I'm a local ISF Calibrator in Detroit and recently worked on a Sony KDSR50XBR1 SXRD and a Sony KDS60A2000 SXRD. After completely tuning each television, both televisions behave the same way after calibration. Both of them are shifting red very very much in the lower shades, very close to black (i.e. 3,4,5 IRE). The red is very noticable in dark scenes of movies.

I've tried everything to fix this problem but it means pushing a lot of blue in the low end and it really destroys the grayscale. I also couldn't get a good gamma either, basically a straight line @ 1.7 or so.

Any suggestions? Especially for the red in the lower grayscale.

Thanks guys!

P.S. I'm aware of the recall on the optic blocks for the televisions but both models are fine with new optic blocks. No green blob.

Greetings

You compromise the bottom end as taught in the class.

Or you add an external scaler for $1500 to $5000 to fix the bottom end.

regards

Check out this post and subsequent posts in the thread where I explain how to adjust gamma on these sets. You should not have a red shift near black with these displays. If I were you I'd start over fresh and apply these techniques.
http://www.avsforum.com/avs-vb/showthread.php?p=13854013#post13854013

I actually think this might be an issue with Sony's SXRD design.

When I first got my HW10 projector (not the same display, I know) I was able to get a near-perfect D65 greyscale without much trouble.

Now that I'm 400 hours in, the greyscale response—particularly reds—is a lot less linear and I need to have 0–20% grey or so with +5/10% red to have a flat response everywhere else.

After a lot of tweaking I was able to minimise it so that it's not much of a problem, but it's still nowhere near as good as it was within the first 100 hours.

Initially the Gamma 3 preset measured 2.3 at the lens, and now it measures 2.1. (unfortunately the HW10 doesn't support image director so you can't use a custom curve)

A friend of mine has also had the same thing happen with his VW60, though his is apparently much worse than mine now.

No disrespect, but I disagree. I've done hundreds of the SXRD RPTVs (and I own one) and a red shift near black is a calibration problem. Of course, there's always a chance it's something I haven't seen yet, but with both of them showing it...
It's true, though, that they do drift much more than, say, a DLP as they break in.

Thanks for the helpful responses! I'm not convinced it's a calibration error though. I worked on a KDSR70XBR2 recently and it did not shift red near black the way the XBR1 and A2000 did. Granted the XBR1 and A2000 had a ton of hours on them and the XBR2 probably had only 200 - 300 hours on it.

These two televisions have a lot of extra processing that you have to turn off or take a close look at but I have tried all of them to get the red out of bottom end. If I remember correctly I tuned the bottom end off of a 20 or 30 IRE window and the top with a 70 or 80 window. No matter what I do, if I get 20 IRE dead accurate, it still shifts red really really bad.

I'm convinced there is nothing I can do about the gamma.

No disrespect, but I disagree. I've done hundreds of the SXRD RPTVs (and I own one) and a red shift near black is a calibration problem. Of course, there's always a chance it's something I haven't seen yet, but with both of them showing it...
It's true, though, that they do drift much more than, say, a DLP as they break in.
I just find it strange that I had no problem reaching a flat D65 for the first couple of hundred hours, and now that I've left it alone a bit, after 400–500 hours I can't flatten out the red response.

If I use 20–80, I have a dip in the middle, if I use 40–80 I have it flat from 30 up, but below that there's 5–10% too much red.

Maybe it is just a calibration error, as I'm not too familiar with SXRD, but I find it odd that I had no problem getting great results and suddenly am having problems getting a flat response after 400–500 hours.


After a lot of adjustments, I was able to get an acceptable response, but it took significantly longer than it had done in the past.

It would be great if I am doing something wrong though. :) Of course I'm talking about two SXRD projectors with the problem, not rear-projection sets, so it might be a different issue.

I just find it strange that I had no problem reaching a flat D65 for the first couple of hundred hours, and now that I've left it alone a bit, after 400–500 hours I can't flatten out the red response.

If I use 20–80, I have a dip in the middle, if I use 40–80 I have it flat from 30 up, but below that there's 5–10% too much red.

Maybe it is just a calibration error, as I'm not too familiar with SXRD, but I find it odd that I had no problem getting great results and suddenly am having problems getting a flat response after 400–500 hours.


After a lot of adjustments, I was able to get an acceptable response, but it took significantly longer than it had done in the past.

It would be great if I am doing something wrong though. :) Of course I'm talking about two SXRD projectors with the problem, not rear-projection sets, so it might be a different issue.

Same scenario here with my A3000. Very easy to calibrate and great results for the first year. Colors were near perfect and Gamma and Grayscale tracked linearly and was very easy to tweak. I did not check it again for another 3-4 months but when I did, it was a real mess. I thought initially something was wrong with my light meter but now I am convinced that is not the case. A fellow A3000 owner used my Eye-One to compare side-by-side with his Eye-One and got very similar results. He was not having any problems with his A3000 at the time, at least not the problems I had. Gamma had dropped off from 2.15 to 2.20 to about 1.90 or less and was not nearly as linear. My grayscale was a mess too, particularly on the low end. I too found the problem was mostly with Red and I had to cheat the grayscale adjustments to minimize the Red bias on the low end. And finally color points that were near perfect out-of-box had shifted about 10-15%.

