So how can I tell if it is the display, the backlight, or the
inverter?

And could someone give me an idea what's involved in replacing each?

On Thu, 5 Jul 2012 22:35:42 -0700 (PDT), spamtrap1888
wrote:

So how can I tell if it is the display, the backlight, or the
inverter?
And could someone give me an idea what's involved in replacing each?

Clean the screen. Fire up the laptop and get it to boot. Darken the
room. Use a flashlight at an angle to see if you can detect any text
on the screen. For Windoze, you should see the Windoze logo as it
boots. If you do see an image, your LCD panel is probably fine, and
you are having a problem with the LCD inverter or the CCFL lamps in
the panel.

It's also a good idea to plug an external monitor into the VGA port to
see if the video card is working.

Determining if it's the inverter or the CCFL lamp is a bit tricky. If
you get a pink colored dim background, it's probably the CCFL lamps.
These are a major project to replace. You'll find instructions on web
sites selling LCD panels and CCFL lamps.

The way to be sure is to open the display section and attach a test
CCFL lamp to the inverter.
http://www.insidemylaptop.com/remove-replace-lcd-screen-ibm-thinkpad-t43p-laptop/
The LCD inverter is in the display section of the laptop. You only
need to go to Step #9. The hard part is splitting the LCD frame.
Don't use a metal screwdriver to pry open the display.

www.ebay.com/sch/i.html?_sacat=0&_nkw=thinkpad+t43+lcd+inverter
I usually buy two to avoid surprises. Note that there are different
part numbers for different display types. You'll probably have to get
the number off your inverter to be sure which is the correct number.

Bug me if stuck.
--
Jeff Liebermann
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558

On 7/5/2012 10:54 PM, Jeff Liebermann wrote:
On Thu, 5 Jul 2012 22:35:42 -0700 (PDT), spamtrap1888
wrote:

So how can I tell if it is the display, the backlight, or the
inverter?
And could someone give me an idea what's involved in replacing each?

Clean the screen. Fire up the laptop and get it to boot. Darken the
room. Use a flashlight at an angle to see if you can detect any text
on the screen. For Windoze, you should see the Windoze logo as it
boots. If you do see an image, your LCD panel is probably fine, and
you are having a problem with the LCD inverter or the CCFL lamps in
the panel.

It's also a good idea to plug an external monitor into the VGA port to
see if the video card is working.

Determining if it's the inverter or the CCFL lamp is a bit tricky. If
you get a pink colored dim background, it's probably the CCFL lamps.
These are a major project to replace. You'll find instructions on web
sites selling LCD panels and CCFL lamps.

The way to be sure is to open the display section and attach a test
CCFL lamp to the inverter.
http://www.insidemylaptop.com/remove-replace-lcd-screen-ibm-thinkpad-t43p-laptop/
The LCD inverter is in the display section of the laptop. You only
need to go to Step #9. The hard part is splitting the LCD frame.
Don't use a metal screwdriver to pry open the display.

www.ebay.com/sch/i.html?_sacat=0&_nkw=thinkpad+t43+lcd+inverter
I usually buy two to avoid surprises. Note that there are different
part numbers for different display types. You'll probably have to get
the number off your inverter to be sure which is the correct number.

Bug me if stuck.

What do you mean by dimming? Dim? Dark? off?

Another way to check the inverter is to waive an oscilloscope
probe around the area of the inverter. If it's running, you
can often see the radiation from it on the scope. That won't prove
the inverter board is good, but might lead you to check the lamp
before buying an inverter.
Don't go probing the inverter circuit unless you know what you're
doing. There's voltages there that the scope won't like at all.
You might find yourself with a dead laptop and a dead scope.

Most people don't have 'em, but an ultrasonic leak detector translates
ultrasonics down to the audible range. You can easily hear the inverter
transformer "singing" and hear it change as you adjust the brightness
setting.

I'm interested in hearing how to check CCFL lamps.
In the old days, you could plug on just about any inverter and see if
it lights. Today, the proliferation of connectors and inverter
topologies makes this harder.

I bought a fluorescent lamp checker at a garage sale.
It's basically a transmitter. You waive the antenna around
the lamp and see if it lights. Works great on household fluorescents.
I've been afraid to try it on a laptop for fear that it'd blow up
the display drivers.

I don't have that model, but I did have a thinkpad with a bad fuse.
Was located near the display connector and well disguised. Didn't look
like a SMT fuse at all. Was open and extinguished the backlight.

