X220 IPS Image Retention Problem - Replace LCD?

[dr_st]

My recently purchased X220 (with IPS) displays a pretty obvious case of the image persistence/retention issue (sometimes referred to as "ghosting" or "burn") that has been known to plague these panels.

See for example the following thread and the links contained therein:
http://forum.thinkpads.com/viewtopic.php?f=43&t=97785

In short: After a static image is displayed on screen for a number of minutes, after switching to a different image, the contours of the previous image remain visible, and slowly dissipate over time (also minutes). This applies also to partial images, i.e., if only a part of the image was static, that part will "burn in", while the rest won't.

At some point, after many frustrating months following the release of the machine, Mark at the Lenovo forums released a statement indicating that a newer batch of panels should have the issue greatly reduced and that Lenovo will exchange panels that exhibit noticeable persistence with new ones, under warranty. Here is the original post by Mark:
http://forums.lenovo.com/t5/X-Series-Th ... 363#M30686

There has indeed been some evidence that the later batches of the IPS panels exhibit lesser ghosting, although they are not 100% free of it. For instance, this thread:
http://forum.notebookreview.com/lenovo/ ... els-3.html

But I was hoping for something more conclusive. With many X220 IPS owners here, I hope that someone can share their personal experiences:

• Have you noticed said persistence phenomenon on later revisions of the panel?
• Have you been able to compare earlier and later revisions to say that there the effect is indeed more pronounced in the earlier ones?

Now, here's the thing. My X220 was manufactured in 11-2011. According to the information we've received from Lenovo, all units built after 08-2011 should actually come with the revised, improved panel. See here:
http://forums.lenovo.com/t5/X-Series-Th ... ue#U543137

I guess, it might be possible that my panel was manufactured earlier. Is there a way to find out without opening up the bezel?

Without an earlier unit to compare to, I do not know if my level of persistence is to be considered normal or abnormal. I can try my best to describe what I see.

• Most comments quoted that it takes 10-15 minutes for the "burn" to appear. It is actually far far less than that. Already after 2 minutes of a static image, the retention is visible when changing images. It is of course stronger, if the image was static for a longer period. It takes it about the same time to gradually dissipate as it took to "create" it.
• I've noticed that there is a very strong correlation between the luminance of the color displayed on the new image and how visible the old image is. Notice that I'm not talking about screen brightness, but the color luminance/lightness (on the HSL scale). The retention is very visible on darker shades (seems to peak around luminance level 30-50 out of 240), and becomes less visible as the luminance increases. At luminance levels over 150 it is invisible are almost invisible. In particular I never see it when displaying the pure colors (255 in one or more of the color channels). This is how I first noticed it, because by chance, my Win8 start screen background was set to a fairly dark purple. I then investigated with Dead Pixel Buddy (which is capable of displaying a solid screen of any color), and concluded that the color itself has almost no effect on visibility of the retention. Only the luminance.

So there you have it. Now my choice is whether to push for a replacement of the LCD under warranty or not. So far I have not experience the issue as an annoyance under normal usage (I guess displaying pictures consisting of large chunks of solid low-lum colors is not something I typically do), but then again I have not been using the laptop much (yet).

Thank you for your time reading it.

[RealBlackStuff]

manufactured in 11-2014

[dr_st]

Yeah, I wish. Obviously I meant 2011. Fixed, thanks.

[Puppy]

I have one of last X220 boxes (manufactured 03-2012) so I assume it has the second revision. The issue is still there but a static image (Windows taskbar, main window caption) must be displayed for at least four hours to make its persistence noticeable on grey or light-blue background. I use sceeensaver that displays a blank image after 3 minutes of inactivity and brightness level 6. I agree it is more (earlier) visible with higher brightness but I am a night bird

Still the e-IPS X220 panel is the worst among other IPS displays I have. The one in R51 does not exhibit the image retention issue at all. The A-TW-IPS one in my desktop NEC 2090UXi monitor suffers from that as well in a small area in top-left corner (there is power supply in that location so it might be caused by permanent heat, 7 years - 15 hours daily). Both affected panels are made in LG.Display (the NEC one is designed by NEC but the mass production actually takes place in LG). The R51 one was made by IDTech.

As for the replacement. The panel is very sensitive on pressure/twist applied to the lid that results in excessive backlight bleeding. I have seen badly accomodated panels after repair having excessive backlight bleeding because of the constant pressure made to the panel. I have also experienced something that looks like a stuck red subpixel on dark background visible on maximum brightness level (everytime after power-on due the BIOS bug that always set the maximum brightness level until Windows OS is loaded).

