IC on these boards is "U22".

Let's take Larry up on his offer to design and build the circuit for the early VIMs. But we'll need to see what the demand for them would be first.

OK the I see, that makes it easy there is a lot of room there.

I guessing the circuit has a fet in the feedback of a op-amp and uses the video voltage to control the fet.

I would think it would be easier to mod the op-amp that is there that way the video path will not have to be altered. and I do not think there is a need for a PCB or if one is made it will not have to be a pricey impedance controlled one.

so it looks like mine is the old one
http://www.avsforum.com/avs-vb/attachment.php?s=&attachmentid=7047
I keep on kicking my self we had a VIM go bad on a new 9500 and they threw it out. I should have grabbed it..

This will be tricky it will have to run at full band-with with out any ringing. does any one know what the Gama value is??
Any one using a HTPC it seems a lot easier just too add that value to the LUT in the video card. I think most Video cards have the option to add gamma tables

Mike

I will purchase 3 of the gamma boards, either assembled or kit.

DH

I just looked at the heater circuit on the LVPS
it is just using a lm117 voltage regulator

http://www.qsl.net/ok1duo/lm317.htm
cool webpage to do the math

the circuit in the LVPS has a 120 ohm as R1
and a 1k pot 0-1K for R2
that gives us 1.25v - 11.67V

if you cut the trace and add a 470 ohm resistor in series with a 20 ohm pot
giving us 470 ohm - 490 ohm
we would get 6.1458V - 6.3542V
This also will have less power in the pot.
a value of 489.6 for R2 should give 6.35V

This seems very simple replace 1k pot with 20 ohm pot
one cut and add one 470 ohm resistor.

What do you think?

save them tubes:D

What would happen if the pot became intermittent or opened completely? The output voltage would still rise. How about putting a 25-ohm pot and 100-ohm resistor in place of R1, and the fixed 470-ohm at R2?

Larry
www.fineelectricco.com

>>What would happen if the pot became intermittent or opened completely

it will never be more the 20 ohms unless it is burned
and it is just as likely to burn a fixed resistor
and if say the wiper fell off it would be 20 ohms
aka 6.3542V

it is very unlikely to fail.
The original circuit
would have a large voltage shift for a very small movement in the pot. so just a bump or thermal change would cause it to go out of spec. With this mod the whole range is safe. And is makes for very easy rework.
To move the pot to the other side would take a lot more rework.
I do not know why they made it that way in the first place
you are just asking for it to jump out of spec with such a sensitive setting. I guess that why they used a 25 turn pot
if they had made it like this in the first place they could have use a 1 turn with out any problems. I think this was ether a over-site or done so they could replace CRT's and boost service income.. Ether way it is bad practice to have an adjustment move any more then is needed. In fact I do not know why it is there at all they could have just used fixed parts set to 6.35V
the only thing I can think of they might have thought that a 2nd source CRT would use a different voltage??

Sly

I don't think that is true. I think Electrohome outsourced the power supply to the company in Illinios with a load spec. I don't think this power supply is just in the Marquee. I think it is used in many devices, not just projectors.

Instead of the pot, why not design a circuit, sampling output to use as feed back, to replace the pot. Remember that filament voltage drops to about half when the pj is turned off.

Walt

Why not just replace the pot with a fixed resitor?
-peter

Peter
"Why not just replace the pot with a fixed resitor?"

As age sets in the voltage will probably change and you want to maintain the 6.35, if possible, with a little adjustment.

Walt

But this circuit uses a voltage regulator, so unless the value of the resistors changes (not likley at all) then it should hold the output voltage regardless of the load as long as the load does not go out of spec for the regulator.
I am missing something?
-peter

Is there a mod or tweak or board/part that will:

a) Increase the light output

b) Increase the resolution rating up to 2200 by 2600 or 2048 by 1536 ansi pixels that you see written in the hi-end magazines for the Reference Imaging/Accurate Imaging "9500s".

Sorry if this is a stupid question or has been already answered.

Also, I was wondering if you experts have a list of products or services/mods that you could do for us newbies for a price and their benefits.

Originally posted by docajay
Is there a mod or tweak or board/part that will:

a) Increase the light output

b) Increase the resolution rating up to 2200 by 2600 or 2048 by 1536 ansi pixels that you see written in the hi-end magazines for the Reference Imaging/Accurate Imaging "9500s".

Sorry if this is a stupid question or has been already answered.



You'll have to learn how to read those magazines the same way you learned how to eat chicken -- don't chew and swallow the bones.

"Why not just replace the pot with a fixed resistor?"

you could but it would be hard to find a 489.6 ohm resistor;)
And they are a bit off any ways. you might be able to find
a set the would work it it should stay
but there is the spot for a pot and this way you can make it dead-on.

" Instead of the pot, why not design a circuit, sampling output to use as feed back, to replace the pot. Remember that filament voltage drops to about half when the pj is turned off. "

The circuit does have feedback it is the feedback part that make the voltage adjustment.

There does not seem to be any thing to drop the voltage when it is off/standby, so it must be dropping the voltage on the input it the LM117 voltage regulator.

The difference in turning the pot all the way back and forth after the mod will only change the voltage 0.21V
before the mod it would be 10.42V . And this is with easy to get parts of common values. And just one cut on the PCB. I am trying to keep it simple and easy to do. and should never have to be messed with again.
if it moves a few percent over time no big deal
a few percent of 0.21V is very small but a few percent of 10.42 and is a different story.

OK I just changed my pot to a 485ohm (measures 485) 1% 1/4 watt metal film. The other resistor measures at 120.5 ohms. Using the formula for the lm317 I should see about 6.28 volts.

With the CRT powered up but cool, I measure 6.16 volts on the jumper near the rear of the blue tube. I will remeasure in an hour and see what it is after everything gets warm.

Anyhow if this still hold to be the case that I see 6.16 volts,then I would suggest having a pot etc in series with a fixed resistor as sly suggested such that you can get a total resistance of at least 500 ohms. This should allow more range. I will probably add another 10 - 15 ohms to mine to get to 6.27 or so

-Peter

120mV is not very much I bet that is the drop in the wires from the regulator to where you measured. ??

Originally posted by sly
120mV is not very much I bet that is the drop in the wires from the regulator to where you measured. ??

This is what I was thinking as well. After two hours the voltage was measured to be 6.15 volts so it is holding steady (watched "Momento" which was a good movie). I will increase my R value to be 500 ohms. This should give a therotical 6.43 volts but if I still see the same 120mv drop/error, then I should be at 6.31 which is close enough to 6.35volts. I will of course report back when I try this out (Probably won't be until tueday though)

-Peter

Hi Mike,
I just got around to checking out my video neck boards. The red and green boards have the blue soldered on capacitors (looks kinda kludgey) but the blue board doesn't. The all have H1100's and the metal can Q12 and Q19.

Is it normal for the three boards to be different?

thanks,
bruce

Yes, that's normal. The Marquee has been in production longer than any other CRT projector, and in the last ten years there have been some changes made. The neck board was the item that had experienced most of these changes, but only the very first boards were a source of problems.

The board that does not have the blue caps soldered on it, would more likely develope the "blanking out on bright scenes" problem. But until that happens and if there is no noticable smearing in the image, leave it be and enjoy.

yeah, I've got bigger fish to fry first - just discovered that my filament voltage is over 8 volts!!!!!!!!

Good think that less expensive CRT rebuilder surfaced in another thread here - hope it works out that their quality is good ...

cheers,
bruce

Hi Guy's,

I've got a pic of my VIM below, the upper part shows "U22", but the
RGB jacks are straight. My VIM is layed out different compared to Sly's VIM.
So, to avoid assuming on my part, comfirm for me that I do have the
gamma circuit. Is this what you could call an early version of the newer
VIM?

The lower pic shows R152, this is the resistor to be removed to do
away with sync on green, correct? I just have to hear a yes from someone
before I take that off the VIM. If I'm wrong I don't want to test my skills
getting it back on.

Thanks for starting this thread Mike, your teaching this noob a thing
or two.

Bruce

Originally posted by bob2010
Hi Guy's,


The lower pic shows R152, this is the resistor to be removed to do
away with sync on green, correct? I just have to hear a yes from someone
before I take that off the VIM. If I'm wrong I don't want to test my skills
getting it back on.

Thanks for starting this thread Mike, your teaching this noob a thing
or two.

Bruce

Yes, take the resistor (200 ohm) out. And if your board has "U22" you have the gamma circuit.

Thanks Mike.

Bruce

if you can't see the images try this
http://www.geocities.com/hvps2001/


I thought I would start off by replacing bad parts on the outside of the potting. I will start off with the G2 control circuits. The resistors in mine were way off and they are carbon type (the worst kind of resistor) they get out of spec if they get hot and they also add white noise.
I think this might be what causes some of the faint snow that has been seen in the background by some people


Start by removing the bad resistors
you can see I have removed one already.
156324
http://www.avsforum.com/avs-vb/attachment.php?attachmentid=156324&d=1256590463
All 6 replaced. 156325
http://www.avsforum.com/avs-vb/attachment.php?attachmentid=156325&d=1256590463
Note the Air gap don't let the resistors touch the PCB. They need air flow for cooling. 156326
http://www.avsforum.com/avs-vb/attachment.php?attachmentid=156326&d=1256590463
Next I replace the input filter caps. The new caps are smaller so I scraped the solder mask off the trace and just soldered one pin of the cap to the trace.
Notice the burnt white cap..
http://www.avsforum.com/avs-vb/attachment.php?attachmentid=156329&d=1256592292
Now the next 2 caps and resistors and yes those little guys are 2 watt I rechecked when I got them and I even ran 20Volts through them just to test that they are 2 watts 156327
http://www.avsforum.com/avs-vb/attachment.php?attachmentid=156327&d=1256590489
Notice the air gaps on the resistors the leads help cool them and I have the caps as far as I can get. 156328
http://www.avsforum.com/avs-vb/attachment.php?attachmentid=156328&d=1256590489
http://www.geocities.com/hvps2001/

you will see milky white stuff on the left next to the metal.
is anyones working HVPS have this??

image for the post before this.