On Jul 5, 10:54*pm, Jeff Liebermann wrote:
On Thu, 5 Jul 2012 22:35:42 -0700 (PDT), spamtrap1888

wrote:
So how can I tell if it is the display, the backlight, or the
inverter?
And could someone give me an idea what's involved in replacing each?

Clean the screen. *Fire up the laptop and get it to boot. *Darken the
room. *Use a flashlight at an angle to see if you can detect any text
on the screen. *For Windoze, you should see the Windoze logo as it
boots. *If you do see an image, your LCD panel is probably fine, and
you are having a problem with the LCD inverter or the CCFL lamps in
the panel.

It's also a good idea to plug an external monitor into the VGA port to
see if the video card is working.

Determining if it's the inverter or the CCFL lamp is a bit tricky. *If
you get a pink colored dim background, it's probably the CCFL lamps.
These are a major project to replace. *You'll find instructions on web
sites selling LCD panels and CCFL lamps.

I suspect it is my CCFL lamp, because I still have an image, and it
has a rosy glow. I haven't found a good page explaining how to
replace the CCFL, though. Are they fairly generic, or do I need to
make sure I have an OEM replacement? And can you suggest a good set of
replacement instructions?

The inside-my-laptop site seems to recommend replacing the whole
display even if just the CCFL is faulty. Is replacing the CCFL that
much of a nightmare? My display is actually pretty good, so if I could
just replace the CCFL, I would, rather than take a chance of getting a
crap LCD.


The way to be sure is to open the display section and attach a test
CCFL lamp to the inverter.
http://www.insidemylaptop.com/remove-replace-lcd-screen-ibm-thinkpad-...
The LCD inverter is in the display section of the laptop. *You only
need to go to Step #9. *The hard part is splitting the LCD frame.
Don't use a metal screwdriver to pry open the display.

www.ebay.com/sch/i.html?_sacat=0&_nkw=thinkpad+t43+lcd+inverter
I usually buy two to avoid surprises. *Note that there are different
part numbers for different display types. *You'll probably have to get
the number off your inverter to be sure which is the correct number.


Thanks so much for your help.

On Jul 6, 12:02*am, mike wrote:
On 7/5/2012 10:54 PM, Jeff Liebermann wrote:









On Thu, 5 Jul 2012 22:35:42 -0700 (PDT), spamtrap1888
*wrote:

So how can I tell if it is the display, the backlight, or the
inverter?
And could someone give me an idea what's involved in replacing each?

Clean the screen. *Fire up the laptop and get it to boot. *Darken the
room. *Use a flashlight at an angle to see if you can detect any text
on the screen. *For Windoze, you should see the Windoze logo as it
boots. *If you do see an image, your LCD panel is probably fine, and
you are having a problem with the LCD inverter or the CCFL lamps in
the panel.

It's also a good idea to plug an external monitor into the VGA port to
see if the video card is working.

Determining if it's the inverter or the CCFL lamp is a bit tricky. *If
you get a pink colored dim background, it's probably the CCFL lamps.
These are a major project to replace. *You'll find instructions on web
sites selling LCD panels and CCFL lamps.

The way to be sure is to open the display section and attach a test
CCFL lamp to the inverter.
http://www.insidemylaptop.com/remove-replace-lcd-screen-ibm-thinkpad-....
The LCD inverter is in the display section of the laptop. *You only
need to go to Step #9. *The hard part is splitting the LCD frame.
Don't use a metal screwdriver to pry open the display.

www.ebay.com/sch/i.html?_sacat=0&_nkw=thinkpad+t43+lcd+inverter
I usually buy two to avoid surprises. *Note that there are different
part numbers for different display types. *You'll probably have to get
the number off your inverter to be sure which is the correct number.

Bug me if stuck.

What do you mean by dimming? Dim? Dark? off?

Dim. Going but not gone.


Another way to check the inverter is to waive an oscilloscope
probe around the area of the inverter. *If it's running, you
can often see the radiation from it on the scope.


Most people don't have 'em, but an ultrasonic leak detector translates
ultrasonics down to the audible range. *You can easily hear the inverter
transformer "singing" and hear it change as you adjust the brightness
setting.

I'm interested in hearing how to check CCFL lamps.
In the old days, you could plug on just about any inverter and see if
it lights. *Today, the proliferation of connectors and inverter
topologies makes this harder.

I bought a fluorescent lamp checker at a garage sale.
It's basically a transmitter. *You waive the antenna around
the lamp and see if it lights.

