CurtPalme.com Home Theater Forum

[Nashou66]

Register to remove this ad. It's free!

Yes, they get thm from Farnell and charge 15 bucks!!!

I'll have to search the other us companies.

I dont think the chips would be better since they have no third connector for ground, i like that ground connector idea more that just a chip.

Athanasios

[Nashou66]

Ok here is a link to a short video of the problem. Ignore the yellow/green horizontal bars as thats from the video camera, the problem is the
blanking out of the image. When it shows the security cameras its fine then when it goes to him tying his shoes it gets the horizontal black
blanking effect.

EDIT Video Removed.



Athanasios

[antorsae]

Very weird. A whacky idea maybe: I understand the CLB provides the menu overlay which are sent to the VIM, where they are mixed with the video. It looks like the picture is not loosing sync, so either the contrast/brightness controls are getting crazy or the CLB is sending crazy video overlay (as when you turn on any menu) or overlay masks.

What happens when you display a built-in menu?

[Nashou66]

I agree Andres, After studying the service manual over and over I was looking at a few things and that was my first op amp to look at was the one that has something to do with G2 settings op Amp U105. I didnt change it out as I think its ok but I did change out op Amp U23, I was getting some conflicting intitial
reading from it comparing to my other identical CLM. This I did after i removed all higher value Tants from the 6 convergence amps and op amps U105 and U23 and also from Dacs U37 U39 and went to .1uf tants (original value) these last two op amps control G2 and horizontal and Vertical waveforms respectively. the two dacs send the waveforms to the U23. So my next move is to change out U105, the G2 voltage op amp. After that I was toying with your theory of the internal video generator but this Problem May be on the VIM. As I said before I only had this problem when using my modded VIM which up to the point of changing out the 16 caps to the 22uf caps was working fine, although it was with the 47uf 6.3 caps I accidentaly used before changing them to the 22uf. This is why i think its the CLM unless the CLM some how damaged the VIM but that theory seem remote to me. it be either the SD540i chips or the AD834, but having all three go at once? Last night I watched Casino Royal(it causes the most issue than other DVDs) with my unmoded vim and I had no issues. Although it does blank out each time I changed the aspect ratio control or put up the lumagens onscreen display, but only for a split second, this I think is just the normal blanking of video from the bandpass switch control, like when changing scan rates or resolutions as I have seen it many times before.

To answer your question of the internal patters. I would pause the scene that would get those partially blanked screen and that would hold, meaning the patially blanked out screen will still be there flickering. It is also there when i switch to the internal video patterns. this is why I want to go after the G2 control amp next. I dont think its U103 that sends the patterns for the internal Video though, its more like the clm isnt sending a steady signal to hold contrast or G2
up to the levels it should be at. But U105 had the same internal readings pin to pin with my unmoded CLM's U105. I need to put the video video rca to the scope during the "blanking" problem I'll call it. I'll also do ti with the different VIMs as I am beginning to think it may be the VIM and not the CLM. do you think that maybe the CLM when I had the smaller valued voltage caps on the G2 op amp may have damaged either the multiplies(AD834) or the contrast control on the VIM? i hope its not the AD834's they are 35 dollars each!!!

If you have any more Ideas please let me know.

Athanasios

[Nashou66]

Ok Did some more testing, Hooked up the scope to compare internal patterns with both Vims, they seem similar just taller (higher amplitude?) waveforms on the modded vim. not related but what i realized is why they call the stair step pattern that, when putting it up on the scope its a STAIR STEP!! lol I now it should be obvious to Electronic guys I'm still learning. Ok So back to what I tested next. I brought up the opening bathroom scene when 007 is fighting it brings up the problem, so i decided to lower the contrast
to see what happens. sure enough at around 10 it stabilized. I went forward a few clicks and raised the contrast, this time at 17 it destabilized . So it is something either on the Vim or CLM contrast wise affecting it. I say my modded vim. but to be sure I will change out op Amp 105 anyhow just to make sure. It might just be the modded CLM and Modded Vim are too much for the LVPS.

I only have 10uf caps for C2 and C1 I wonder if i add more capacitance there it would help the other caps at each chip?
Or is this line of thinking wrong ? Experts ?

Athanasios

[Nashou66]

Still trying to figure this out. Added another video. This time using my Lumagen scalers internal patterns as I found out going from the every other Horizontal line to the every other vertical line pattern causes it to blank out, also on the full white field on my lumagen. I also have the LG Bh100 blu ray player on so the lumagen syncs to its signal to output the 1080p signal. When I have no external signal on and only use the Marquees internal video it will not blank no matter what pattern I put up or how High i put contrast.

CLM Problem Video : second one

EDIT:7:22pm I decided to try one more time with another CLM I have even if the setting are not the same.
With the modded VIM it wouldn't even turn on, no Video, the crackle would happen but no video. So i put the Un modded VIM in and it turned on. put back the Modded VIM no go. So I think its the modded Vim not the Modded CLM which is good news !!now this is where i need to investigate more, either Something must have been wrong before with the vim before the CLM upgrades or when I had the wrong voltage caps in on the CLM it did something to the Vim.
I added a video of the scope on the Good vim to show how the scope pattern for the every other vertical line should look.


