<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0"
	xmlns:content="http://purl.org/rss/1.0/modules/content/"
	xmlns:wfw="http://wellformedweb.org/CommentAPI/"
	xmlns:dc="http://purl.org/dc/elements/1.1/"
	xmlns:atom="http://www.w3.org/2005/Atom"
	xmlns:sy="http://purl.org/rss/1.0/modules/syndication/"
	xmlns:slash="http://purl.org/rss/1.0/modules/slash/"
	xmlns:series="http://unfoldingneurons.com/"
	>

<channel>
	<title>Tom Gidden &#187; ibook</title>
	<atom:link href="http://gidden.net/tom/tag/ibook/feed/" rel="self" type="application/rss+xml" />
	<link>http://gidden.net/tom</link>
	<description></description>
	<lastBuildDate>Sun, 01 May 2011 10:35:37 +0000</lastBuildDate>
	<language>en</language>
	<sy:updatePeriod>hourly</sy:updatePeriod>
	<sy:updateFrequency>1</sy:updateFrequency>
	<generator>http://wordpress.org/?v=3.3</generator>
		<item>
		<title>Replacing the backlight on an iBook G4</title>
		<link>http://gidden.net/tom/2007/10/02/ibook-ccfl-replacement/</link>
		<comments>http://gidden.net/tom/2007/10/02/ibook-ccfl-replacement/#comments</comments>
		<pubDate>Tue, 02 Oct 2007 11:12:44 +0000</pubDate>
		<dc:creator>Tom Gidden</dc:creator>
				<category><![CDATA[Techie]]></category>
		<category><![CDATA[apple]]></category>
		<category><![CDATA[backlight]]></category>
		<category><![CDATA[ccfl]]></category>
		<category><![CDATA[fix]]></category>
		<category><![CDATA[hardware]]></category>
		<category><![CDATA[ibook]]></category>
		<category><![CDATA[laptop]]></category>
		<category><![CDATA[lcd]]></category>

		<guid isPermaLink="false">http://gidden.net/tom/2007/10/02/ibook-ccfl-replacement/</guid>
		<description><![CDATA[For a while I've been noticing significant degradation of the backlight on my dilapidated iBook G4.  So, yesterday, my dad and I spent a few hours replacing the CCFL backlight bulb.