I don't have that model, but I did have a thinkpad with a bad fuse.
Was located near the display connector and well disguised. *Didn't look
like a SMT fuse at all. *Was open and extinguished the backlight.

Backlight is still working somewhat. A lot of good tips, thanks.

spamtrap1888 wrote in message
...
On Jul 5, 10:54 pm, Jeff Liebermann wrote:
On Thu, 5 Jul 2012 22:35:42 -0700 (PDT), spamtrap1888

wrote:
So how can I tell if it is the display, the backlight, or the
inverter?
And could someone give me an idea what's involved in replacing each?

Clean the screen. Fire up the laptop and get it to boot. Darken the
room. Use a flashlight at an angle to see if you can detect any text
on the screen. For Windoze, you should see the Windoze logo as it
boots. If you do see an image, your LCD panel is probably fine, and
you are having a problem with the LCD inverter or the CCFL lamps in
the panel.

It's also a good idea to plug an external monitor into the VGA port to
see if the video card is working.

Determining if it's the inverter or the CCFL lamp is a bit tricky. If
you get a pink colored dim background, it's probably the CCFL lamps.
These are a major project to replace. You'll find instructions on web
sites selling LCD panels and CCFL lamps.

I suspect it is my CCFL lamp, because I still have an image, and it
has a rosy glow. I haven't found a good page explaining how to
replace the CCFL, though. Are they fairly generic, or do I need to
make sure I have an OEM replacement? And can you suggest a good set of
replacement instructions?

The inside-my-laptop site seems to recommend replacing the whole
display even if just the CCFL is faulty. Is replacing the CCFL that
much of a nightmare? My display is actually pretty good, so if I could
just replace the CCFL, I would, rather than take a chance of getting a
crap LCD.


The way to be sure is to open the display section and attach a test
CCFL lamp to the inverter.
http://www.insidemylaptop.com/remove-replace-lcd-screen-ibm-thinkpad-...
The LCD inverter is in the display section of the laptop. You only
need to go to Step #9. The hard part is splitting the LCD frame.
Don't use a metal screwdriver to pry open the display.

www.ebay.com/sch/i.html?_sacat=0&_nkw=thinkpad+t43+lcd+inverter
I usually buy two to avoid surprises. Note that there are different
part numbers for different display types. You'll probably have to get
the number off your inverter to be sure which is the correct number.


Thanks so much for your help.

+++++

Felt-tip pen marking junctions of materials and photographs while taking
apart is advisable. Don't pull on anything that looks as though it should be
free, there maybe something at the other end unseen , that is locking it in
place.
Without all the framework stuff, the tubes are about as fragile as spaghetti
and not much thicker.

On Fri, 06 Jul 2012 00:02:54 -0700, mike wrote:

Here's another article on the problem:
http://reviews.ebay.com/Diagnosing-laptop-backlight-problems?ugid=10000000005308726

What do you mean by dimming? Dim? Dark? off?

When the CCFL bulbs crap out, the display initially becomes pink
colored. As more air leaks in, the CCFL lamp becomes dimmer and
dimmer until it finally quits altogether. Some inverters have
protection circuits to detect CCFL faults and will turn off the
display after a few seconds. I forgot what the thinkpad inverters do.
Note that this is for CCFL backlighting, not LED backlighting.

Another way to check the inverter is to waive an oscilloscope
probe around the area of the inverter. If it's running, you
can often see the radiation from it on the scope. That won't prove
the inverter board is good, but might lead you to check the lamp
before buying an inverter.

Good idea.

Don't go probing the inverter circuit unless you know what you're
doing. There's voltages there that the scope won't like at all.
You might find yourself with a dead laptop and a dead scope.

Bah. I use two 10:1 probes setup for differential input. My probes
will handle 1000VDC so there's not going to be a problem with the
typical 500-700V peak output from the inverter. However, if you
forget and set the probe to 1:1, there's going to be an expensive
problem in the scope.

Most people don't have 'em, but an ultrasonic leak detector translates
ultrasonics down to the audible range. You can easily hear the inverter
transformer "singing" and hear it change as you adjust the brightness
setting.

Great idea. I have an HP 4917A and never thought to use it to check
the inverter. Thanks.

I'm interested in hearing how to check CCFL lamps.
In the old days, you could plug on just about any inverter and see if
it lights. Today, the proliferation of connectors and inverter
topologies makes this harder.