If this is a problem with SXRD technology, what do you guys think the chances of Sony taking care of the problem are? Probably slim, but like I said before, they have had serious defects with these models.

Personally my A3000 with thousands of hours on the bulb measures about the same as it did new, after adjusting white balance and using a higher iris. The brightness of the bulb has dropped and the unadjusted white balance has clearly changed, which can both be recognized without measurements. I can only guess that those items are due to the bulb aging, so generally I don't see anything significant that has clearly changed with time on my SXRD.

I have a feeling it is a defect/problem with the early model SXRDs. Both the XBR1 and A2000 were first models. Like I said before, I didn't see any issues with an XBR2 I worked on. Looks like Sony found those problems and fixed them.

I strongly feel this is a calibration issue. If it were a blue cast (elevated blacks with a blue hue), that would be different; I have seen some of them, and my first set had that.

I have calibrated many (a hundred or more) early model SXRDs, in addition to owning one. I strongly feel this is a calibration issue. If it were a blue cast (elevated blacks with a blue hue), that would be different; I have seen some of them, and my first set had that.

How do you explain my experience then ? I did calibrate my set... grayscale, gamma, contrast, and black level. There was no need to even attempt color controls because color was nearly perfect out-of-the-box. I checked my A3000 frequently over the first year and only ever made very minor tweaks. Everything was holding up nicely but when I checked again after another 4 months, it was a mess, and I had not made any changes at all to the set during these 4 months.

I don't see how this can be a calibration issue. I was using the same source equipment, the same light meter, the same software (HCFR) on the same laptop.

I should clarify based on your comment. When I say it was a mess, Red was now reading way too low particulary on the dark end of the grayscale. I had to add more Red but if I brought Red up to match Blue and Green, I would then have a visible Red bias on the low end. Perhaps this is what you are referrring to as a "blue cast" ?

I've found, as you have, that SXRDs drift a lot. Gamma, grayscale, and even primary points on the A3000 seem to drift more than would be expected from normal lamp aging. This SXRD drift seems to be normal. It's a very good idea to touch up an SXRD calibration once a year, or more often if you DIY. I'm not sure why your set didn't drift much the first year and then drifted a lot. But the issue reported in the original post can still be a calibration related issue (correctable with a certain calibration technique and not a manufacturing defect).
I hope I am not coming across as a Mr. Know-It-All, by the way. It may sound like that, but it's not my intention at all.

I've found, as you have, that SXRDs drift a lot. Gamma, grayscale, and even primary points on the A3000 seem to drift more than would be expected from normal lamp aging. This SXRD drift seems to be normal. It's a very good idea to touch up an SXRD calibration once a year, or more often if you DIY. I'm not sure why your set didn't drift much the first year and then drifted a lot. But the issue reported in the original post can still be a calibration related issue (correctable with a certain calibration technique and not a manufacturing defect).
I hope I am not coming across as a Mr. Know-It-All, by the way. It may sound like that, but it's not my intention at all.

I appreciate your feedback and don't think you are coming across as a know-it-all. I think you are just relating your experience as I am. BTW, I did try a new lamp and the only difference was slightly more light output. My original lamp was at only 1500 hours or so. I tried everything I could think of to rule out the TV, including using both an HD-DVD and Blu-Ray player, but the results were nearly identical.

In the end, my A3000 still looked pretty good after compensating for the bias. But in the end, I got rid of it because it simply bothered me that the set had drifted as much as it did and because it was no longer possible (for me) to get it back to its original state of calibration. Just hope my new Panny plasma is more stable over the long term.

Chad B have you experience the A3000 from the first 500 hours to something
like 3-5K hours? If so how far has the primaries and secondaries shifted?

I did not know the hours, but I did one when it was fairly new and then again a year later, and the primaries had shifted significantly. The first time the normal color space was far more accurate, but the second time it was a toss up between normal or wide as to which one was most accurate. That set also developed a positive blue at the low end of the grayscale, which was easy to calibrate out.
On the other hand, my XBR1 SXRD's primaries don't shifted noticeably, but it's grayscale and gamma do shift enough to be in desperate need of a tune up every year or so. If I neglected it and never touched it up it'd still be much better than if I hadn't calibrated it, though.

I hope I am not coming across as a Mr. Know-It-All, by the way. It may sound like that, but it's not my intention at all.
It doesn't come across that way to me Chad. This is my first SXRD unit, and I wasn't expecting a significant drop in performance within the first 4 months of use.

The HW10 only has a 2-point adjustment though. If I remember correctly, the only gamma options in the service menu were panel uniformity ones rather than ones that affect the whole screen at once. (so it's something like 4000 adjustments that need to be made)

This amount of drifting (will it settle down at all?) combined with the projector shifting focus as it heats up and the poor motion handling is starting to make me regret my purchase.

That and the poor convergence (though Sony's adjustments really help improve the appearance of it) is almost making me wish I had gone for a DLP machine. (though I did find rainbows, dithering and motion handling to be a problem with them…)