CurtPalme.com Home Theater Forum

[mamawaldee]

Register to remove this ad. It's free!

My PJ has been stored under a blanket, but I used to uncover it every few months and fire it up for 20 minutes or so to keep the cathodes fresh. I used to do this with a source connected and verified an image on the tubes. Somewhere along the line I stopped using a source and checking the tubes.

One time I forgot to remove the blanket from it, and it apparently overheated. By the time I remembered my mistake I ran downstairs to find it had shut down, but it was already cool. I fired it back up and it sounded normal, as in fans running, etc.

So I figured it was OK, until one time I put a source to it and discover no image. So I don't really know when it quit working exactly.

Filaments light up, all voltages seem OK, but I have scan fail and no HV. EHT hold down is not active.

I emailed Curt with the above summary. Curt sent me an SMPS, and the two deflection boards from a working 1209. I swapped all three in there and get the same result. At this point Curt mentioned the decoder could be involved but he is too busy, so sent me here.

I have the service manual for the set, I'm also a EE with access to all the tools at work to service boards, but I do not have extender cards.

So far I have traced the scan fail line through all of the schematics to become familiar with it, what I see is that the decoder is the last stop prior to EHT. It looks for the presence of the 5V digital supply and if the 17V is present without the 5V present it will yank the scan fail line low. So I pulled the decoder from the set and the scan fail is still there. The only other sources for pulling scan fail low, other than on the frame itself, are the horizontal, vertical, and G2. I don't see anything on the frame schematics that could cause scan fail.

Recall, SMPS, H and V are all swapped, so with decoder out, that tells me the scan fail source is on the G2 board itself.
Scan fail main pullup is on G2, so the rail originates at G2. It has to loop through the vertical board as the first stop and then goes to the horizontal board. So I put the decoder back in, and pulled horizontal and vertical boards out.

The set acts almost the same way, I get all of the SMPS voltage LEDs OK and scan fail on the G2. The only difference is the set refuses to power down normally, it cycles on and off with a period of a few seconds until I kill the main. I put all the boards back in and get back the original behavior, as in it will power on/off normally.

The G2 diagnostic LEDs do the following:

On power-up, scan fail and everything but HTHD comes up immediately out of standby. Then HTHD comes up full, then goes dimmer and with slow pulsation. You can hear some faint ticking that may go with the pulsing, sounds like modulation of magnetics in the SMPS, the kind of audible change that happens when tubes go from black to displaying an image.

So I have the G2 board on the bench, nothing is visibly wrong with it on the top side. But underneath this G2 has a lot of rework on it already with traces cut and extra diodes attached to it, but it worked fine for years. I checked a couple power transistors and don't find any shorts, but have not looked at it further.

I figured it was time to post here at this point and try to get some advice.

Thanks in advance,

-Scott

[km987654]

If the deflection boards are Okay and thats an assumption even though you have changed them then its likely to be in the HV section so EHT, Quad or splitter or a combination of.

You should hear a little HV crackle as the set powers up. Do you hear that?

Of course the deflection coils form part of the scan fail circuit so there could be a problem there also.

[mamawaldee]

Trouble is I don't really remember what it normally sounded like. There is no HV crackle, but there is a chirp which is either the SMPS or the HV trying to start, and then the scan fail kills it. I would need a multichannel scope hooked to it to get a picture of the exact sequence. What is bugging me now is the G2 has scan fail illuminated with the deflection boards out. In this case the HV should be off, but scan fail should be off too. Unless the failure is on the G2 board itself. I checked the -170 and 210V comparators and related parts and everything measures OK. Other than power up the TL084 circuits on the bench I can't test that circuit live. I'm going to disconnect the pull-down transistor on the G2 board that hooks to the scan fail line, so G2 is unable to pull scan fail low. Then I'll put all the boards back in and find a good spot to purposely disable the EHT generator so it can't burn the tubes. I'll look for a fuse I can pull that will do this properly. Then fire it up and see if the scan fail is gone. At least then I'll know what board it's on.