It's on most of them, even the working supplies. I gave up on trying to figure out what it is or where it comes from.

When I bring up the convergence pattern at 15.75Khz, it is jittery, mostly in the horizontal direction such that I see three vertical lines for every real vertical line. At any higher frequencies I get a stable pattern. Does this jitter indicate some flaky components? When I look at an NTSC input (say, from a VCR or DVD player connected to the decoder board), I get a pretty good picture with some horizontal jitter at the top of the image. (gotta get working on my HTPC!). Any suggestions?

Also, does the anamorphic squeeze mod also squeeze the overlay graphics (e.g. the slider bar that shows when you click on Contrast). My overlaid graphics are running off the bottom of the screen, and there are some convergence zones where I don't see the box at all.

thanks,
bruce

For those of you who wants to build up the gamma correction; here it is.


Does it make sense to build a gamma correction for green and red too?

Thanks I thought it would be.
it is primer for the potting
see SS4155 http://www.mgchemicals.com/products/silicones/primers.html

and I think the potting is/should be TSE3331 HV and Thermally Conductive
http://www.mgchemicals.com/products/3331.html

Originally posted by mp20748
I gave up on trying to figure out what it is or where it comes from.

There are some things better left unknown :rolleyes:

If someone can slap together a gerber and BOM I could look at making a small run of these. Would of course give free samples to all who help!

If someone can slap together a gerber and BOM I could look at making a small run of these. Would of course give free samples to all who help!

Hey Sly, I'm suprised someone has not asked you yet. What model camera and what resolution are you using for those pictures. Those are outstanding!

I'm going to guess at least 4 megapixels, probably more.

Also, did I get lost here? What parts and numbers are you replacing in these pictures? I can see your heading that states HVPS, but what are the part numbers you are replacing the bad ones with?

Brad

Originally posted by Tester007
If someone can slap together a gerber and BOM I could look at making a small run of these. Would of course give free samples to all who help!

A gerber? I'm thinking the BOM is a bill of materials list...but a gerber? What the heck is a gerber? Did I miss something?

a gerber plot, as in a layout for the PCB

Gerber is a file format used for PCB construction(and other photo etch processes). It contains the data needed to create the etching masks for the PCB. There is another file that is used for the drill data to make the holes.


Ken

OK...Cool...Never heard it called that.

Like I said before, I can take the schematic and make the PC board layout, make the file, and get boards etched and drilled. If needed, I can source the components and solder them and pigtails for PJ connection.

Here's a couple of screenshots of a PC board layout I made from a magazine article:

This one (www.fineelectricco.com/PCBoard1.jpg) includes the componenet layout.
This one (www.fineelectricco.com/PCBoard2.jpg) is just the trace and hole layout.

The boards are complete with plated-through holes, but not silkscreened with component layout, which is an extra charge. The circuit we are interested in is simple enough.

Larry
www.fineelectricco.com
My system (www.fineelectricco.com/System.html)

You might want to try eagle www.cadsoft.de
the free version will work for this. It runs under linux and windows. as for a pcb place AP is the cheapest I have found http://www.apcircuits.com/

The circuit is easy not what I would expect..
I was thinking it would use a FET in the feedback .
The hard part is going to figure out the size of trace to use to get 75ohms any one good at SPICE??

ps I use a Panasonic PV-SD5000 3.3 Mega Pixel for the pics

Originally posted by BLS

Also, did I get lost here? What parts and numbers are you replacing in these pictures? I can see your heading that states HVPS, but what are the part numbers you are replacing the bad ones with?

Brad

so far I use the same values for the resistors
r9,r10,r15,r16,r21,r22 560k 2watt
r5,r46 220 ohm 2watt
c1,c2 .47uF 630V
c6,c3 1uF 400V

you should be able to read the values in the pics.

Allied part #
BC COMPONENTS 2222-468-28105 Capacitors, Metallized Polyester, Radial, 1.0uF, 400 Volts 648-3080


BC COMPONENTS 2222-468-40474 Capacitors, Metallized Polyester, Radial, 0.47uF, 630 Volts 648-3116


OHMITE OY564K Resistor, Ceramic, Axial, 2 Watt, 10%, 560k 296-5456

BC COMPONENTS 5083NW220R0J12AFX
Resistor, Power, Film, Axial, 2 Watt, 5%, 220 Ohms (Pack of 100) 649-6128

Eagle will work, but not many places will take the format. APC does good work but on small runs charge an arm and a leg. I had a run done last fall of small 1x3 pcb's and had to do a minimum of 500 of them and got charged 5$can each for them even though I only needed around 200.
I have a place in Canada I can get the silk added on for free, gold plating for free (if wanted) and a good price on small runs although the original setup is always a few bucks. I've done hundreds of thoudsands of bucks of biz with him so I get very good deals. I've been manufacturing pcb's and circuits for sale for about 3 years now so have been to most places... from little 1"x3" pieces (daughterboards) to larger more complex 100 piece plus boards, thru-hole and smt and I am set up in-house to do that assembly also. The only downside is, is I have always hired another person to do my layouts. If that schematic is correct, then it should not be a problem although I need to know values and types for components to get the drill sizes correct for the holes. I would also have to pay him so if anyone here was adept at that that would save some $$.

Where would this connect? I have a non-gamma board, and one with the circuit and there is quite a difference all over the board so I don't know where it would go. But just as important as making the thing is a good mount if possible. If someone could send me a pic of where on the board it goes, maybe we could get lucky on the trace spacings and get a nice low-profile right angle header to mate to and build that into the circuit. If not as Larry said, wiring is the only way.

But is this even worth getting into cost wise? I would love to offer free samples as I said earlier to any who help, but it would not be cheap to be done. You can't go in and just ask for 20 pcb's..well I guess you could but its going to be expensive. Prototype pcb's (you usually get 5) run in the neighbourhood of 40-50$can each and you have to get 5 minimum. There is no silkscreen, no solder mask, no drill holes different then about 5 drill sizes they use since they don't want to be changing a lot of tools. I would never use one of these boards for an actual use. It is only one step over home-done etching so you definitely want an actual run done to keep quality up, which then goes to minimum amounts. What would people be willing to pay for a gamma circuit and is it worth it?

asdesign1@yahoo.com

prototype pcb vs final version.. no comparison.

http://www.protodevices.com/b1.jpg
http://www.protodevices.com/b2.jpg

I guess AP likes us in the US I do small runs all the time with them like 20 pcbs for under $100 us

As for how it should go I would think if it had a RCA connectors in and out it could just plug it in the green RCA and plug the blue cable into it. Then just fly a the power wires and solder them where ever there is a good spot.


bye the way nice pic of your smartcard reader;)


oops I meant blue:p

I would suggest a very small square board, that could be attached floating above the main board on wire stand-offs (1/2"). That small board should have the mini RCA jack on it, and it should have a mini RCA plug that would plug into the "blue" out on the VIM. The only other thing that would need to happen from there is to attach both + and - 5 volt supply and make sure that the small board is properly grounded.

20 for under 100$us?? The film cost is usually more than that! Was that just single sided or smt?

I just dug up last bill from apcircuits (they do all my proto's) for a batch of prototypes.
Setup-67$,
4 boards @ 4.6"x3.1" - 55.29$,
and extra drill 13.50 for a total of 145.30$can for 4 boards...and thats bare, no silkscreen or solder mask. Approx 100$us.
Whoever you talk to there must be cutting you a very good deal! Most of the time its not even worth it for a company to set up the machines just to run through 1 sheet.

Mike,
wire standoffs poses a threat of shorting something if it shifts. What I would suggest, and I only have the board 'with' the gamma circuit to look at is to use one of three methods. If the non-gamma board is different then neither of the first two might work.

1) use the area near U35 to mount using double sided tape. If circuit is small enough this should be no problem.

2) use one of the two holes drilled through the board on the connector end to use one of the plastic PC-motherboard spikes to mount it in. It wouldn't be exactly firm in its hole, but would be fine.

3) the best way as I see it is to have a small pcb with a male RCA pointing downward on the bottom, and a female on the top, possibly at right angle in case height is an issue. The pcb would 'plug-in' directly to the old RCA jack and then the wire plugged into the jack on the top. All you would need to solder would be a power source and ground. This would be very firm and give the shortest path, greatest ease of install. (I think this is Sly's idea also)

I use a company in Ireland or Germany called www.pcb-pool.com. I guess you have something like that in the USA. The price is 32.7 USD per 15.5 inch^2 for a doublesided PCB. You don't pay any startup cost at all. You don't need to make Gerber files. They take your PCB file directly.
I can put a test with my next PCBs. I will send files in three weeks time.

Kind regards, Claus

Originally posted by Tester007


Mike,
wire standoffs poses a threat of shorting something if it shifts. What I would suggest, and I only have the board 'with' the gamma circuit to look at is to use one of three methods. If the non-gamma board is different then neither of the first two might work.

1) use the area near U35 to mount using double sided tape. If circuit is small enough this should be no problem.

2) use one of the two holes drilled through the board on the connector end to use one of the plastic PC-motherboard spikes to mount it in. It wouldn't be exactly firm in its hole, but would be fine.