I've had the iBook for over three years now, and over the past couple of years it's been in use pretty much all day every day.  It's my only computer now, other than the little headless Linkstation in the cupboard.  It's been my trusty steed, but it just wasn't built for this kind of workload.
I've also taken it apart more than a few times.  The hard drive has been replaced at least four times -- I've lost count -- and it's even had the motherboard and screen transferred into a new case when the old one wore out.  It's also been back to Apple to fix what seemed to be a broken or loose display signal cable.
Anyway, the main symptom has been a gradual dimming and "colour warming" of the screen, along with a patchy backlight at times.  When I say colour warming, I mean that the entire screen has been getting more red and yellow, with a poor quality white.  In addition, the bottom left quarter of the screen seemed to be distinctly darker than the rest, especially when the screen came out of "reduced brightness", ie. the dimming that occurs halfway before "display sleep".  In fact, I couldn't let the screen dim or sleep without it being much darker on wake up for a while.
I was managing to correct the colour warming issue by tweaking the display colour calibration, but it was getting harder and harder to keep a wide enough gamut for graphics work.
So, after some research, I partially disassembled the iBook's screen case to find out the LCD model.
If you've never taken apart an iBook before, let me tell you: it's really not fun.  It's difficult to do without scratching or damaging the case... even with correct use of the "black plastic stick" (informally known as a "spudger") that Apple insist is the correct tool for the job.  iBooks just aren't really user-serviceable.  For the most part, this isn't a problem, because the memory and AirPort card are accessible from under the keyboard.  However, I've usually had to replace the hard drive instead, which requires almost a complete stripdown of the machine.
NOTE: I must make it clear that the information below is offered purely as a suggestion, and should not be relied upon.  I was taking a risk with this disassembly, and it was largely trial-and-error.  If you use any of this information yourself, then it's up to you, but really don't blame me if you mess up.  It's quite possible I did it wrong, and also possible that you have a different model anyway.  I'm mainly writing this up because no-one else has, and I could have done with some hints myself!  Consider yourself warned.  Definitely Your Problem, not mine.
The case is tightly held together with snap tabs (if that's the correct name), and special sticky tape is liberally scattered pretty much everywhere inside.  All of this needs to be carefully peeled up and reused, if you don't have access to replacement tape like Apple.
Opening up the screen case, on the other hand, is far easier:  just unscrew the four visible hex screws and prize the white case off with your fingernails.  Even my weak-ass fingernails are good enough.  Be careful not to trap the signal cables, though.
From here it gets delicate.  I decided to take this process very slowly, and also asked my tech-savvy dad to sit in and second-guess me all the way, while supplying a second pair of hands.
Handy Hint #1:  When I first replaced the HDD in my old iBook, I forgot to protect the screen, and I also forgot to secure the hanging AirPort antenna plug.  As a result, when I'd put the machine back together I noticed a fairly deep scratch in the middle of the screen.  Very upsetting.  So, before going any further, tape some bubble wrap, cardboard or other protection to the screen surface.  Be sensible, though, and don't go using duct tape or something like that on the delicate surface.  Also, put something soft on the work surface: you'll be putting the LCD face down quite a bit.
On opening the screen, you'll see a lot of tape and stuff.  I carefully removed the obvious bits, along with the foil tape covering the display connector on the back of the LCD.  The aim is to totally remove the LCD unit from the case, so figure it out from there.
I then partially removed the LCD shield:  the thin metal case.  There were four small screws on the sides of the shield holding it in.  I removed the bottom screws and lifted the shield just enough to look at the back of the LCD unit itself.  From this I could see the make and model of LCD:  Chi Mei N141XB-L03 Rev. C1.
From what I could find out, this make/model is meant to be quite easy to disassemble... I'd hate to see what a difficult make/model would be like.
I ordered the correct replacement lamp from Sparesweb for £22.79 all inclusive, and it arrived the next day.
The bulb itself is incredibly fragile.  It's very light and made of very thin glass.  I've seen other web pages suggest buying more than one just in case, and I can see the logic.  However, I risked buying just one.
Next thing is to get to the fitted bulb.  This requires quite a lot of fiddly dismantlement, and there doesn't seem to be anything on the web saying exactly how to do this, at least for this make/model.  So, we approached it very slowly, carefully, thinking and discussing every movement, and picking the correct tool to use, and the best orientation for the screen.
The backlight sits in a channel at the bottom of the screen, and miraculously manages to evenly illuminate the entire screen from there.  I still have no idea how it manages to be so even.
The visible back of the unit is predominantly white plastic which I assume is the main diffuser.  When switched on, this plastic glows white:  it's the bit that you can see through the Apple window in the back case.  On this screen, there's an obvious thin green circuit board covering the top quarter of the back, covered by a stuck-down clear plastic sheet.  This sheet wraps around the top edge of the screen and is stuck down.
The bottom 2cm or so of the screen is a metal panel, held in by two crosspoint screws.  I guessed correctly that the backlight is under this panel.  However, the panel wraps around the bottom edge, and the main LCD metal frame then wraps over the top of this panel to hold it in.
After staring at this for quite a while, coupled with some reluctant experimentation, we decided that it wasn't possible to lift this panel without removing the main frame, and even if we could, the frame would still obscure the backlight.  So, I carefully slit the top edge of the plastic cover over the circuit board, and peeled up two copper foil tabs.  After doing this, the only thing (apparently) holding the frame in place is the little tabs.  These were easily opened as normal by levering with a small screwdriver.
