CurtPalme.com Home Theater Forum

MYoung · Joined: 24 Feb 2007 Posts: 369 Location: Madison, WI

Well, now that I've obtained a NEC 6PG with minty tubes I don't have much to lose if I screw up wear-leveling my Runco IDP-980's green tube phosphor. I hooked up a PC to it, maximized the raster to past the edges of my screen, installed GIMP, put it in fullscreen mode, and carefully drew in the wear pattern, then applied a slight gaussian blur to feather the edges just a bit. We'll see how it goes! I'm probably just wasting money on electicity. Oh well. Any guesses on how long before I start seeing wear-leveling results? I'm thinking I might need to re-draw the pattern due to drift but we'll see.

**Curt Palme** · Posted: Wed Oct 07, 2009 2:51 am Post subject:

Nice, I'm REALLY curious how that works. Keep us posted. Contrast and brightness at 100%. Smile

MONITOR IT CAREFULLY!

perisoft · Joined: 29 Aug 2007 Posts: 2920 Location: Ithaca, NY

Ack! If you have a reasonably good digital camera, you should use that to create the wearmap rather than trying to do it by hand. If you do it by hand, you're going to end up with little brighty-darky-ribbons around the edge of the wear area.

There are a few tricks to using a camera to set up the wear map, though - you have to do some critical thinking, particularly in order to compensate for distortions in the camera. But it's the only way to hope to get a reasonable result.

Actually, there is *one* other way I can think of - to fire up photoshop full-screen, so you can draw on the wear area. Get yourself a wireless mouse, and get comfortable with the airbrush... now make a 50% gray (green) image, and go through and 'paint out' the worn area with white, so the screen appears to be a solid color again. You have to do this really carefully - basically it'd be a way to "hand-feather" the edges instead of just doing a gaussian blur like you did. See what I mean?

MYoung · Joined: 24 Feb 2007 Posts: 369 Location: Madison, WI

I agree that a taking a picture of the wear pattern, perhaps right off the tube face, and applying various tweaks would yield the best result, but I also figure that also requires exponentially more dicking around to get right.

I used GIMP in fullscreen mode and used a wireless mouse to draw the wear pattern. For some parts I turned down the contrast to 0 and peered through the lens to accurately select a point on the edge. I used the "paths tool" in GIMP to draw a pretty accurate outline, basically very carefully selecting points along the edge in line plot fashion. I don't think ribbons will be too bad as I did apply a slight gaussian blur. I'm going to monitor it though and make sure the drift isn't too bad and make sure the feathering is at a good amount. It really can't get too much worse than it already is and if it does, oh well.

One complication is that the projector was definitely in two setups. There are really two wear patterns, though they are nearly over each other and one is more pronounced. I may wear the virgin phosphor to a level between them.

Mark_A_W · Posted: Wed Oct 07, 2009 4:51 am Post subject:

I played all of Half Life 2 on a PG Xtra at 100% contrast and didn't burn/wear the tubes at all.

So, um...good luck, perhaps it will wear faster than mine did....But if really comes down to how long are you willing to leave a 600w heater on?

Zebu Fellenz · Joined: 21 Dec 2006 Posts: 2567

I did the same thing on my Marquee 8000, it helped but I ran my pattern for somewhat more than 1,000 hours, and it left a slight bright line like Perisoft mentioned as my pattern was slightly off on one side.

On my G70 I also have uneven wear, but I haven't done anything special to deal with it, it does seem to be becoming less visible as the hours rack up, possibly the difference in wear between an 8k hour tube, and a 7600 hour tube is much less than the difference between a 400 hour tube and virgin phosphor?

It does work, but it takes a long, long time.

Good Luck

1031 · Joined: 22 Mar 2006 Posts: 657 Location: Finland

It is wise start that wear leveling at small contrats (10 min) then add more contrast (maybe another 10 min) and little more..Until tube is warm enough, then take full boost out of it. Cold tube can crack if temperature rises too fast.

MYoung · Joined: 24 Feb 2007 Posts: 369 Location: Madison, WI

Hmm, 1,000 hours is around 42 days. Though I'm thinking I'll certainly see a difference at 20 days and maybe even 10 days. Maybe if I redraw the wear pattern every few days that will reduce overlap and underlap. I'll certainly redraw it at around 20 days as that's when the outer phosphor should be around the light output as a slightly lighter area on the edges of the main wear pattern. I should take a decent before and after picture.

ecrabb · Posted: Wed Oct 07, 2009 2:44 pm Post subject:

I'd guess it'll take WAY less than 1,000 hours to start seeing a change... Full white is WAY more beam exposure than the phosphor ever sees with actual viewing material, so I think your 10-day assessment is right on to start seeing a difference. I think I'd be monitoring it daily and maybe moving the raster around a few clicks at the same time.