3) the best way as I see it is to have a small pcb with a male RCA pointing downward on the bottom, and a female on the top, possibly at right angle in case height is an issue. The pcb would 'plug-in' directly to the old RCA jack and then the wire plugged into the jack on the top. All you would need to solder would be a power source and ground. This would be very firm and give the shortest path, greatest ease of install. (I think this is Sly's idea also)

Stand-off is the same procedure the manufacturers use on retro upgrades. And mounting too close to U35 would be a bad idea because it is a DAC, this is a high bandwidth multi-layer video board, it's very important that the circuit is mounted within that area of the board, that section is isolated from the rest of the VIM.

The procedure for mounting is not that important, but locating the board outside of the video section of the VIM could induce noise in the blue. There are mixed signals (analog/digital) on that board, and much care should be taking to keep them apart, also keep in mind that the neck board is actually a very high power high frequency amplifier.

Sly,

the work you did on the HVPS, was that for repair or upgrade? Or should
I just be thinking don't fix it if ain't broke.

Can anyone recommend a sourse for the 20ohm 1W pot on the LVPS
for filament voltage. I looked at Digikey and everything in 1W is huge,
3/4" long. Is this the norm for this type of pot? Actually I was trying
to keep the same lay out so it could be adjusted from outside the case,
does this go out the window with pot replacement?

Bruce

"was that for repair or upgrade?"

well I would call it a repair with upgraded parts

The old parts were out of spec so I would call that a repair. but the new parts are better and might fix the snow problem so that would fall under upgrade.

for the LVPS it looks like a 475 fixed and a 20 ohm pot is the right mix taking in account the IR drop from the Vreg to the crt. I will try to post the digikey # when I get home.

when using AP you must use the free drills or it will cause the cost to skyrocket. Also the setup is fixed so the cost of 4 vs 20 is almost the same. and I make my boards small
I like to do SMT for most things. it is so much quicker to build and to rework with SMT.

Sly,
On the upgrade vs. repair question...were you actually having a problem when you changed components? Aside from testing and them testing out of spec, was there anything else going on to clue you in?

When will you be testing for the "snow" issue?

my HVPS had a a resistor burn in the potted part
so I am now going through the whole thing replacing all bad parts.

I have never noticed any snow my self but other people have seen it . And it is well known that carbon resistors cause white noise and if the resistors are damaged that the noise is worse. mine were way off (very damaged) now this has nothing to do with the failure but is way out of spec.
and should be fixed

and the caps were a bit weak and have heat damage.

so this is like replacing worn-out shocks on a car with new better quality shocks that have a lot longer life.

As for the caps they were showing there age and were getting weak but I was more worried that the might short later on doing damage. The working voltage on this type of cap go down as the cap ages. Just like the HV caps in the potting.

So far all the parts I have replaced is the same value.
just new and in spec parts and the new type of resistors should not go out of spec like the carbon type did.

This is not a mod. I not changing the value of any of the parts. The only miner thing is the .47 600V cap is replaced with a .47 630V cap but that is just the voltage the new cap comes in and having more voltage margin is OK.

so think of this as more of a preventive maintenance.

FYI

digikey 3299X-200-ND 20 OHM 3/8 SQ CERM MT SL POT

For use with a cut and a 475 ohm resistor soldered over the cut trace.

Okay, I'll "spill the beans" and tell you guys where I get my boards from. It's called ExpressPCB (www.expresspcb.com). The board CAD software is a free download. The mini-board option is $62, flat, for 3 boards, which can be cut into smaller ones if you just duplicate the pattern on each board. The price includes:

Double-sided boards
Plated-through holes
No silkscreen or solder masks
Board size must be 3.8 x 2.5 inches
Shipped the next business day

The circuit is probably small enough to fit 6 on each of the 3 boards, which means 18 boards for $62. Now that ain't bad at all. Even 3 per board is still a bargain. All you do is design the layout, upload your file to them, and you get the boards a couple of days later.

Larry
www.fineelectricco.com

Sly, yes I know non-standard is extra which is what the 13.50$ was. But even if you take that off its still expensive.. however looks like Larry has found a pretty cheap place!
Definitely a good price for doing Proto's. Thanks Larry!
But in an actual run they look to be 50-70% higher than I have paid for a final version (mask..etc).
So I think its up to what anyone wants.. a non-masked, self-cut board or a sanded, masked pcb. I personally feel if we are doing this as an upgrade then do it properly. Else it may be making something worse.
I will take the schematic and send to my pcb guy and ask him what it would be to do make a run as he can whip off a Pcad layout easily. Then I guess people can make a choice, but I think I am going that way. Mike I am going to PM you in a week when I get back in town with anything you think may be a problem or your ideas on it and what to look out for then I'll run a batch.

This thread is getting interesting!

p.s. Sly and others stock on pot is getting low at D-Key so you can get the same (a little cheaper) at www.alliedelec.com part#754-2702 as well.

Does anyone know where to get a replacement fan with the same specs for power supply? My rear one was making an odd noise a few days ago so I think I might as well replace it while I have my green tube off and I have it out for pot replace.

In regards to the LVPS and heater voltage,

I changed my resisistor to a 500 Ohm 1/4 watt metal film 1%. It measured at 496 ohms at room temp. With this in place I now measure 6.26 volts at the connector near the blue tube. This measurement was taken when cold and also after watching a 2 hour movie and was the same.

I am going to guess that replacing your pot with a fixed 500 ohm resistor will yield similiar results for others. (you certtainly would not exceed 6.35 volts unless something else was wrong in this circuit.)

-Peter

I have used ExpressPCB too.
And I just used PCBexpress it is easy to confuse the two withe names like this. It is a different place I used them because they did smaller trace/gap and I needed it for a part that had very fine pitch leads but it cost a lot more..
by the way every place I have ever went takes the Gerber that eagle puts out. I can't say the same for pricey Cadence tools.


Just remember this is a circuit running a Ghz speeds (edge rates) and the PCB acts like caps and inductors and transformers at this speed also the traces need to be 75 ohm impedance
or else you will get reflections. Normally this type of circuit and PCB is run through SPICE to get every thing right.

Exactly... This is why I am trying to point out getting it done for the cheapest like a prototype on FR-4 material is probably not the best idea when the whole point is to make things better.. :)

My shop does regularily FR-4, Polyimide, BT, MC-5, GML-1000 and can get others with surface finishing of H.A.L., Entek 106A, or Immersion Gold. I am going to be gone for a week, but when I get back should have his reply on costs and I would like to float the final idea and layout past a couple people, Sly included if you would? I have a few friends with Marquees and would like to make some boards the best they can be.

Good job guy's. I'll take it that we'll have a retro-fit blue gamma circuit available some time in the near future.

There's one more thing to do on the VIM, and then we'll move on to the convergence and vertical boards, but before we move on make sure to resolder the center contact on the three mini RCA jacks, and also do clean this board with denatured alcohol.

Here is another nugget:

Replace the three mini coax cables that are connected from the neck boards to the VIM on the mini RCA's - use commscope 73501S (mini 75 ohm). Make sure that all three cables are the same length.

Note: Only the early vintage 8000/9000 VIM boards did not have the gamma circuit.

Greetings Mike! Anything special we should use for the new rca ends that attach to the coax? Is there a particular brand you like over others?

It would be good if someone could get a shot both front and back of ether the factory doter PCB or of the new VIM PCB where the Gamma circuit is. I think we should try to do the exact same layout. because the layout of the traces is part of the circuit. I do not know how many layers the VIM is so we might not be able to see all of the traces.

These pics help at all? Let me know what you guys need and I'll post 'em
here or email them.

Bruce

backside

Great pics bob

As I thought layers on the inside can't be seen
but the top looks like is has a lot of the circuit:)
so I think this is very good info.



Also I think I have finished the schematic of the HVPS
I hope I have not made any mistakes.
POWER PC BOARD
ASSY 404251-

Hi Sly

This is awesome. Thanks a lot for the great work. This will make our Marquee investment even better. Now we should have a chance of repairing it if we run into bad luck.

Kind regards, Claus

OK guy's, let's stay on track and keep some perspective here. There are areas of these projectors that has certain SAFETY limits. We should stay focused on "maintenance and upgrades" and be very careful to not violate any of the critical safety areas of these remarkable machines.

Every CRT projector sold in the US, Canada and most of Europe, must meet certain interferance and radiation requirements, before they can be sold in these countries. In the US it's the UL, FCC, FDA to name a few.

The FDA (food and drug administration) department of health and human services (DHHR) regulates radiation limits for CRT projectors. The FCC (federal communications commission) regulates "interferance" and let's not forget underwriters laboratory (UL).

The Marquee in its original design has all of the above approval. And we should be very careful to maintain these very critical safety concerns. These HVPS are manuafactured and potted to keep these supplies from being a safety issue, they are very carefully designed that if for any reason the supply could contribute to the emission of X-radiation, the supply would not come on. This is one of the built in protect features.

One of the reasons why that power supply is potted is to help seal and keep that safety system intact. By removing the potting material and replacing parts, that internal system has been violated, therefore the proper equipment MUST be used to calibrate the two mini pots (R35/R36). One is for overvoltage protect and the other is to set the high voltage at a precise 34.9KV.

Unless you have these capabilities, I would suggest that if you enter the potted area, do it with the understanding that you have violated the safety system, and that your projector or the projector that the supply will be operating in could be a X-radiation hazard. Make sure that you indicate that on the supply where it can be visually seen.

Now, during the continuation of this thread, let us not get into anything that could bring any potential harm or danger to anyone. I ask that any further discussion on the repair of the HVPS be not discussed on this thread.

One last thing on the gamma. Heard back today and the shop I go to recommended not to do it and that they were not capable.
Part of response was:

"Material is also specific and many cases require teflon base.Very difficult to plate.Sorry to say for high freq.pcb. Desiging pcb lay out is also critical."