Athanasios

[1031]

Have you cheked are those two 3 amp fuses ok on your modded vim? There are also two fusible resistors for +-15volt lines. Those all are located near multipin connector.

[Nashou66]

Jarmo, we think It might be one of the following, the X-OVER-I, the SPOTKILL , BAND-SWITCH or SCANFAIL. And the outputs from CA339. These have specific voltages that if they either go higher or lower will cause the image to blank out. TSE also said that :

[AnalogRocks]

[AnalogRocks]

[Nashou66]

[PaulB]

Athanasios,
The SMT tantalums under the blue balls, you have them down as .1uf 16v but if I'm right (!?) the markings on them, V104, indicate that they are .1uf 35v - see the pdf here http://www.acaltechnology.com/download.php?id=110006 (3rd column in, last page).

Does that make a difference at all? I ask because I can't find .1uf 16v SM Tants anywhere.

[Nashou66]

Good eye Paul !!
As long as they are higher than the 15 volt rail, Higher voltage is actually a bit better. Get 50 volt .

Athanasios

[PaulB]

Excellent Athanasios

Not so much a good eye as desperation trying to find the right part - not knowing what I'm looking at seriously hampers me but I'm learning!

[Nashou66]

Paul, you could leave the ceramic caps there if you want I just used the tants since i have a crap load laying around. the next CLM I do for my other PJ I plan to try it leaving allt he ceramics there and adding the other caps on top, I think i mentioned that in the original post. Mcpherve even sugested adding another 100pf cap to those as well, I just dont think there is enough room especially if using radials as well as SMT's.

Athanasios

[PaulB]

Understood, there are many ways to skin a cat! I must admit I'm inclined to just put the blue rads on the existing smd's. Its easy enough to replace them should a better variation be found. Not sure how much better one method over another would be though, difference may be minimal pq-wise but as a complete ignoramous on the subject, I may very well be (and most likely am) wrong.

I've stopped doing the mods for the moment as you are aware but I think that with all the experimentation yourself and Jarmo are doing, its probably better to lag behind a few weeks to see if you guys have a brainwave about something you did earlier and decide to undo days worth or work! Especially those 126 caps for the CLM, I ain't doing these 'til you're happy it all works - your problem with the VIM scares me! I know its probably not related to the CLM work but it did raise another alarm bell of the dangers of doing this mod stuff.

Hope the VIM gets fixed soon, finger crossed my friend.

[Nashou66]

I agree with you on your second point about the Vim, I am confident though we'll find the reason, it could be thats it just the older style 2006-03p vim. I have not done any mods to the newer Vim as I want to do that with a camera and document it for the thread, also It will be tested in use for reliability before i post anything. The CLM mod was tested with the unmodded vim and I only put the modded Vim in as I saw it under a bunch of part boxes on my bench, I forgot I had it out of the PJ. Only when I put it in and first put in Gran prix and then Casino royal did i discover the problem, i assumed it was the CLM as the Vim worked before with half the caps changed on the CLM. Now it could still be a contaminated control signal is making its way to the vim that could have its origins from the CLM. The signals for the current limit, spotkill and scanfail come from either the CLM,VDM or HDM. That is why i need to make an extender card to be able to scope the VIM easier than connecting wires to those test points. I'm hoping the Striped PCB card comes from Farnell today when I get home so I can start making the extender card.

Athanasios

[Nashou66]

Well i made my extender card . Any one who is testing and doing mods and has a scope and wants to measure points in the marquee should make up a few of these boards. I am going to make a few more for testing the CLM as well.

Mentioning the CLM i have good news so far, I do not think the CLM is the culprit for the blanking(spotkill) going on. I measured the test points Scott told me to on my Modded vim and a few of those points would cause spotkill to happen. So I put in my good vim and no trouble at all, every test point was fine. So it looks like something went on my VIM, I think i may have shorted something on it when it was under the pile of stuff on my bench( a piece of stray wire or solder may have gotten on it). I have watched a few movies with the other VIM and it was the original VIM I had in when I completed the entire CLM mod so I am almost sure the 22uf caps on the run of op amps did not cause the failure. I will put them back in and test again with my good vim. When I took out the 22uf caps I replaced them with .1uf tants. I will keep those in and add the 22uf's as WTS's suggestion that the lower value cap helps out in the high frequency filtering.

Athanasios

[PaulB]

[Nashou66]

On the +/- 15 volt rails there is a run of op amps 8 all together, 6 are right in a row and the other two are in different locations, one next to the stig board connector and the other one U23 i believe I think its a tl084 those are on the 15 volt supply rail if you look at the schematic . those are suppose to be 22uf tants (blue ones) I accidentaly put the lower voltage (6.3) 47uf caps there oops!! the 22uf's are 16 volt or higher, you could even try 33uf's as Ken Hottie mentioned in the original AVS post.

[PaulB]