[mamawaldee]

Here is an update,

I pulled the replacement H, and V deflection boards, decoder, and G2 and attached test points all along the scan fail loop driver transistors. The scan fail is coming from the H board only. Next, I added more test points all over the H board. By turning P2 on the SMPS all the way counterclockwise, I can start the machine and avoid deflection hold down on the H board. By increasing it very carefully I was able to get the HV to come up.

It was a little more complicated than that however, its a combination of run-time and having multiple sources connected to the machine. By changing from port 1 to port 5 back and forth and adjusting P2 with it running about an hour. Most of the time the hold down will kick in and require a power cycle to clear, but on a few occasions the HV would come on.

When this happened it would last from a few seconds to a minute, and the images on the tubes for red and green are sharp, the blue is a weaker and less focused. When the HV is starting to die, the video gets all torn up, but the deflection remains good (stable raster). Spot killer is working, and the tubes are not damaged.

During this experimentation I swapped a few boards, my original H board works the same as the replacement. My original V board will not produce an image, all I get is an unstable ABL line. My original SMPS will only make the tubes flash, and there is a distorted image in the flash. I swapped back to all replacement boards, did one final test with stable images, and called it a night.

This success never repeated. The following day with the same procedure all I got was flashes from the tubes, and much more difficult to obtain even that.

I pulled the convergence board, that just made the machine shut off after a few seconds. Power it back up, it shuts off again. H-board hold down and scan fail were the same regardless. Put convergence back in, pulled the H shift. Same H board behavior, but no more shut downs. Pulled Port 3, G2 and decoder at same time. Result was same H board behavior, hold down and scan fail.

During this, I found that the heatsink on the decoder is roasting hot. I found the thread on AVS forum where somebody had this same problem on his 1208/2 and there was an open resistor feeding one of the three regulators on that heat sink. My resistors are good, and the board looks pristine otherwise. I think something is toasted on this board, but Curt does not have any RGB driver/quad decoders in stock anymore.

I know its a long shot, but I'm asking if there is anybody in the greater Detroit area that might be able to assist in debugging this machine?

Regards,
Scott

[km987654]

[Pokemon640]

i have some spares for your unit, i have a TON of cards in my storage and by accident i purchase a pair of cards recently.
I got R7617481 RGB Input Controller and R7621745 RGB amp + R762174S (Quad decoder)
Record a video (with sound) and upload it to dropbox, to check the problem easier. +record the diagnostic lights.

~Andreas

[mamawaldee]

Hi guys. I appreciate the help and advice. I'm presently securing some more replacement boards from within the US, but will certainly keep you guys in mind for if/when the machine comes back to life.

Some updates, my machine will not run with the convergence tray removed. It will power up via the built in remote, but shuts off after around 10 seconds. Several people have told me this is not normal, perhaps that is true for an otherwise working machine, or a machine with a different software revision. I have the convergence card on the bench and already checked the 12 output devices for shorts, everything looks OK thus far. After checking a few more things I plan to reinstall it with the power supply jumpers removed to eliminate loading on the G2, but leave the I2C bus still alive to see if it will run that way. Maybe the software shuts the machine down if it sees the Bella chips missing on the I2C bus.

Interestingly I found this on the forum where another guy experienced the same behavior:

http://www.curtpalme.com/forum/viewtopic.php?t=41565

Unfortunately that thread never stated root cause.

@km987654, my service manual shows the scan fail line looping through the decoder just prior to the EHT, so I'm not sure how the EHT can come up with that board missing. I could make that circuit with a jumper and run the PJ off RGB otherwise, I think that would work. I wouldn't want to do that with the EHT in there though just in case. I still have good tubes.