I do have to agree there is REAL DANGER in this potted part too.
The caps can and will keep the High voltage inside and if you open it you will find the voltage.
and there are big caps with up to 12KV each!!!
And you need special equipment to repot it.
A vacuum chamber oven .
and the potting stuff is not nice ether you need a VENTED vacuum chamber oven. I have all this stuff so I going for it but it is not going to save me any cash. I am doing it just to understand it better.
and it is a lot of work... I do not think I would do it again

The stuff that I posted on the out side is safe to do. it is just bringing old components in spec
and it has nothing to do with the safety systems it is mostly the G2 circuit. that just happens to be on the HVPS PCB

and NEVER NEVER touch the sealed pots on the HVPS
they are sealed for a reason..

Now back to our regularity scheduled program...

FR4 is OK that is what is VIM is. But it is a impedance controlled PCB and that cost as well as it is many layers.

This type of circuit can be done it just needs to have the impedance matched
but this it the hardest type of circuit to do.

If we can get the trace set to 75 ohm use the same layout
I think it might work.. we might have to tune each one
use a gimmick or two. ( you ask whats a gimmick)
it is 2 wires twisted together with the insulation still on.
yon then trim and twist/untwist to change very very small amounts of capacitance. used in old TVs in the UHF tuner. And also use to null the parasitic capacitance of a transistor.

And if you use a HTPC I still think it would be better to just set the blue gamma in the RAMDAC I think all video cards let you set the gamma.

Mike
I have been looking for the mini RCA type male connectors to use when replacing the input cables but I can't find them anywhere.Do you know of a source?
Thanks
Andrew

Just a thought, but since Mike wants to keep this thread on topic as to maintanance and some updating, and I agree, why don't we start a separate thread for those that have a few extra boards around that want to try some actual tweaking?

Andrew,
I've been using the ones that's there. The mini RCA's are hard to come by. I have a lead on some and if that works out I'll post back.

Guy's,
tweaking is something that most of us who are into electronics, and are somewhat skilled with a soldering irons loves doing. But unlike the common audio amp or other piece of electronic gear, these high end CRT projectors can be a real source of danger if care is not taking when tweaking or trying to make the product better than what it is.

There are three areas of concern when randomly tweaking these beast, and they are:

- Low voltage power supply. It has special testing for interferance and isolation (Hi-Pot) that is conducted at FCC compliance labs. These test are required for FCC and UL compliance.

- High voltage power supply. These supples are tested and should have at least two stickers or labels on them indicating that they are safety compliant for X-radiation safety (DHHS).

- Neck boards. These boards on the Marquee has been modified at the factory for FCC compliance on the earlier models, the two white wires that are attached to these boards are called "FCC connectors" they were added on the earlier models to add better grounding to the neck boards for a much better FCC compliance.

All of the above modules or boards have "safety critical components" in them, and only the original part should be replaced when replacement is needed, therefore "tweaking" is not a good idea in these circuits, because if the wrong or a different component is used it could disable the protect circuits (safety), and cause the CRT's anode voltage to exceed the safe limit, and if that happens the projector could turn into a low grade X-ray machine. The reason for leaving these components in place is the assurance of safety. These projectors have been flawlessly safe over the years, thanks to strict government regulation, and also to the fact that the power supplies are provided from industrial experts in supplying safety compliant products.

So if you're into tweaking, please be very careful with the above three modules.

I don't think I would touch them. What I am more interested in is tweaking some of the input area much like what has been done for the smaller Marquees.

Tester007,
as we go along I'll point out some areas were I think improvements can be made, but I'll not get into suggesting too much, because I think that the proper upgrades is all that this projector needs. There are areas where improvement can be made for better stability, and so far we've covered a lot of them. Once we get to the vertical and convergence boards you'll see what I mean.

keep in mind that the Marquee was not a projector that was manufactured for a year or two and then it was discontinued, it's now been in production for over tens years, and the engineers at Electrohome was very involved in this projector for that ten year period, they did not move on to some other product, therfore it has experienced much attention over that period of time.

Any image of performance improvements to the Marquee is only relevant to the quality of the source material and setup skill involved.

Unlike many other CRT projectors the Marquee has an almost directly coupled video chain, with only one coupling cap in each video channel.

We're gonna have some fun with the next two boards...

The mini-RCA's is used for the IF connection on some tunner modules I have not seen were to get them yet.
But they are not that un-common. Becaus RCA does not make them and RCA is a trademark they seem to go by the name of phono.

I found some right angle phono plugs at http://www.connect-tech-products.com/ part# ctp-175-12
these are 2.35mm, any idea if these are the right size?

I wet to work and grabbed a set of calipers
and the VIM's mini RCA's measured 6.35mm with a 1.7mm PIN

so it looks like CTP-174-6A-1
and
CTP-175-11 with 1.7mm pin option
is what we can use..

So Bob found them...

so I will ask around for a SPICE expert to help me figure out the trace thickness and I think we are good to go..

Try Here for mini phono plug http://www.excess-solutions.com/audio_connectors.htm
Andrew

Guy's
good job on the research, and good luck with the gamma circuit. And thanks for helping me with this. let me assure you that if you're keeping up with this and making the recommended changes where needed, you should see a change in the performance of your Marquee, and you'll also gain a better appreciation for it when we're done with this thread.

I've exausted myself last week wearing many hats, so I had to push myself away from the table for awhile to chill out, now all is cool and we can continue with where we left off.

We covered the VIM's and neck boards, we've also did a little poking into the low voltage and high voltage power supplies, and we'll revisit that for a little more discussion later on. From time to time I'll throw in a performance mod. But I'll not share all that I know ;)

Next we'll look at the vertical, horizontal and convergence boards. These will be more like reliabilty upgrades and maintenance, and I'll show how to minimize drifting and make the projector more stable.

Tomorrow...

Is there a lot of differences between the Marquee and ECP 4101 or does some of these mods apply to the ECP 4101 also?

Thanks, Henrik

Henrik,
the ECP series is very different from the Marquees. I don't have much experience with the ECP's as I do with the Marquees, and maybe someone who has that experience will start a thread like this for the ECP's. The Marquee is a different beast.

---------------------------------

Today we'll start with the deflection and convergence boards, but before we start there, we'll perform some overall maintenance on the projector as a whole, this is where the bed sheet comes in. Note: the following could slightly effect your fine setup.

- place the sheet on the floor below the projector (place around if floor mounted), you will need a vacuum cleaner (with nozzle) and brush to loosen dust with, canned air, lens cleaning cloth and glass cleaner.

- remove the three lenses from the front of the projector (non LC models only), and mark or label them as to where they came from (r, g, b)

- now remove the bottom (if ceiling mounted) cover (2 screws) and rear cover (6 screws) that cover large heat sink on the rear of the projector, as well as the cover over the power supplies (4 screws).

- with the lens removed (non LC models only) and the three covers removed, dust out the projector as much as possible, making sure to also remove all dust from the vent holes in all covers. Use the brush (or canned air) to delicately loosen any dust from exposed areas inside the projector, you can also pull out the focus (between the red and green CRT's) and the horizontal (between the green and blue tubes) boards and dust as much as possible with canned air.

- place vacuum nozzle on the vent holes on the neck boards, and the hole openings were the wires are coming out of the CRT housings.

- remove both power supplies and remove their covers and dust out the insides using canned air.

- dust out the area where the power supplies were remove from, and reinsert the power supplies.

- now we clean the surface of the CRT's using a glass cleaner, and we clean the surface of the lens with a cloth lens cleaner, and then re-attach the lens to CRT they were removed from.


----------


Later today we'll start on the Horizotal deflection board.

Mike, I have been following your thread so far doing the necessary maintenance, upgrades, and in some cases, mods that you recemmend. Everything is great so far. I have a M8000 manufactured in 1995. My question to you is, when I put the projector into standby when done using, the fans small squeaking noise has they shut off, is this normal in your experience? Is there a way to lubricate them while I have my machine apart, or do they need to be replaced? Your or any other person's advice on this would be greatly appreciated. I figured now is the best time to take care of this. Thanks in advance!

Paul

Paul,
there are five fans in a Marquee. Two are in the low voltage power supply (right where the power cold connects), and three on top of the projector (where the mounting bracket attaches). You should first try to locate where the noise is coming from (which location). If it's the three on top of the projector it could be that a small group of wires had become loose from the harness and are touching one of the fans. If that's the case then you have to remove the cover off the fans and secure the wires away from the fans, I use wire ties for this.

If it's coming from the power supply you'll have to contact Tim Martin for the fix.

-------------------


Ok, it's now time for the fun part. We are going to cover three boards, and all three of these boards play a part in convergence drift. First we'll start with the horizontal deflection module (between the green and blue tubes).

For this you'll need a good soldering tool, solder and a 220 mfd @ 160vdc cap (C50). And remember that we clean each board as we go with denatured alcohol.
http://www.geocities.com/hvps2001/hdm/hdm_c50.jpg
Replace "C50" it's in the modulator circuit on that board, The modulator circuit supplies the operating voltages for the deflection power stage (power mosfets), this circuit varies the voltage for the fets based on the scan rate that the project receives. If this cap leaks, it'll cause erratic width performance, or inability to achieve max horizontal width.


Then we'll move on to soldering various contacts on the board. If you remove this board and closely examine the contacts you'll immediately notice the need for soldering. The contacts that will need this soldering are the the three power mosfets and two transistors that are mounted to the heatsink.
http://www.geocities.com/hvps2001/hdm/hdm_fet.jpg
The next items for soldering are the five relays and six coils that are mounted
next to the three large relays.
http://www.geocities.com/hvps2001/hdm/hdm_solder.jpg
Next, inspect the three yoke connectors that are attached to the board, look through the plastic connector and look for burns or dark marks in the area where the contacts meet. If you see burn or darkness on that connector, it'll need replacement.