This is where the stress level increases.  I became very aware of the delicate nature of the sheets of plastic and glass involved.  With the main structural device out of the picture, everything gets more worrying.
At this point, the two silicone caps holding the backlight ends were accessible.  The backlight has a pink and a black cable dangling a little plug at one end. The black cable is lightly stuck in place along the bottom edge.  It's also clear at this point that the metal panel I mentioned is, in fact, a channel or gutter, with a tall J shaped cross-section.  The bulb sits in the bottom of this channel, which then acts as the reflector.
In hindsight, I think we did the next bit wrong.  It looks like the channel's intended to stay fixed in place, and the bulb slid out from the side by unsoldering the wires in situ.  Unfortunately, we didn't figure that out.  Instead, we ended up carefully flexing it back and working with the bulb directly.
In our defence, without knowing any better, unsoldering and sliding such a delicate part without knowing how everything's set up inside is just too nerve-wracking.
The cables have very little slack:  especially the black cable.  They are soldered very close to the glass, and then run 180&#176; back on themselves along the tube.  The silicone caps are fairly complex in design, and it's vital to pay attention to their orientation.
We unsoldered the old bulb, and managed to remove it without breaking.  As expected, the bulb was slightly discoloured at one end.  Two tiny clear rubbery (silicone?) rings were threaded on the bulb at 1/3 and 2/3rds of the length, presumably to space the bulb from the reflector.  I carefully worked these down the length of the tube.
I then rolled them onto the new tube at roughly the right positions.  We trimmed the new tube's leads down to approximately 2mm, and carefully soldered the black wire on.  We quickly found that we hadn't angled the black wire back far enough, so the cap wouldn't fit.  After resoldering, it seemed to fit better.  Then, I carefully guided the tube into the channel, and then fitted the pink cable the same way.
After mental gymnastics, plus trial-and-error, we managed to fit the caps in the right way and then got the channel back in place on the screen.  It was a tight fit, and took a lot of patience to get it sitting correctly.
From there, it was just reversing the procedure.  Before finally fitting the shield, we switched on the iBook and tested the screen, fully expecting that we'd broken something permanently... it's just too delicate for it not to go wrong.  Surprisingly, all worked properly.
Once reassembled, we relaxed for a while.  In fact, we took a number of short breaks, just to stay focussed, as it's fairly nerve-wracking doing stuff this delicate and critical to your main (and only) machine.
Upon switch on, I was struck at (a) how bright, and (b) how blue the screen was.  I was worried for a while that there was a signal cable problem and the red signal was failing.  However, after booting, it was clear that all was well.
Initially, and for the first few hours, the bulb was definitely bluer than expected, but this calmed down after a while.  There was also an odd effect that the screen got redder as it was tilted away from the viewer.  This also went away.
Quite quickly, I figured out why the screen seemed so blue.  Firstly, it was because I was so used to a worn-out backlight with an orange cast. The real cause, however, was that I was still using the massively overcompensating ColorSync profile!  Once I selected a generic profile, everything looked a lot better.  Even so, I've chosen a lower colour temperature than I would usually, as the bulb is still quite blue.
After writing that last night, I revisited it just now, comparing it against my external LG Flatron L1730B.  Unfortunately, it doesn't compare that well.  The white is still far more cyan than I'd like, and I just can't get ColorSync to give me a nice, clean white.  It's still better than it was, though.
I seem to remember reading that these CCFL bulbs can sometimes take a few weeks or months to properly settle down, so we'll see.
Conclusion
I must say, replacing this backlight was a real bitch, mainly due to lack of instructions, coupled with the delicacy of the task at hand.  In hindsight, I'd probably just get an entire new LCD panel next time.  It's nice to know that I can replace the bulb myself if necessary, though.  Much more complicated than it should be, but not totally impossible.  Thing is, this is officially not a user-serviceable part, but it wears out over time, and is also likely to wear out well after AppleCare expires.  A bit of a crappy deal if you ask me.
If all goes well, this iBook's going to be retired soon, but I don't want to get another laptop with a CCFL backlight now that LED-based ones are available.  However, the 17" MacBook Pro I want still uses CCFL, so I'm stuck with this iBook for the time being.
Anyway, if you're intending to undergo this task, I hope you find these notes helpful.  Unfortunately, I didn't take any photos while disassembling it which I'm sure would be incredibly useful. However, I had far more critical things to worry about at the time!  Instead, I took these photos after reassembly, so they're not that helpful.  Even so, I've put some notes on the photos on the Flickr pages which could be handy.]]></description>
		<wfw:commentRss>http://gidden.net/tom/2007/10/02/ibook-ccfl-replacement/feed/</wfw:commentRss>
		<slash:comments>1</slash:comments>
		</item>
		<item>
		<title>Hitachi GST (no longer) Sucks</title>
		<link>http://gidden.net/tom/2006/09/11/hitachi-gst-sucks-ass/</link>
		<comments>http://gidden.net/tom/2006/09/11/hitachi-gst-sucks-ass/#comments</comments>
		<pubDate>Mon, 11 Sep 2006 16:10:48 +0000</pubDate>
		<dc:creator>Tom Gidden</dc:creator>
				<category><![CDATA[General]]></category>
		<category><![CDATA[Techie]]></category>
		<category><![CDATA[4k120]]></category>
		<category><![CDATA[customer-support]]></category>
		<category><![CDATA[failure]]></category>
		<category><![CDATA[hard-drive]]></category>
		<category><![CDATA[hitachi]]></category>
		<category><![CDATA[hitachigst]]></category>
		<category><![CDATA[ibook]]></category>
		<category><![CDATA[laptop-drive]]></category>
		<category><![CDATA[rma]]></category>
		<category><![CDATA[travelstar]]></category>
		<category><![CDATA[warranty]]></category>