I have an inverter used by gamers to light up the inside of their
tower cases glued to a board. I've ripped off connectors from dead
LCD inverters and built a connector octopus. I've only found 4
different HV connectors so it's not an ugly mess yet.

I bought a fluorescent lamp checker at a garage sale.
It's basically a transmitter. You waive the antenna around
the lamp and see if it lights. Works great on household fluorescents.
I've been afraid to try it on a laptop for fear that it'd blow up
the display drivers.

Don't do it. Chances are good that you're going to apply HV to some
place that won't appreciate the voltage. Incidentally, I have a
customer that keeps blowing up monitors, mice and keyboards. It's
obvious that it's his "negative ion generator" that's causing the
problem, but he insists on using it near the computah.

I don't have that model, but I did have a thinkpad with a bad fuse.
Was located near the display connector and well disguised. Didn't look
like a SMT fuse at all. Was open and extinguished the backlight.

I saw that on a Thinkpad T22 or T23. If you short something in the
external LCD connector, it blows the fuse, which also kills the LCD
display. The clue is that BOTH the internal LCD and external video
are dead. I found a rather vague article on the problem:
http://www.tomshardware.com/forum/65493-35-strange-thinkpad-video-problem
Finding the fuse was easy. There was an F(something) reference
designator silkscreened nearby.

Incidentally, I have several USB microscopes that are quite handy for
looking at SMT components that my rapidly fading eyesight can't see.

--
Jeff Liebermann
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558

On Fri, 6 Jul 2012 08:02:07 -0700 (PDT), spamtrap1888
wrote:

I suspect it is my CCFL lamp, because I still have an image, and it
has a rosy glow.

Don't assume it's the CCFL quite yet. It's best to verify that the
inverter is doing its job before attacking the LCD panel. That means
hanging another CCFL lamp across the inverter output to see if it
lights up, and applying voltage from an external LCD inverter to see
if the CCFL lamp is any good. I would not try to replace the CCFL
lamp until you've clearly identified it as the problem.

Also, as a contingency, I would buy a replacement LCD inverter on eBay
simply because they're cheap and I hate waiting for parts to arrive.

I haven't found a good page explaining how to
replace the CCFL, though. Are they fairly generic, or do I need to
make sure I have an OEM replacement? And can you suggest a good set of
replacement instructions?

I just wrecked a Toshiba panel so I may not be the best person to ask.
These are the link I bookmarked for the exercise:
http://www.laptoprepair101.com/laptop/2007/12/09/replace-laptop-backlight-ccfl-lamp/
http://www.lcdparts.net/howto/instruction/ICCFL1.aspx
It is NOT easy and you will probably wreck a few panels as part of the
learning process. If you can find a dead LCD monitor, it might be
useful to tear the panel apart for practice.

You can find panels on eBay, but they's not cheap:
http://www.ebay.com/sch/i.html?_nkw=t43+lcd+panel&_sacat=0&_odkw=t43+lcd&_ osacat=0

If you look at the first link above, you'll probably see the burned
spot on the CCFL terminal and rubber boot. This is where the HV has
arced over to the frame. It's also where the CCFL lamp terminal got
hot, and slightly charred the rubber boot. I've had to use high
voltage RTV (from my TV repair kit) to make sure that the insulation
is intact on reassembly.

The big danger in replacing the CCFL is in peeling or even touching
the row/column connections going to the LCD panel itself. Break the
connection and the panel is dead. I can sometimes use my hot air SMT
desoldering station to repair the damage, but my batting average is
lousy.

The inside-my-laptop site seems to recommend replacing the whole
display even if just the CCFL is faulty. Is replacing the CCFL that
much of a nightmare? My display is actually pretty good, so if I could
just replace the CCFL, I would, rather than take a chance of getting a
crap LCD.

What's your time worth? About $100 will get you a new display with
CCFL lamp. My shop time is $75/hr. I can do the CCFL job (including
checking out the laptop, cleaning out the dust, etc) in about 2 hrs
for a total of about $170. Or, I can replace the panel in 1 hr, for a
total of $175. Given the high chances of failure with replacing the
CCFL lamp, I would suggest a panel replacement.

It really depends on which LCD panel you have.
http://www.thinkwiki.org/wiki/Category:T43
The T43 comes in 1024x768 or 1400x1050 in 14" and 15" panels. If you
have a 1400x1050 15" display, it might be rather expensive to replace.