This simple procedure greatly stabilizes this board, and it'll protect the board from failure, because it's very important that these contacts are tightly soldered, if not, as the board heats up, you could experience drifting or even board failure.

If you can't see the images try: http://www.geocities.com/hvps2001/hdm/

I keep getting asked about the LVPS mod I did so here it is.

This is the trace that needs to be cut.
http://www.avsforum.com/avs-vb/attachment.php?attachmentid=156330&d=1256592674

And this is the 475 Ohm resistor installed. I used a 805 size part rated at 1/8 watts you might want to use a 1206 for more wattage just to play it safe.
http://www.avsforum.com/avs-vb/attachment.php?attachmentid=156331&d=1256592674

Now just install a 20 ohm pot.

Thanks Sly, the pictures helped immensly in understanding what was going on. I'll let you know how it turns out.

Sly
the picture are missing!!!

could your correct it please

thanks


Abshar

Strange...I couldn't see any of the pictures in Mike's post until I visitied the Geocities page and came back. Now, because they are "in memory" I can see them all fine on this page.

I can't see the ones from Sly about the mod though...what is a link to the website with these pics Sly?

Jay,
If you right-click on the picture that isn't there (the box with the x in it), then go to properties, you will see the website where the picture is stored. Just copy and paste into your browser, and do the same thing with Mike's pictures. Put them into memory, and then come back to this thread. Don't know why you have to go through all that work, but it works.

Paul

DoH! Thanks Paul!

I edited my post to include the link. I think they mark a cookie or something just to force you to see one of there ads. after you are there reload works so it is not just a cache thing. That what I get for being a cheapskate;)

Mike, is it possible to kinda get a bit of a partslist for future things you will be going over? Digikey charges 8$ per ship so it would be nice to get most of the things at once if possible... We have one or two half decent parts places in town here but they usually have to order in the odd stuff anyways as well.

Tester,
we're almost finished with part upgrades. I'm going through my notes, and once we get to the vertical and convergence boards, we'll be doing something a little different from what we've already done, and there are some more parts to replace, but I'll rather post as we go.

I've shown the results of these upgrades to a few forum members in my area, they have seen the results to include witnessing the convergence from a cold start to a one hour later warm up, and that was after a partial upgrade, and I've since made a few more very noticeable improvements. Now they've insisted on me doing the same to their projectors.

I want to be clear that I'm not saying that we're able to make the Marquee better, though there are a few components that could use an upgrade, but they were in the very early version Marquees, and as you have seen from the difference from the neck and VIM boards changes, Electrohome has already made their own upgrades. What we're doing is applying some or most of them to the early version projectors, with the intent of bringing the early version projector as close as possible to the later versions.

So far some of you have presented some highly workable ideas, and I'm looking forward to hearing the results (gamma, filament, etc.). there are some amazing minds on this forum.

Tomorrow - convergence board.

Now I'll share a real nugget.


First we'll start on the vertical. This board has three separate vertical amps that drive the yokes, and they are directly controlled by the DAC's (digital to analog converters) on the board. There are 8 resistors I'll recommend for change on this board. Changing the caps on this board did not make any measurable (scope/meter) change or difference on this board. But replacing certain resistors did.

On the early version boards, you'll see that resistors R720, R721, R820, R821, R920 and R921 are mini one (1) ohm resistors. On the later version boards, there are two of these one ohm resistors in parallel, they are the plus and minus 15 volt supply rails for the vertical output IC's. So that would mean that the early version boards had a one ohm resistor on each of the IC's supply, and the later version boards have one half ohm (1/2 ohm) on each supply rail. In other words the upgrade was a doubling up of these resistors. I only had three problems with the design of these projectors, and the use of these mickey-mouse resistors is one of them. They're supposed to be fusable metal film resistrors at 1/2 watt :rolleyes:

Your choice would be to upgrade by adding the other resistor in parallel to the one in place, that's if you have the early version board, or do as I do on both the early and later version boards -- remove all of them and put in real fusable resistors. You'll also want to replace R5 and R6 with the same resistors also.

solder each lead on the three vertical output IC's, and remember to remove all solder flux (denatured alcohol), flux is good for solding, but bad for the circuits.

The above greatly improves the vertical boards stability.

---------

Now for the convergence board. This board is very well designed, with only one exception, yep, they used the same little mickey mouse resistors on this board. The early version boards had one each, while the later version boards had a double stack. There are twelve of them on the convergence board (24/stacked on the later version). R118, R117, R218, R217, R318, R317, R418, R417, R518, R517, R618 and R617. There are also 48 1.2 ohm (same mickey-mouse resistors) right near the power transistors.

And while you have this board out, make sure you solder the many leads for the 30 power transistors that are heat sinked. If you look very closely at the contacts on these transistors you should see the need for soldering, especially on a high use projector, take your time and solder all 90 contacts.

Next, you'll want to clean that heat sink surface insulator (denatured alcohol) and apply a very light film of petroleum jelly on that insulator. Before reinserting the board, use a ruler or other flat device and align each transistor so that each one would be perfectly inline with the heatsink surface. Now take each heat sink clamp and bend back for better tension. Do the same for the vertical and stimator boards (if it has stimator board).

Now, I'll find a hole to crawl in, because I've just shared one the nuggets I did not want to share :(

No...stay out of the hole....you are getting sleepy....don't stare into the light...noooooo. :D

Thanks Mike,
I finally started my upgrades yesterday, beginning with the HDM. I am still waiting my green to return from the rebuilders, so I don't want to do much without being able to test in between. I'd hate to "knock" something loose and not know it until I was completely finished...especially the removal/replacement of all those resistors.

Plus, if I wait long enough, I can order all my parts at once! :)

Thanks again for this great comntribution!

And Mike, a question, on the need to resolder...

All of mine on the HDM looked really good under the magnifying lamp...

Are these cold solder joints from the factory or do they go abd over time from heat etc? My unit has only about 3500 hours on it, built in 1996 so I am unsure whether I should be seeing this.

I'm doing all the resoldering anyway, just was a curiosity question.

Geocities is not the best place to try to link pictures from, because they don't allow direct links to files.

I am willing to donate whatever web space is needed for the pics and web pages related to the project and give it a permanent home as an article on my Techweenies site.

If the folks doing this are interested, you can send me a PM and I will coordinate it with you.

Mongo

Jay,
I'm not sure but I believe the weak soldering joints would be on most of the early version boards, not really noticed on the vertical though, but it's best to solder them as well (Preventative Maintenance). I think it may come from heat and high hour usage, because the transistors on the heatsink get really hot. I had the same problem on the 9500 that I have here. It took awhile to isolate where the drifting was coming from, but I did get to put my finger on it (literally). Another way to check for this is to let the projector warm up for about twenty minutes, bring up a converged grid and press on the convergence board from the center, if the convergence shifts, it's the connections. You can also do this with the horizontal deflection circuit, bring up the grid and then touch the area where the coils are with a plastic non conductive, non magnetic object.

You may not have noticed this on your Marquee, but you've probably heard this before: "after the projector has been on for awhile I have to touch up on the convergence" The Marquee if properly setup and operating properly should show very little drift from cold to warm up. I have a few customers who has complained about this drifting, so I rolled my sleeves up and went to work here in my shop, and I had the perfect Marquee for it, it had high hours and it was a little drift happy. It's now very tight, and I like showing it off, but it required a little work to achieve that stability.

The good thing that I really like about the convergence and vertical boards is the stabilty of the output devices, especially the power transistors on the convergence board. These transistors can get very hot, yet they only show variance if they are not properly mounted to the heatsink. And I always revitalize the rubber isolator when I replace a semiconductor, I don't trust dry contact, it works fine when the isolator is new, but after awhile the surface hardens and looses its effectiveness.

Mongo, thanks for the offer. I'm sure it's something we'll probably need. I'm waiting on the gamma project, and would like to hear how the power supply filament fix pans out. I think that with our cooperative work effort, we'll keep this wonderful work of electronics at the apex of video display.

Mike would it not be a good to use heat-sink compound instead of Vaseline??

I have Gamma circuit in eagle I just had no time last week to do any thing. I hope to do the layout tonight.. Then all that will be needed it to adjust the trace width so it is 75 ohm.

Sly,
sure, heat-sink compound should be fine, as long as it can revitalize that isolator, that's the real goal. I did not mention heat-sink compound because some of it can be so thick and pasty. I also use the small tubes from Radio Shack, it goes on just like petroleum jelly, but make sure you clean the isolators surface first.

Oh, and don't forget to use this same procedure on the stigmator board (next to the convergence board), for those who have 8500's, 9500 and 9000's

Hi Mike

Thanks for the latest installment of Marquee maintenance. I am
having difficulty locating "fusable metal film resistors". Digikey has
some described as "metal oxide resistors" or as "metal film resistors".
They carry the YAGEO RSF series which offers a .47 ohm or a .51 ohm
at 1 watt resistor. The BC and Panasonic resistors are only offered in 1
ohm or higher at 1 watt. will the YAGEO be OK even though it is not
described as fusable? They all have a 5% tolerance.

Thanks

DH

DH,
here you go:

http://www.micro-ohm.com/

You'll want Part# FRN1/4 - 1 ohm 5% tolerance.

Mike

Thanks for the info.

DH

OK I found a cool free program to do the math for the trace widths.

http://www.hp.woodshot.com/appcad/version302/setup.exe

it won't be long now...

I think I almost got it done.

I think I will add 2 10uF caps on the power coming in
it looks like is will cost about $1 each for the raw PCB's from AP.

take a look .

Kudos!

Mike

Hellloooo Mike

I think there is still one board to go. I am waiting with iron in hand O Great One. Please impart more wisdom to those, like me, who kneel at your feet.