@Andreas, I have the older style quad decoder setup with the two big boards sandwiched together. In the manual those numbers are 7621175, section-J in the manual. I don't see the number you are referencing. When I got boards from Curt, the freebie he sent me was something not found in my machine. In the service manual that extra board is found in section 2719-RGB-IN, full number R762719. It looks like an RGB switcher with the quad decoder attached to it. Anyway Curt told me not to plug that one in there, it would blow things up. Maybe its for a data model machine.

Lastly, I noticed my convergence output schematics are missing the dynamic voltage booster circuits +/- VDYN which are shown in the block diagram on page-7. The block diagram shows a charge pump topology and there are six power device on the heat sink which do not belong to the power amplifiers. Two of then are in the schematic and are diodes, but the four transistors are not anywhere in the schematic. Somehow I think I'm missing a page and was wondering if anybody might have it.


Regards,
Scott

[km987654]

In terms of the decoder board its clearly different to the 8 series projectors but i would think a loop through cable attached to one or more connectors on the RGB input and switching board would deal with that. Of course you could just replace the decoder.

[Curt Palme]

I sent Scott an email, but I was thinking:

It's possible that the scan fail is fixed. If the master H width trimpot on the power supply causes the SF light to go on and off at distinct points of the trimpot travel, then chances are the SF line is doing what it is supposed to do. generally the red SF LED goes out completely when the projector is working, but perhaps on this set it is lit dimly when things are working normally?

The EHT board has several fusible resistors on it that can fail for no reason. If any of them go open, the EHT board will go dead.

Just a thought.

[Curt Palme]

I sent Scott an email, but I was thinking:

It's possible that the scan fail is fixed. If the master H width trimpot on the power supply causes the SF light to go on and off at distinct points of the trimpot travel, then chances are the SF line is doing what it is supposed to do. generally the red SF LED goes out completely when the projector is working, but perhaps on this set it is lit dimly when things are working normally?

The EHT board has several fusible resistors on it that can fail for no reason. If any of them go open, the EHT board will go dead.

Just a thought.

[mamawaldee]

Taking videos is possible of course. If the next set of replacement cards does not make any difference I'll make videos with a scope connected as well.

Does the HV come up if there is no signal connected to the machine? I can't remember how it used to work. What I do remember is that the machine would not generate an image if there was no source connected, not even from internal test patterns, But I also think I remember there being ABL lines on the tubes without a signal connected. Just ABL lines, no visible raster. I'm trying to figure out why the decoder is in the scan fail loop and what damage could result from defeating it.

[Curt Palme]

Yes, the HV will come up with no signal. The 1208 chassis is sometimes a bit weird, some won't fire up without the convergence tray, others will. Ditto for no image/menus with no input signal.

[Curt Palme]

Yes, the HV will come up with no signal. The 1208 chassis is sometimes a bit weird, some won't fire up without the convergence tray, others will. Ditto for no image/menus with no input signal.

[mamawaldee]

OK, latest update, I checked all of the power semiconductors on the convergence output and added a test point to the current limiter signal that feeds back to the convergence driver. Couldn't find any bad devices, and I partially traced out the boost converter stages which are missing on my schematics.

Also, the Port-3 and H-shift boards are out of the chassis.

Replaced the board in the tray, and tested. No change, so I didn't break anything. Then pulled the jumpers which provide power to the convergence driver. Tested, at this point the convergence load is gone on the G2 board and it is completely quiet, no more ticking sound. And it keeps running, no more shutting down like when the tray is removed. So the issue must be lack of I2C communication to the Bella chips.

No change regarding the scan fail though. So I pulled the tray and pulled the jumpers which provide power to the dynamic astigmatism power amplifiers. No change to the scan fail condition. So at this point I'm satisfied that the convergence tray has nothing to do with it.