[dr_st]

I have one of last X220 boxes (manufactured 03-2012) so I assume it has the second revision. The issue is still there but a static image (Windows taskbar, main window caption) must be displayed for at least four hours to make its persistence noticeable on grey or light-blue background.

Well... That in itself does not do much to contradict my findings. On light backgrounds (grey, blue, whatever) there is no visible retention on my panel as well.

It would be more interesting to see if your panel exhibits, similar to mine, visible retention on darker backgrounds. The difference on mine is really tremendous. Dark colors show retention after 1-2 minutes. Light colors don't show it, basically, ever (maybe after a few hours they would, but I never have static images displayed for so long).

Then again, maybe you don't want to find out. If your panel turns out to be just like mine, and you end up unhappy because of that, I don't want to be held responsible.

Still the e-IPS X220 panel is the worst among other IPS displays I have. The one in R51 does not exhibit the image retention issue at all. The A-TW-IPS one in my desktop NEC 2090UXi monitor suffers from that as well in a small area in top-left corner (there is power supply in that location so it might be caused by permanent heat, 7 years - 15 hours daily). Both affected panels are made in LG.Display (the NEC one is designed by NEC but the mass production actually takes place in LG). The R51 one was made by IDTech.

That's a no-brainer. 99% of the IPS panels out there are manufactured by LG-Philips. Some of them are great, others are bad. Quality control issues are all over the cheap ones. Just look at all the issues that people have been reporting with the DELL IPS monitors. That's what you get from LG when you want to buy panels from them cheap. $50 is how much extra Lenovo charged for the IPS panel on the X220. You can imagine that the difference for them was even less. And you can then imagine what kind of crap they must have been buying for this money.

As for the replacement. The panel is very sensitive on pressure/twist applied to the lid that results in excessive backlight bleeding. I have seen badly accomodated panels after repair having excessive backlight bleeding because of the constant pressure made to the panel. I have also experienced something that looks like a stuck red subpixel on dark background visible on maximum brightness level (everytime after power-on due the BIOS bug that always set the maximum brightness level until Windows OS is loaded).

Those are valid concerns. Let's just say that I will think hard if I want to bother with a replacement, especially since that one may not be any better. I'll have to start using the laptop regularly and then see if the retention really bothers me or not.

But somehow I think I'll have to go with the tradition of replacing at least one LCD per privately owned Thinkpad in warranty, and eventually replace that darn thing.

[cros]

Coincidence - I was about to post a message about this. I got a new X230 recently. I couldn't find an unused model with IPS, so bought a TN one and a fresh 04W3462 (genuine boxed Lenovo part, supposedly free of persistence issues) to put into it.

Swapping panels was a cinch: 2 screws on the bezel, 4 on the panel tabs. The most difficult part is removing the bezel. I found that starting on the outer edge of the lower centre worked best. The claws on the outer edge are shallower and things unzip nicely from there.

First thing: the IPS is beautiful compared to the TN panel! There is no meaningful comparison: proper blacks, relatively wide angles, no blue cast, more subtle colours etc. IPS lifts the whole machine with a premium feeling.

I saw image persistence right after the first power up. Windows popped up an "Internet Connection" tooltip for just a few seconds and its black outline was "burnt" into my grey-ish wallpaper afterwards. Moving a window through the area cleared the issue instantly. Strangely and luckily, the panel hasn't done this since. Rising temperature? AC bias re-centering? I'm happy with it now anyway...

You might not have seen this service bulletin: http://download.lenovo.com/express/HT051485.html It serves to classify ghosting as normal - not much help - and adds nothing to what has already been said on the forums.

Significant backlight bleed was very noticeable on two corners, at least in tough conditions: black background, dim environment, high backlight settings. Luckily this was simply due to pressure being applied to the panel by the bezel. This LCD is very sensitive to pressure and is simply taped at the edges rather than having a frontal metal frame/surround like other screens (the pursuit of thinness?)

Adding a few plastic shims between the bezel and lid back was enough to remove the bleed spots. Two bezel claws were simply tugging too tightly into the panel.

I'm really happy with it now, but can see how any flex/pressure carried by the LCD will cause bleed. I may have to revisit my installation as things settle...