A Dark Side Disciple

Ok, we'll get back to this in a day or so. We have the focus and control boards, and then we'll go over what we have covered, and I'll throw in another nugget or two...

mike

I got some sample jacks from Connect-teck very nice guys.
so I will be sending out the PCB as soon as I check that the jacks look OK.

I got my Gama PCB back and have finished building the first one. I have not powered it up yet.. I need to find a split power-suply I can use and good pulse gen. to test/tune it..

Betcha Mike's got a bunch of split power supplies (+/-5V) ala MP-1.

Cary

PS. You can also get +/- 12V from a PCI slot. I use a riser card with a thru-hole PCI connector on it and solder wires to the back of it to get at +/- 12.

I found a supply + - 5V


There is some over shoot I trying to get rid of.
the mini RCA is crap and causing some ringing on its own..

I think one of the mods should be to remove them all and replace with something that is 75ohm impedance.

I have been busy so I have not spent much time on this
but I am in the tuning stage now..

Originally posted by sly
I found a supply + - 5V


There is some over shoot I trying to get rid of.
the mini RCA is crap and causing some ringing on its own..

I think one of the mods should be to remove them all and replace with something that is 75ohm impedance.




Sly,
the circuit (by itself) should not cause overshoot, nor should the mini RCA's. It sounds like the chip could be oscillating, and that's one of the problems when using high bandwidth Op amps, that's where the bypass caps come in, and that's also why they are to be placed as close as possible to the chip. Your board design looked good (I'm not sure here), so I would not suspect the board design, and therefore redesign the circuit itself to tame the "overshoot".

Why would you use a separate power supply when there's +/- 5 volts on the board?

I would think you should mount the circuit board as close to the main board as possible (I would mount it right on top of the RCA jack). Get your power from the board, but use 100uh inductors (one for each +/- power rail) for 'decoupling' from the main boards supply.

I wouldn't worry about the RCA's, your problem could be more related to grounding, decoupling or board location.

I am using a lab power supply because I have the board in the lab not in a PJ.
I have a pulse gen. feeding it and then the output goes to a scope. I had to make the 50 to 75 ohm network for the output of the pulse gen. And a 75 to 50 ohm network for the scope. This is how I am tuning it. So far its needs a 1 pF cap in the feedback loop the bypass look good but there is a tad bit of drop on the edge so I am going to add some small caps on the power I think the .1 is to big for the real high frequencies. I have only spent about 1 hrs on it so far I been busy doing other stuff.

It is getting there. When I think it looks good then it will go on the VIM and it is plugged on the connector there is no way to get it any closer then that..

I know the mini RCA cause some ringing because if i run it through the connectors going from the generator to the scope my nice square wave gets distorted with some over shoot. This just makes the tuning a bit harder thats all. I am feeding it with very fast edge's off a HP pulse gen. probably a lot faster then it will ever see in the PJ but it is good to tune too.

Originally posted by sly
I had to make the 50 to 75 ohm network for the output of the pulse gen. And a 75 to 50 ohm network for the scope.

This is incorrect. The input to the scope is not 50 ohms, it is a high impedence on the order of 1-10 megohms. With a 10X probe it's higher still.

The output from the board should be terminated with a 75 ohm termination (assuming that the load presented by the next stage is 75 ohms). Ignore the load presented by the scope.

Marvin Match

Unless he's not using a scope probe but a coaxial input to the scope.

Cary

Originally posted by jcmccorm
Unless he's not using a scope probe but a coaxial input to the scope.

Cary

Nope. The direct input to the scope is 1 megohm shunted by 20-35 picofarads. Thru a 10X probe it's more like 10 megohms and 2-4 picofarads.

A far cry from 50 ohms, probe or not.

If he's going directly into the scope, then he needs to use 75 ohm cable and a 75 ohm thru termination right at the scope.

Marvin

Originally posted by Marvin Match
Nope. The direct input to the scope is 1 megohm shunted by 20-35 picofarads. Thru a 10X probe it's more like 10 megohms and 2-4 picofarads.

A far cry from 50 ohms, probe or not.

If he's going directly into the scope, then he needs to use 75 ohm cable and a 75 ohm thru termination right at the scope.

Marvin

Yep, and the scope should not effect the circuit at all, even if the circuit is connected directly to the scope (1 meg) using a 75 ohm terminator.

Most scops have a 50ohm term you just hit the button and the 50ohm lites up...

I had a bad SMA I have replaced it and I am getting very good results now. I added 2 .01uf to the bypass and the 1 pF on the feed back. They stole my scope to do work, how dare they.... so I moved down to my lab and used my old 1Ghz analog scope ( I trust it more anyways the digitals lie)
I took some snaps I will download latter.. but I think it is working as well as the op-amp will let it. If the 1pF is left off there is a bit of over shoot.
If there is some role-off elsewhere we might want that boost in the hi-end..

...and they have the 50ohm input because you need this when using a 50ohm coax as your probe.

My TEK2230 just has two 1M/20pf inputs but the digital at work lets you switch to 50ohm.

Cary

the normal thing if you cant see any thing try http://www.geocities.com/hvps2001/gamma/


so here is the circuit board

http://www.avsforum.com/avs-vb/attachment.php?attachmentid=156333&stc=1&d=1256594876

here it shows the gamma boost in vs out notice the ramp change.
http://www.avsforum.com/avs-vb/attachment.php?attachmentid=156334&stc=1&d=1256594876

here is the speed test
http://www.avsforum.com/avs-vb/attachment.php?attachmentid=156335&stc=1&d=1256594876
and here is my mess with my old analogue scope
the ramp Gen. and the Pulse Gen. power supply etc.
http://www.avsforum.com/avs-vb/attachment.php?attachmentid=156336&stc=1&d=1256594876

Originally posted by mp20748
Guy's,
the MAX4224ESA (Dallas Semiconductor) chips arrived on Friday. I've already installed them on a neck board, tested and viewed. So far this chip seems to be a perfect substitute. Could not even notice any difference in the gray scale.

This chip seems to be a good replacement for the now obsolete CLC449.

That completes the neck board, now we'll pick back up on the VIM...

Where can one get this chip. I tries everywhere and they only do quantity of 100.

Is their somewhere I could purchase 12 or so?

Thanks!

jrichards,
Go directly to the Dallas semiconductor website. From there you can request samples and get your 12 for free! This is the only place that I know of where you can get them in small quantities (less than a hundred). Good luck!

Paul

Thanks!
They are sending me 10.

I have a Marquee switcher which has the 409 chip in it's output circuit. Do you think it would be worth while changing it out with the new part? Also is their any other mods fot this switcher which might help it's performance?

Originally posted by jrichards
Thanks!
They are sending me 10.

I have a Marquee switcher which has the 409 chip in it's output circuit. Do you think it would be worth while changing it out with the new part? Also is their any other mods fot this switcher which might help it's performance?

Yes, remove the CLC409's, and I'm sure there's other things that could improve on the switchers performance, but I can't say because it's been a while since I've seen one.

The MAX4224 would definately improve that switcher...

I been very Busy but I thought I would start a parts list for what has been covered so far.

I think I got everything.


Neck

6 22uF 160V 105 degree
3 MAX4224

VIM

6 MAX4224

Vert squeeze

3 56.2K ohm digikey # 56.2KXB

HVPS G2

r9,r10,r15,r16,r21,r22 560k 2watt
r5,r46 220 ohm 2watt
c1,c2 .47uF 630V
c6,c3 1uF 400V

Allied part #
BC COMPONENTS 2222-468-28105 Capacitors, Metallized Polyester, Radial, 1.0uF, 40
0 Volts 648-3080


BC COMPONENTS 2222-468-40474 Capacitors, Metallized Polyester, Radial, 0.47uF, 6
30 Volts 648-3116


OHMITE OY564K Resistor, Ceramic, Axial, 2 Watt, 10%, 560k 296-5456

BC COMPONENTS 5083NW220R0J12AFX
Resistor, Power, Film, Axial, 2 Watt, 5%, 220 Ohms (Pack of 100) 649-6128


LVPS heater fix

20ohm pot
digikey 3299X-200-ND 20 OHM 3/8 SQ CERM MT SL POT

475ohm 1206

Vertical

fusable metal film resistrors at 1/2 watt
R720, R721, R820, R821, R920 and R921 R5 R6
change to .5 ohm

Convergence
There are twelve of them on the convergence board
R118, R117, R218, R217, R318, R317, R418, R417, R518, R517, R618 and R617

and
There are also 48 1.2 ohm near the power transistors.


horizontal deflection module 220 mfd @ 160vdc cap (C50)

Is this thread dead? I was hoping to get the last of the information needed to finish my Marquee 8000.

I think Mike is busy building a transcoder. I still have resistor mods to do. They just came in. I won't be ready for the rest for a cuppolla days yet.

Walt

I'm ready when you guy's are.

I forgot to mention to remove the contrast modulation board (color correction module), right next to the ACON board (above the control module). This module is not in all Marquee's, look for it on the 8500, 8500LC. 9000, 9500LC.

This board will allow color correction in nine different zones. Much needed for multi-display commercial systems for edge matching and such, but not needed in video applications, and could effect the overall purity of the image - REMOVE!

----------

Right now I have a 8500LC that has the full maintenance performed on it. It's a customers unit that replaced another 8500LC he had hung in the same spot. So we were able to clearly see what improvements the maintenance and upgrades can do, and I have not done a calibration as of yet, but we have some pretty tight lines, and this is with VDC rebuilt tubes.

-----

I'll be making another visit to Arizona to visit Tim Martin real soon, and from that visit, there should be something very exciting to look forward to;)

Originally posted by mp20748
I'm ready when you guy's are.


Now that's a silly question :D I think we're all on the edge of our seats Mike!