SC

Kiev Savoie · Joined: 25 Oct 2007 Posts: 432

Very interesting. How many hours are on that tube?

perisoft · Joined: 29 Aug 2007 Posts: 2920 Location: Ithaca, NY

Kiev Savoie · Joined: 25 Oct 2007 Posts: 432

I have a G70 with only 900 hours on it but showing mild wear on the green tube. It took me three weeks to notice it, but now I can't help but see it in every bright scene. Some "Home Theater Technician" preset the contrast at 80 and set it up using only about 60% of the raster. Rolling Eyes

Luckily, it did not stay that way for long, another thousand hours and it would have been toasted.
Can't wait to see how this works for you!

Kiev Savoie · Joined: 25 Oct 2007 Posts: 432

Mark_A_W · Posted: Thu Oct 08, 2009 8:33 am Post subject:

Mark_A_W · Posted: Thu Oct 08, 2009 8:34 am Post subject:

MYoung · Joined: 24 Feb 2007 Posts: 369 Location: Madison, WI

Several years ago I played Grand Theft Auto Vice City and San Andreas on the PC from start to finish on a 1271Q. It must have put some wear on the tubes. I cannot say for sure how much though because, well, I didn't really care. Laughing

Everyone should have a beater CRT projector!

MYoung · Joined: 24 Feb 2007 Posts: 369 Location: Madison, WI

Well, I'm about 120 hours into the wear-leveling process. It is working well. I get the feeling I'm around half done. There are actually 4 wear patterns that I'm trying to level out, one by one. First comes leveling out the outer virgin phosphor to the next, much darker shade of the outer most shade, then work my way in. This projector was setup from scratch at least twice judging from the patterns and in the middle it has a slight overlapping cross wear pattern of 16:9 and 4:3. We'll see how successful I am at getting rid of those. They are noticeable at full white, though knocking out the virgin phosphor will be the key thing and that's working pretty well. That's what was blatently obvious when watching bright scenes in movies, at least to me. I've probably redrawn the wear-leveling pattern 4 times so far in case there's lots of drift. I'm going to make an animation of the drawn wear patterns to see how much drift there actually was. I do notice slight ribboning, as perisoft suggested would happen, but it's very slight and I think you'd have to see peak white projected and carefully analyze the image to find it. We'll see if there's ribboning on the other parts. I think redrawing the wear pattern and applying the slight gaussian blur is helping in large part prevent that in large part.

The problem I see with taking a picture of the wear pattern and making a wear-leveling pattern based on that is all the time to get the pattern to fit correctly, plus the need to recreate it or carefully adjust it to account for drift, whereas just redrawing the wear pattern with GIMP's paths tool while projecting full white takes a few minutes. I think that you could add control points to the image if you take a picture of the wear pattern and use those points to line things up, but having software to properly distort the image to fit might be a challenge.

Pictures (left pic is at 25 hours and right pic is at 120 hours):

Not the best pics as you can see the retrace 2/3 down on the left and a bit more than 1/3 down on the right, but you get the point.

CIR Engineering · Posted: Mon Oct 12, 2009 4:36 pm Post subject:

Looks like you are making good progress Thumbs Up

I don't think the tube has wear from two different setups though, I think it has 16x9 and 4x3 wear in the same raster. Either way, you will probably have to do two different wear patterns to fix it as you have already surmised.

Also, I think you are correct on the amount of time it will take to level. At 100% contrast and 100% brightness with full white you are cooking the phosphor really fast. What took thousands of hours to burn in will probably take less than 300 to level out.

Keep it up, this is a fantastic experiment.

If you want it to go even faster, raise the green gain to max as well. But you may not want to bother as you are already making good progress and raising the gain will make the tube run even hotter.

craigr

CIR Engineering · Posted: Mon Oct 12, 2009 4:37 pm Post subject:

Actually, now that I look harder at your pics I am sure that you are correct in that there were at least two different full installs on that tube...

craigr

MYoung · Joined: 24 Feb 2007 Posts: 369 Location: Madison, WI

Yeah, it's hard to tell from the pic as it's a low rez pic taken with a point and shoot camera without a tripod, but on the upper-right part of the one on the left there's a sliver of wear that's between the darkness of the middle wear and the virgin phosphor. I already wore the outer phosphor past that shade and hence needed to redraw the wear pattern.

I'm approaching attacking the 16:9 and 4:3 patterns.