Overall, the X230 seems as well built as other Thinkpads I've spent time with (570, T23, T42, X61, X200). The battery, keyboard mounting and LCD bezel needed slight TLC to reach perfection, but the others weren't 100% from the factory either. The island-style keyboard feels OK and looks modern, but I don't see why the 7th row had to be removed. (PrintScreen in place of the second Windows key - why?). The signed BIOS update mechanism is a pain too, meaning the WLAN whitelist is unlikely to be bypassed. 802.11ac or Broadcom WLAN (OS X) will probably never be an internal option for this machine.

[ZaZ]

$50 is how much extra Lenovo charged for the IPS panel on the X220. You can imagine that the difference for them was even less. And you can then imagine what kind of crap they must have been buying for this money.

Well, had Lenovo charged $250 for a better panel a good chunk of people, if not the majority, would have complained it's too expensive. From Lenovo's perspective they can't win. I'm sorry you got a bad panel, but I'm not necessarily sure that it makes for a trend. My X220i, ordered the first day, never had any ghosting issues. I hope the new one, here tomorrow, is the same.

[dr_st]

Well, had Lenovo charged $250 for a better panel a good chunk of people, if not the majority, would have complained it's too expensive. From Lenovo's perspective they can't win.

They can't win, and neither can we. The race to the rock-bottom prices is a given, and something has to suffer. Quality is among them. I don't want to speculate about majority, but I'm sure more than a few people would be happy to pay $250 more for quality (the same people that didn't have a problem paying $3000 for their Thinkpads back in the days). It's just not an option given to them among any of the major brands any more.

I'm sorry you got a bad panel, but I'm not necessarily sure that it makes for a trend. My X220i, ordered the first day, never had any ghosting issues. I hope the new one, here tomorrow, is the same.

There is no need to feel sorry. But there is no way to avoid the fact that it is a trend. The hundreds of people that complained all over the internet from the beginning and until now, the fact that Lenovo bothered to issue a service bulletin stating "this is a normal characteristic of the crappy panel we use" make pretty clear that it is a trend.

What I hope to find out is whether the situation is the same for all panels or just some. Or more precisely, whether the degree of the retention varies sufficiently between panels to qualify some of them as issue-free (because on a purely technical "yes/no" level I am sure all of them suffer from the issue, since it's probably inherent to their manufacturing process).

Unfortunately such information is difficult to retrieve without the same person comparing panels side-by-side in the same conditions. Because all of us have different eyes and tend to pay varying degrees of attention to certain things, and because the issue is overwhelmingly more noticeable on certain backgrounds than others.

There was one person in the huge thread on the Lenovo forums that for many pages was repeating that his panel does not show any retention (he was pretty much the only respondent there claiming so), until one day he found out that it, in fact, does, under certain conditions.

I haven't spotted it myself on day 1. I would probably not have noticed it at all, if I had not, by chance, set the Win8 start screen background color to a rather dark one. I can believe that quite a few people who don't use solid dark background for anything they normally do, may never see the issue, even if it's there.

Now, having investigated it, I am fairly confident I can easily reproduce the conditions on any panel, and very quickly determine if it exhibits the retention to a similar degree of my panel or not. However, not having access to any other panels to sample, I have to rely on others making the investigation for me.

So, thanks for your input. Hopefully with enough inputs, there will be enough basis to form a decision.

Swapping panels was a cinch: 2 screws on the bezel, 4 on the panel tabs. The most difficult part is removing the bezel. I found that starting on the outer edge of the lower centre worked best. The claws on the outer edge are shallower and things unzip nicely from there.

That's good to know. Lenovo's manual does not provide instructions for swapping just the panel, only the panel with the LCD cable (which of course requires basically the whole machine to be gutted open).

I assume when you did the swap, you used the original cable that came with the machine and attached it to the new panel. What kind of connector is it? Is it solid enough to detach/attach without fear of damaging it or of it coming loose later?

[cros]

The X230 HMM includes the steps for changing the panel alone. There are also a few videos of it being done on YouTube.

You use the same cable whether TN or IPS. The connector is about 1.5 inches wide and easy enough to release with a fingernail or screw driver once the securing tape is peeled back. It's reasonably delicate looking - don't pull by the wires!

It's unlikely to come loose later: the friction fit is good, the tape is there too and that area of the cable doesn't move or strain when the lid is opened and closed.

[dr_st]

The X230 HMM includes the steps for changing the panel alone.

Hmm... Good point. I assume the panels and connectors are the same for X220 and X230? If so, it just means they haven't figured out this shortcut when writing the X220 HMM.