Hahaha, I was thinking the same thing Jay! I think Mike sent us students off to do our lab experiment and was waiting for us to come back with the lab report and say we're ready for the next step. :)

I love this thread...

Cary

WRT .7v vs 1v input levels, does anyone know if in practice the Faroudja DVP5000 has sufficient output? Its specs say 700mv nominal, with specific signals indicated at 1v. I run its output through an ehome switcher which then drives the pj. In almost all cases when 5000 is in use the pix seem darker and noisier compared to other inputs. But the 5000 processes only DVD which is a lower quality source to start with, compared to HD sources which also connect to the switcher.

--Bill

Originally posted by jrichards
I have a Marquee switcher which has the 409 chip in it's output circuit. Do you think it would be worth while changing it out with the new part? Also is their any other mods fot this switcher which might help it's performance?
--------------------------------------------------------------------------------
To which Mike said
Yes, remove the CLC409's, and I'm sure there's other things that could improve on the switchers performance, but I can't say because it's been a while since I've seen one.

The MAX4224 would definately improve that switcher...

The 409's are on the RGB outputs, but 430's are on the H & V outputs. Would there be any benefit in upgrading those as well (with a suitable sub of course).

Mike, I have an extra Marquee switcher which doesn't work quite right (power control problem, not video) which I could loan you if you wanted to look into fixes/upgrades for them.

--Bill

Mike, could you please comment on these issues?

In an earlier message detailing resistor changes on the vertical board, you state that later rev boards will have each of the six resistors listed piggybacked with a like type. On my board only R721,R821 & R921 are piggybacked -- the other three are not. Is this of any significance?

The resistors in question both on vertical and convergence boards are 'mickey mouse', yet they appear to be identical to every other resistor on the board. Are they really a different type while looking similar, or is that type of resistor simply a bad choice for use on the rails? (I believe you, I'm just curious as to the why)

On the convergence board details you mention there are 48 1.2 ohm resistors right near the power transistors that should be replaced also. On my board there are 24 1.2 ohm resistors (R168,169,170,171, 268,269,270,271, 368,369,370,371, 468,469,470,471, 568,569,570,571, 668,669,670,671. And evens in each hundreds number group, and odds in each number group are in parallel with each other (e.g. R168,R170 are paralleled, R169,R171 are as well, up the list). Is my board somehow different that earlier ones? And each of these 24 should be replaced with the same value, 1.2 ohms?

Thanks for the information you have been sharing. It certainly has been an eye opener!

--Bill

bblue,
on the switcher issue: If you can, don't use it. If you use it have it upgraded with the later version chips, to include a few other components.

convergence board: There are 48 resistors total. 24 (if the later version board, 12 if the early boaard) are 1 ohm, and they would be double stacked. The other 24 are 1.2 ohm and they as well are double stacked and near the heatsink (power transistors). I've listed the replacement part number and where to get them on this thread.

Starting next week we'll look at a fully done Marquee 8500LC. This Marquee will have a complete set of VDC rebuilds, to include all of the suggested tweaks, mods, maintenance and a few special tweaks ;). Thanks to my friend James Bond (Bethesda MD), this display will be done using his Home Theater setup, and he's giving me permission to do an open inventation to anyone who'll like to attend. we'll also bring back and post some screen shots from this gathering. Actual date to be determined.

And we'll also debut my new Transcoder on this same projector, using a JVC D-VHS (HDTV) deck. But the primary source for the gathering will be a HTPC with a modified ATI 9700MP-1 card.

I've had a few request for when William Phelps will be back in our area, and if we can get enough interest we might be able to get him out around the time of the gathering or anytime soon after that. William is a high end setup/calibration expert, so he only does the very best in the technology. And he's Vidikron trained on the Marquee, so he has the proper qualifications for this projector, to include being one the few best on the Sony G90, plus he really knows and understand the critical technical stuff, as it relates to setup. He also does the high end lamp projectors, and he has some lamp clients in this area who praises him, and claims that after he had completed their projector, that they had noticed a huge difference in the image. Now, the only way I know to improve the displayed image on a lamp projector is to turn it off, they seem to look better that way to me :D

"Modified 9700MP-1"?? Modified beyond having the MP-1 attached? Something new?

Cary

Originally posted by jcmccorm
"Modified 9700MP-1"?? Modified beyond having the MP-1 attached? Something new?

Cary

Cary,
I meant to say that the video card will be modified with the MP-1. So far the mod itself is the same.

Also, the screen shots will be of test patterns. I don't believe that a screen shot could properly represent image quality. I intend to show a few pics of lines and flares and such...

How about setting up an NEC XG series thread like this one?

Craig

Hi,

and what about doing similar for sony 1292 (streaking issue ;))

Anyway, wonderfull stuff mike.

--Patrice

but .. but .. but ..
Mike, on the (my) convergence board the 24 1.2 ohm resistors are already in parallel pairs, 2 for each 6 output transistor sets. There are no piggybacks, and the schematic shows the parallel pairs, each resistor with its own part number. So are you saying to replace each 1.2 with a 1.2 ohm 1/2 watt fusible version?

The 24 1 ohm resistors (which look like 12 1 ohm piggybacks) are also shown exactly that way in the schematic, as individual 1 ohm resistors with unique part numbers. Are you saying to replace each unique resistor (retaining piggybacks) with 1 ohm 1/2 watt fusible, or each piggyback pair with a single 1/2 ohm 1/2 watt fusible?

On the vertical board, there are six 1 ohm resistor in piggyback pairs on the + supply rail. The - supply rail has a corresponding single 1 ohm. This is exactly how the schematic shows it to be, including unique part numbers for each resistor in the piggyback pair. Are you suggesting that both rails be replaced with a single 1 ohm or single 1/2 ohm 1/2 watt fusible?

Are all of these replacements due to potentially flakey resistors? Or design/performance considerations?

I called MicroOhm and was told their FRN series was special order only and 4 weeks out with a 1000 item per value minimum. Are there other options? No one else I've asked has ever heard of them.

On the switcher, unfortunately I'm pretty much stuck with using it due to the number of inputs I have. You mentioned the chips 'and other components' -- is there any documentation that goes into details of what else to replace, or is that a subject of a future thread here, or can you detail it?

--Bill

Bill,
I'm not sure why Electrohome stacked the resistors on the later boards. I've never liked the ones that's in there. Just let your projector warm up and hit them with a can of freeze spray. The single resistors will be effected to some degree, but the stacks will be stable. I was using another brand of resistors in this circuit that was .47 ohm. Tim Martin found the MicroOhms for me, and I've been using them since. Excellent resistor, very very stable. In fact, they are so stable that you could even use one rather than two.

It's been awhile since I seen that switcher, but if it has the CLC409 in it, you should replace them. You'll also have to make a few other changes but get me a diagram of the circuit and I'll get you through it.


Later on this thread, I'll post some patterns, and I'll show what happens when the electronics and the coils/magnets are properly tweaked and tuned. You'll see why it's important to have the electronics at peak as well as being able to work with the coils and magnets. The Marquee is very easy to setup, and it can be done by almost anyone. But we're going to look at it on another level.

Now, I'm a little low on my resistor stock, but I'm sure some of you have purchased the 1000 resistors for your one Marquee projector. So that we all can get up to speed, how about sharing or selling a few to those who don't have them as of yet!?

Mike, thanks for the clarification. So the bottom line is to match the net values with better resistors single or dual. Are the MicroOhms you like only the FRN series, or can you suggest more readily available substitutes? And where *do* you find 1/2 watt resistors only 4mm in length?

Unfortunately the switcher is one schematic I don't have, but will try to locate it.

Looking forward to the next installments!

--Bill

Originally posted by mp20748
Starting next week we'll look at a fully done Marquee 8500LC. This Marquee will have a complete set of VDC rebuilds, to include all of the suggested tweaks, mods, maintenance and a few special tweaks ;). Thanks to my friend James Bond (Bethesda MD), this display will be done using his Home Theater setup, and he's giving me permission to do an open inventation to anyone who'll like to attend.

As the owner of an 8500 with wife permission to start building my basement theater within the next 3 months, I would LOVE to attend to actually see both the modded vid card as well as a tweaked Marquee in person.:) :) :)

John

Hello

My order of 1000 1 ohm 1/4W FRN resistors arrived yesterday. I plan
to use 350 or so of them on my two Marquee PJs and my buddy's
Marquee. Total cost was $69.07. I am willing to sell the remaining 650
for 7 cents each plus package and shipping. They are loose bulk, not
on a tape or reel. PM me here or email at dhoward@oasisoainc.com
if You are interested.

DH

Fun Thread. I have a horrible dilemma !!!

While I can chime in on TONS of stuff on this thread I have to stay quiet cuz the changes I make and know how to make are worth money to me.

IT SUX !!!!!!!!!!!!

It really sux...

So while I would love to offer just loads of mods and changes and insight I can't. Im sorry.

One thing I do have to point out. In the replacement of the carbon comp resistors in the HVPS... These are carbon comp for a reason. A number of reasons. These really need to stay carbon comp. Look at the voltages across them. This is a carbon comp thing guys. BE CAREFUL.

I recently aquired a Marquee 8000 and I noticed severe streaking from the Blue tube at 1024*768 resolution with absolutely no streaking on the other two tubes. The neckboards are all equipped with the H1100 op amp
and the two metal can transistors. There is no sign of any heat burn on the resistors and all of the solder joints look fine.

Form prev monitor repair experience this streaking was mostly due to bad coupling capacitors.

Any ideas are welcome.

Laz

Chris, Having seen your projector at CEDIA, I can agree...it sux that you can't chime in with us :D

Originally posted by Chris_Stephens

One thing I do have to point out. In the replacement of the carbon comp resistors in the HVPS... These are carbon comp for a reason. A number of reasons. These really need to stay carbon comp. Look at the voltages across them. This is a carbon comp thing guys. BE CAREFUL.