		<guid isPermaLink="false">http://gidden.net/tom/2006/09/11/hitachi-gst-sucks-ass/</guid>
		<description><![CDATA[UPDATE January 2007:  I got the replacement drive, and it has been working fine for a few months now.  Hitachi's service was quick and painless, and while I'm still not 100% sure about the quality of Travelstar drives (since I've had them fail since the IBM days), I'm satisfied that the problem's fixed.

Unfortunately, while reassembling the iBook I managed to gouge a scratch in the LCD screen, courtesy of the sharp edges of the Airport antenna cable.  Fortunately, my mother said she doesn't notice the scratch, and so was okay with swapping iBooks with me.  As a result, the iBooks have been taken apart and put back together many times, and it's difficult and fiddly each time.
UPDATE 13/9/06:  I finally got a good reply from Hitachi on this issue, and it looks like running the DFT tool isn't actually required to get an RMA.  So, I'm going to open the iBook up and get the drive sent off in the next day or so.  This is a major PITA, as I won't have my music/photos accessible for the time being.  I'm still slightly honked off with the sloppy customer service.  Anyway...
The Hitachi Travelstar 4K120 drive I fitted in my iBook earlier this year has started making horrible clunking sounds.  The clunks seem to coincide with the drive seeking for data after a short idle time.
Anyway, since this is my main (and only) machine, I'm understandably a little worried, so I backed up the drive to my external Firewire drive, and sent off a note to Hitachi's warranty returns contact page.
The note explained the problem, and went on to explain that I have an Apple Macintosh laptop (iBook G4) which will not run the "Drive Fitness Test" tools required to diagnose the problem.  It also explained that I don't have a PC at hand to test the drive; I don't have the required 2.5" IDE cable adaptor to plug it into such a PC if I did have one; and that the iBook takes a long time to take apart and put back together, which I don't really want to do if they're not going to authorise an RMA.
This morning I got a reply:

From:     HGST.NoReply%SUPPORT [AT] support [DOT] hgst [DOT] com
Subject:  Hitachi GST Support Request #XXX-XXX-XXX-X
Date:     11 September 2006 11:17:00 BDT
To:       tom [AT] gidden [DOT] net
Reply-To: support_uk [AT] hitachigst [DOT] com

Dear Mr Gidden,

Thank you for contacting our Hitachi Technical Support Center.

From the information we have regarding your problem we cannot make
a reliable diagnosis of your problem and how to resolve it.

In order to gain more information about your problem you need to run
a utility called Drive Fitness Test on the drive. This will establish
the cause of your problem.

Please download Drive Fitness Test from the following URL:

http://www.hitachigst.com/hdd/support/download.htm

If possible we would recommend backing up any important data at this
time before running the test.

Run the quick test first, and if you receive the result 0x00, then run
the advanced test.

If you receive result code 0x72, 0x73 or 0x75 please log on to
http://www.hitachigst.com/warranty in order to create an RMA Number
for your hard drive.

For any other result code please contact us again

Best regards
XXXXXX XXXXXXX
Hitachi Technical Support Center

From this, I can tell that they didn't actually read my original note.

From:    Tom Gidden
Subject: Re: Hitachi GST Support Request #XXX-XXX-XXX-X
Date:    11 September 2006 11:31:11 BDT
To:      support_uk [AT] hitachigst [DOT] com

Hi,

As I said in my original message, I can't run Drive Fitness Test
as the drive is installed in my Apple Macintosh laptop, and DFT is
only available for PCs.

I don't have the equipment necessary to connect the drive to a PC, as
it's a 2.5" drive.

Regards,

--
Tom Gidden

And the reply...

From:     HGST.NoReply%SUPPORT [AT] support [DOT] hgst [DOT] com
Subject:  Hitachi GST Support Request #XXX-XXX-XXX-X
Date:     11 September 2006 13:01:24 BDT
To:       tom [AT] gidden [DOT] net
Reply-To: support_uk [AT] hitachigst [DOT] com

Dear Mr Gidden,

Thank you for contacting our Hitachi Technical Support Center.

Simply connect to the internet and point your browser to

http://www.hitachigst.com/warranty/

After you have run DFT and checked the drive Click the link called
RMA Request Process. You can then work through the wizard on screen to
authorise your drive for warranty.

If when using this system you are unable to return your drive you will
be given a telephone number to call for further support and advice.

Once an RMA is created you can check the status of your request by using
the RMA status tool at this link

http://www.hitachigst.com/warranty/jsp/arma51.jsp

Best regards
XXXX X
Hitachi Technical Support Center

I can see that I'm not actually getting anywhere.  Yet another case of sucky customer service where the customer isn't getting listened to.  I also object to the fact that I need a PC to process a simple warranty claim.  Macs are common enough nowadays that it's a bit unfair when drive manufacturers only have DOS-based x86 diagnostic tools.
At this point, the drive is starting to clunk a whole lot more, so I'm going to have to bite the bullet and take it apart anyway, and try to get it hooked up to my dad's PC somehow.
This is the second drive I've bought this year, and the second one that's started making clunking sounds, and neither of them have actually got moved about or anything.]]></description>
		<wfw:commentRss>http://gidden.net/tom/2006/09/11/hitachi-gst-sucks-ass/feed/</wfw:commentRss>
		<slash:comments>2</slash:comments>
		</item>
	</channel>
</rss>

