CurtPalme.com Home Theater Forum

[perisoft]

OK, I decided I didn't like the motion detection; it didn't seem to be working reliably. So I wrote another bit on Focus!Focus!:



So, y ou can see the webcam window there, the crosshairs showing where I'm recording spots, etc.

It occurs to me that if you pointed the webcam at the tube right, it'd be a good way to make sure your phosphor usage is maximized - a lot easier to tell whether you've got the thing centered by doing that!

You can set how quickly it runs through the graph, choose whether to clear on wrap or not (by not clearing I can leave it running and it'll still show any incidence of dropped contrast). It's kind of specific and doesn't have obvious things like saving, scrolling, etc etc. But for now it works.

It might be interesting to use for doing astig stuff too, since you'd be able to see the history of changes, not just numbers.

Anyway, there you have it. So far no drop outs...

[perisoft]

Ah hah! I reproduced the problem while messing with my graphs.

It seems to happen when you change the contrast on the PJ! If I run the contrast up and down,
it'll drop the green out once the contrast hits about 50. In this graph, the first couple are just holding
down the contrast button and running it up as fast as it can go; one is tapping the button so it doesn't
rise as quickly. Rise time appears to affect the likelihood of dropouts, and as you can see, they flicker.

You're seeing the values drop to zero because I'm looking at the bloom from the tube, not the tube
face. It's not dropping to zero.



I checked the tube filaments now that I can trigger the problem - they stay lit. ABG makes no difference
on or off.

This explains why I was seeing it once in a while when testing the projector but not with a static screen.

I tried changing the signal into the PJ but that didn't trigger it - that said, I was turning green off and on,
not ramping it like I was doing for the contrast. That's the next step.

Ideas?

[CIR Engineering]

Probably not the tube

Neck card, B board, or a Y board is what I would guess.

craigr

[perisoft]

[CIR Engineering]

Yeah, the minicoax, not BNC.

As I said, swap blue and green neck cards to test the neck cards.

craigr

[perisoft]

[perisoft]

OK, I swapped the G and B neck cards. Now the G works fine and the B is *always* dim - very dim; I have to crank its gain all the way up and turn R and G gain down by 70% to get it close to balanced. Ramping up and down contrast on a full white screen doesn't affect any of the channels now - it kind of looks like what was an intermittent problem on that neck card is now a permanent one?

B doesn't seem to be messed up in any -other- way, though. Not flickery or bouncy, etc.

It's possible I didn't get something hooked up right, but I was pretty careful.

Next thing to check for?

[CIR Engineering]

So it's almost certainly a problem on the green neck card. I don't have any for the G70, and since this is Curt's forum you should probably see if he can service it for you. Or just see if Curt has one that he can sell outright. The neck boards are pretty much interchangeable between green and blue so he'll probably have one he can sell... and they are not too expensive either.

If you are at all good with soldering, you may want to look over all the solder joins on the green neck card before you write it off. There very likely could just be a bad solder joint, and a quick touch with the iron may fix it good as new.

Honestly, this is about the very best outcome you could have hoped for An easy to find board that is not expensive.

craigr

[perisoft]

[CIR Engineering]

[perisoft]

The neck cards all look the same. The R and G are marked (by hand) though; either there's a small difference or someone was cautious.

I swapped G and B; and like I said, the problem went to B but is no longer intermittent.

I decided to swap back to make sure there's not something else wacky going on; the problem moved to the green tube very nicely; still constant. It seems a bit worse than it did with B, but presumably they respond in somewhat different ways...

So yeah, pretty much I gotta get a neck card, I guess. Or get this one repaired. So, until I can find the time to get all that done, the Sony waits... Hopefully a fix will get it working well.

[CIR Engineering]

Neck card is no big deal. Check for a bad solder joint first, or just buy a new one from Curt.

craigr