I know the voltage.
The OX/OY Ceramic Composition
should be a good replacement.

From there spec.

"The OX/OY Series of fixed ceramic resistors are ideal for circuitry associated with surges, high peak power or high energy. They offer enhanced performance in high voltage power supplies, R-C snubber circuits, and inrush limiters. The OX/OY resistors can often replace carbon composition resistors which can be difficult to source. "


The OY has a 400V working voltage and a 20KV peak voltage
there is 2 in series giving the circuit 800V working
looking at my drawing there seems to be a cap with a voltage of 250V feeding it and the there is other caps that seem to be only good for 100V

So did I mis something???
or is there a error in my drawing??

The specs look very good for this part..

Fun Thread. I have a horrible dilemma !!!

While I can chime in on TONS of stuff on this thread I have to stay quiet cuz the changes I make and know how to make are worth money to me.

Chris, think of it this way. It's very unlikely that the folks here would represent a very high percentage of potential customers for you. And for the small percentage that might, only a handful of those might have the skills to do these things.

Unless a competitor of yours in the same city or surrounding area is also posting here it seems like you'd be pretty safe!

[Of course there may well be other points of view -- I'm just trying to coax you into sharing some 'secrets'! There's a lot of pretty sharp people here, and you would probably benefit from it as well.]

--Bill

Laz,
Q13 is more likely your problem. It's one of the six RF power transistors that's mounted to the heat sink. You'll have to desolder the leads from the PC board and perform a transistor front to back test with a meter.

Chris,
thanks for stopping by, and please feel free to contribute whatever. I understand your dilemma, because I've struggled with the same thing with this thread -- I have a bag of tweaks myself;)

Our intent with this thread was to focus mostly on "maintenance" and "upgrades", but I would welcome anyone that wanted to add whatever they wanted. The only thing that we'll try to avoid is any mod or tweak that could create a safety issue. That's why we did not get into the HVPS section, as you seem to be familiar with the concern there.

Originally posted by jlmclemore
As the owner of an 8500 with wife permission to start building my basement theater within the next 3 months, I would LOVE to attend to actually see both the modded vid card as well as a tweaked Marquee in person.:) :) :)

John

If you have to ask your wife's permission, you can't come over.

Originally posted by James Bond
If you have to ask your wife's permission, you can't come over.

I didn't need permission to attend, Only permission to spend an excessive amount of money to build my theater. I've learned she can get little fussy when I forget to mention the "little" purchases I make.:D

I was wondering if the fan tray can be removed from the top??

this is what I am going to try to do.
I have a 8000 on the ceiling. and it is not coming down. I got a 8500 to upgrade it. so I was going tr replace all the parts but leave the chaise hanging. it looks like I can slide out the CRTs one at a time.

After that, can I get the fan tray out ??
I would like to replace the fans.

or am I kidding my self..?

Hi Mike

You mentioned as Q13 being the suspect, is Q13 part of the G1 drive amplifier. Where could one find this particular high frequency RF transistor as I came up emty handed when I contacted my local radio repair dealers.

They seem to carry all the other MRF transistor except for this one.

Laz

Hello

The FRN 1 ohm 1/4 watt resistors I offered earlier in this thread are
all gone. Five of our forum members bought them.

DH

Laz,
Q13 is part of the final output stage. It's one of the four power transistors that are "a mirror cascode differential amplifier" (two complimentary cascode amps NPN/PNP).

These transistors were made by Motorola, the NPN (MRF548) may still be available, but the PNP (MRF549) is discontinued. Motorola sold off their existing lot, and they are very hard to come by.

I would suggest that you contact Tim Martin for a replacement board.

Well it looks like Q13 is fine,

I desoldered the pins and did a transistor check base to emiter and base to collector 0.7V one way and open when reversed and emiter to colector open both ways

Could it be a leaky cap?

Laz

Sly, sure seems like it would be a lot easier to replace the whole thing intact. Don't you have 3 or 4 friends that could help? I would only consider the component replacement on the old chassis as a last resort.

I mounted the 9501LC here on one of the Chief motorized lift platforms. The cost of the Chief was something I agonized over for a while, but after installing the projector from the floor to its operating position by myself, then later dropping it down for some cleaning and returning it to operating position and to finding all physical settings intact, made me realize it was money well spent.

You might want to consider adding something like that if you take the chassis down, since no pj is forever.

--Bill

Originally posted by Aksor
Well it looks like Q13 is fine,

I desoldered the pins and did a transistor check base to emiter and base to collector 0.7V one way and open when reversed and emiter to colector open both ways

Could it be a leaky cap?

Laz

At this point I'd recommend that you replace that board. The problem could be any number of components on that board. And you'll have to be very careful, because a slip could take out the tube. There are both safety and protection circuits on that board, and they operate around that same differential circuit, and they have 'critical safety components' that are necessary for proper operation of that board.

another source is Curt Palme. He has "new" reject neck boards that have good transistors on them for $10 each plus $10 shipping (when I bought them a few months ago)

cheers,
bruce

Can anyone give me some readings of a good MRF549 transistor located on the neckboard.

I got a few neckboards and some of the transistors collector emitter reads as open and the same if I reverse the probes but some others read open in one direction and a high value the other direction.

The base emiter junction is 0.6V and the base collector junction reads 0.6V when forward biased is the only consitent reading i can get.

I know collector emiter should be open in both directions unless the transistor contains a diode between the collector and emiter terminals.

As I have no specs on the MRF549 its hard to tell what is good.

All of the readings were performed with a Fluke DMM on the diode check function.

Laz

Well I took out my CRT's out of my 8000.
And the fans come out very easy. so I do not have to take the PJ down. the 2 things I have noticed is my CRT's show no wear whats so ever. I have a lot of time on them.
And none of the ones at work have ether. So I was thinking how come I have never burned the CRT's on any of the electrohomes over the years...
the closest thing was on one of the ECP 4500 we used along time ago I white window poped up for 1 or more days could have been a week no went in to the room and the PJ was not in standby and a window poped up.
and that only just left a hint of a mark that seems to go away after a few weeks.

The only thing I can think of is I run them fast the ones at work run only at hi refresh rates and at hi rez. 960x680x112Hz
or 1280x1024x76

my 8000 runs only 1920x1080x72Hz
and 1920x1080x60Hz interlace

could this be a factor???


The 2nd thing is the air direction the fans blows from the bottom of the PJ to the top. This works fine if the PJ is right side up but if it is on the ceiling would it not be better to revers the fans so the air goes from top to bottom aka (the air goes up). So it it not fighting the convection??

Or if I am to keep the air flow the same would a remote blower blowing into the PJ be the best. This has a lot of advantages one the noise but the best is I could put a air filter on the blower so my PJ never gets dusty.



Any thoughts on the matter??

"The 2nd thing is the air direction the fans blows from the bottom of the PJ to the top. This works fine if the PJ is right side up but if it is on the ceiling would it not be better to revers the fans so the air goes from top to bottom aka (the air goes up). So it it not fighting the convection??"

Sly, I plan to cover the cooling system on the Marquee. You'll want to leave it as is, I'll explain later. You have the most advanced cooling system of any CRT projector manufactured. You'll see why the Marquee was the ONLY CRT projector that would handle very high temperature operating 24/7 environments, with only a neck board failure from time to time, and I mean very high temperature rooms, and this is where every other CRT porjector failed. Even the Sony 1292 with its 2000 fans.

We'll pick up on the "Maintenance" and "Upgrades" after the weekend gathering. I'll throw in a couple more nuggets, and I may reveal how to hot-rod this projector (but I don't recommend it). It's easy and it'll make the picture look brighter and punchier, BUT you'll have to see what it does to the video chain, and hopefully I'll have the scopes hooked up for this. The best way to improve on the performance of a Marquee 'IF' all is well with the electronics and tubes, is a proper setup and the input source is the most important factor. If there's a certain level of noise (hash) in the input source (processor), it will increase on it's own along the video chain. And no amount of caps or mods will correct on it, because it's a part of the signal and the bandwidth of the projector is such that it'll pass it along with the signal, because there are no signal filtering networks in these projectors. The goal should be to look for a cleaner processor, rather than looking for a miracle tweak or mod. I have a bag of mods and tweaks myself, but they'll only make a difference based on these four things: Source, cables, setup and the condition of the projector. I'll demo this on next Saturday with 'My' transcoder and a HTPC with 'MY' Mp-1 mod. We'll look at (clean) DVD and HDTV sources. The 8500LC at the gathering will be Special ;)...

The gathering next week (5-17-2003) will be in Chevy Chase Maryland, not Bethesda. It will be right off of 410 (east west highway). Shoot me an email if interested (mp@wisdom-technologies.com). We plan to start around 3PM.

We have much more to come -- stay tuned!

Here it goes. I replaced Q13 on the neckboard with a new transistor.
streaking problem seems to be gone but for some reason when I turn on the projector with no input connected I can see the red and green raster but no the blue.

If I bring up a test pattern the blue pattern is visible on the screen


I switched neckboards between the green and blue the same thing. I then measured the G2 voltage on both the green and blue neckboards and the blue looks like its 100v less than the green.



Could this be a HV Power supply issue as it I could see a raster before I screwed with q13 except I had really bad streaking.

I need some of your expertise in this issue

Laz

I just did another check it looks like a G2 issue I coonected the green G2 to the blue neckboard and the Blue G2 to the green neckboard and I can now see a raster on the blue tube that is the same intensity as the red.

I measured the red G2 and its is the same as the green tube.

Is the G2 level derived from a voltage divider inside the HV power suply?

Maybe I have a bad resistor

Laz

You know that G2 is adjustable (per tube) in the service menu right?

Cary