If you have the time, and are willing to take the chance, I would try
the CCFL replacement. The only catch is that the lamps are about
$15-$25/ea including shipping, which makes this a potentially
expensive learning experience.



--
Jeff Liebermann
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558

On 7/6/2012 8:56 AM, Jeff Liebermann wrote:
On Fri, 06 Jul 2012 00:02:54 -0700, wrote:

Here's another article on the problem:
http://reviews.ebay.com/Diagnosing-laptop-backlight-problems?ugid=10000000005308726

What do you mean by dimming? Dim? Dark? off?

When the CCFL bulbs crap out, the display initially becomes pink
colored. As more air leaks in, the CCFL lamp becomes dimmer and
dimmer until it finally quits altogether. Some inverters have
protection circuits to detect CCFL faults and will turn off the
display after a few seconds. I forgot what the thinkpad inverters do.
Note that this is for CCFL backlighting, not LED backlighting.

Another way to check the inverter is to waive an oscilloscope
probe around the area of the inverter. If it's running, you
can often see the radiation from it on the scope. That won't prove
the inverter board is good, but might lead you to check the lamp
before buying an inverter.

Good idea.

Don't go probing the inverter circuit unless you know what you're
doing. There's voltages there that the scope won't like at all.
You might find yourself with a dead laptop and a dead scope.

Bah. I use two 10:1 probes setup for differential input. My probes
will handle 1000VDC so there's not going to be a problem with the
typical 500-700V peak output from the inverter. However, if you
forget and set the probe to 1:1, there's going to be an expensive
problem in the scope.

You've got better probes than most people. Probes are significantly
derated at higher frequencies. What's the peak voltage out of the
inverter when the lamp is open?
Bottom line is that most people asking the question asked here
would be better off not probing the output of the inverter.

Most people don't have 'em, but an ultrasonic leak detector translates
ultrasonics down to the audible range. You can easily hear the inverter
transformer "singing" and hear it change as you adjust the brightness
setting.

Great idea. I have an HP 4917A and never thought to use it to check
the inverter. Thanks.

I'm interested in hearing how to check CCFL lamps.
In the old days, you could plug on just about any inverter and see if
it lights. Today, the proliferation of connectors and inverter
topologies makes this harder.

I have an inverter used by gamers to light up the inside of their
tower cases glued to a board. I've ripped off connectors from dead
LCD inverters and built a connector octopus. I've only found 4
different HV connectors so it's not an ugly mess yet.

I bought a fluorescent lamp checker at a garage sale.
It's basically a transmitter. You waive the antenna around
the lamp and see if it lights. Works great on household fluorescents.
I've been afraid to try it on a laptop for fear that it'd blow up
the display drivers.

Don't do it. Chances are good that you're going to apply HV to some
place that won't appreciate the voltage. Incidentally, I have a
customer that keeps blowing up monitors, mice and keyboards. It's
obvious that it's his "negative ion generator" that's causing the
problem, but he insists on using it near the computah.

I don't have that model, but I did have a thinkpad with a bad fuse.
Was located near the display connector and well disguised. Didn't look
like a SMT fuse at all. Was open and extinguished the backlight.

I saw that on a Thinkpad T22 or T23. If you short something in the
external LCD connector, it blows the fuse, which also kills the LCD
display. The clue is that BOTH the internal LCD and external video
are dead. I found a rather vague article on the problem:
http://www.tomshardware.com/forum/65493-35-strange-thinkpad-video-problem
Finding the fuse was easy. There was an F(something) reference
designator silkscreened nearby.

Incidentally, I have several USB microscopes that are quite handy for
looking at SMT components that my rapidly fading eyesight can't see.

Got a recommendation for a cheap one of those?
I got my hands on one once with a killer zoom, but didn't think
I'd use it. Wish I'd bought it.
I do have a dental camera with fiber optic light source. Looks like
a colonoscopy machine. Gets real close, but takes up way to much space
to justify infrequent use.

The ideal solution would be a high resolution camera with autofocus
and zoom. Need to get it far enough away to make room for the
soldering iron under it.
I still haven't mastered the coordination needed to look one direction
at a display and solder in another direction.

On 7/6/2012 8:02 AM, spamtrap1888 wrote:


I suspect it is my CCFL lamp, because I still have an image, and it
has a rosy glow.

You shoulda led with that statement.
That's common among IBM laptops.