I have test points on every circuit that can produce scan fail. If I turn the horizontal width trimpot (SMPS-P2) counterclockwise, the G2 board asserts scan fail. Somewhere about 1/3 turn counterclockwise it will release. At that point the set will run without power oscillation behavior. I then measure the horizontal deflection detector voltages. All three of them are approximately 12V. Then increase P2 slowly with a computer source on port-5. At some point the horizontal deflection detector voltages decrease and begin to bounce between 12V and a low voltage. At that point the HV comes up and I get flashes from all three tubes that seem to coincide with the signals on the deflection monitors. The scan fail LED is going from bright to dim, but not quite off.

I repeated this test with the 15kV lead to the G2 board unplugged. It made no difference. While the tubes flash, there are distorted images in the rasters. Blue image is definitely weak and defocused.

So I pulled the G2 board and the EHT. Now I'm just measuring the scan fail detector voltages to the remaining boards while playing with the P2 control. All of the same signals are on the h-board. I pulled the switcher and RGB driver and decoder and then the SMPS makes chirping sounds. It needs those two boards to complete enough of the feedback loops to remain stable.

The decoder contains the Bella IC which controls the horizontal drive. Maybe that has something to do with it. It's like the drive signal is too weak to keep the deflection detectors to stay off, but increasing P2 too much causes the hold down to engage.

Nothing else to do with it until I get more replacement boards. I'm leaning towards the decoder somehow causing this, but the scan fail line from that board never gets asserted.

[mamawaldee]

Minor correction to the above, regarding the convergence tray, the convergence output has it's power jumpers disconnected, not the convergence driver. So all of the power amplifiers in the convergence tray are now disabled, but the signal generators are active. Hence the convergence tray is effectively removed from the system. The controller board can communicate over the I2C bus and is no longer shutting the machine down. So this particular quirk appears to be software related. Probably because the dynamic astigmatism is an add-on to the 1208/2 the software is not prepared to not have the Bella ICs missing so it just throws an error and shuts the machine off. Regardless, with all of the power drain in the convergence tray gone, that maddening ticking sound the G2 makes is gone and makes debugging a lot more pleasant.

[geisemann]

Hi Scott

Boards on the way.


www.eisemann-theater.com

[mamawaldee]

Awesome, thanks Greg.

[geisemann]

Liquidation of parts

If anyone needs Barco parts or projectors I am liquidating parts and projectors send me an email but not on here. Mods of barco projectors are limited now as we are moving to Digital Projector improvements. Mostly Sony SXRD.

Have shelves and shelves of barco parts.

www.eisemann-theater.com

[mamawaldee]

OK, the latest update.

After receiving more boards and doing minor repairs on some of them, the projector is running, mostly.

The confirmed casualty list thus far:

RGB driver/Quad Decoder (blue channel damaged, no contrast)
G2 board: (convergence power supply wasted, also causes scan fail, but not from the scan fail circuitry)
Vertical deflection: (black rasters, rolling ABL on all three tubes)

Changing the above boards has it working to see the other things that are still not right. Here is what remains thus far:

Scan fail is enabled when machine is cold. Let it run for 5-10 minutes and scan fail goes away and HV comes up. Once warm it will come up reliably.

Ticking/snapping sound coming from splitter, along with image shrinking when it happens. Probably arcing internally.
Does it on power up, easy to see in the image. After power up, this arcing sound stops and the image is stable.
Tick occurs on power down also, but image is blanked so cannot see image effects.

I think I need an EHT, quad and splitter.

There is some random video artifact that looks like a single line wiggle that jumps around. It's not real bad, you have to look into the tube to see it. Probably related to power supply noise, deflection or the EHT. Hard to tell. Its on all three tubes.

I'd love to be able to buy a Moome card before the sale ends in three days, but I'm not sure this machine is worthy of it.


Curt, I sent you an email...

[mamawaldee]

Anybody know what happened to Curt? He no longer answers any emails.

[mamawaldee]

Well he got back with me. Looks like he is out of the game.

Going to make some want to buy ads.

[HaydnG90]

Time to buy those old parts mules. Luckily most CRT's are modular so replacing boards is a cinch.