Replacing a lamp is a crap shoot.
I've done some that had removable panels that gave
access to the lamp.
Others have it buried DEEP in the frame and you have
to completely disassemble the display.
Be very careful when you unfold the boards with the
display drivers. It's not difficult to turn a dim
display into one full of bright lines.
Don't mix up the order or direction
of all those plastic sheets that fall out.

Like most things, it's a LOT easier the second time
you do it.

mike wrote in message ...
On 7/6/2012 8:02 AM, spamtrap1888 wrote:


I suspect it is my CCFL lamp, because I still have an image, and it
has a rosy glow.

You shoulda led with that statement.
That's common among IBM laptops.

Replacing a lamp is a crap shoot.
I've done some that had removable panels that gave
access to the lamp.
Others have it buried DEEP in the frame and you have
to completely disassemble the display.
Be very careful when you unfold the boards with the
display drivers. It's not difficult to turn a dim
display into one full of bright lines.
Don't mix up the order or direction
of all those plastic sheets that fall out.

Like most things, it's a LOT easier the second time
you do it.


Amazing technology in their own right those sheets of plastic - I've never
seen a laptop with a noticeably dim centre of the display or farthest edge,
if only one light tube.

On 6/07/2012 3:35 PM, spamtrap1888 wrote:
So how can I tell if it is the display, the backlight, or the
inverter?

And could someone give me an idea what's involved in replacing each?


I once replaced both the inverter and the tube on a Thinkpad, only to
discover that the actual problem was a break in one of the conductors of
the ribbon cable joining them.

Sylvia.

On Sat, 07 Jul 2012 02:46:02 -0700, mike wrote:

You've got better probes than most people. Probes are significantly
derated at higher frequencies. What's the peak voltage out of the
inverter when the lamp is open?

I don't know, but I can measure one after I'm back from my
mini-vacation. I don't like running the inverters without a load as
I'm afraid the xformer will arc over. So, I've never bothered to
measure it. Also, some inverters will shut down if they see an open
or a shorted load.

My probes are nothing special. I bought them on eBay from some random
Chinese source. I got tired of constantly fixing my collection of HP
and Tek probes. So, I bought 10 new probes and am slowly blowing them
up as I blunder forward. By using two of these probes with
differential inputs to the scope, I essentially double the maximum
voltage rating of the probes.

What I really should use is a 100:1 probe that can handle 1200VDC:
http://www.dealextreme.com/p/250mhz-oscilloscope-probe-max-1200v-35610


Bottom line is that most people asking the question asked here
would be better off not probing the output of the inverter.

Maybe. However, I figured that anyone that owns an oscilloscope
should know something about how to use it and about its limitations.

Incidentally, I have several USB microscopes that are quite handy for
looking at SMT components that my rapidly fading eyesight can't see.

Got a recommendation for a cheap one of those?
I got my hands on one once with a killer zoom, but didn't think
I'd use it. Wish I'd bought it.
I do have a dental camera with fiber optic light source. Looks like
a colonoscopy machine. Gets real close, but takes up way to much space
to justify infrequent use.

Same problems. I bought several different styles on eBay for about
$20/ea. Something like this one:
http://www.ebay.com/itm/221064301923
All are rather crudely built, fat, etc. No adjustable illumination.
Very crude mounting hardware. The software universally sucks, but
fortunately there are substitutes. A major problem was that the
required drivers all interfere with each other, with my microscope
camera, and with my other USB video cameras. I don't know what to
recommend because nothing works perfectly.

The original plan was to strap the camera on the back of my hand to
aid in soldering. I'm getting old and my eyes are deteriorating. I
have trouble seeing SMT parts and thought the camera will help. It
does, but due to the lack of depth of field, the ability to keep the
camera in focus is limited. I've switched to surgeons eyeglasses,
which are much better.

The ideal solution would be a high resolution camera with autofocus
and zoom. Need to get it far enough away to make room for the
soldering iron under it.

I just put a layer of cellulose wrap over the camera lens to keep the
flux fumes from trashing the lens. Doesn't affect the image very
much, but does keep the lens clear. What I would really like is a
pseudo color far-IR camera, so I can see how hot things are getting
when I use my hot air SMT desoldering gun.

I still haven't mastered the coordination needed to look one direction
at a display and solder in another direction.

Yeah, that was another problem. I did OK, but was still somewhat
clumsy. The surgeons glasses were much easier.

--
Jeff Liebermann
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558

On Friday, July 6, 2012 1:35:42 AM UTC-4, spamtrap1888 wrote:
So how can I tell if it is the display, the backlight, or the
inverter?

And could someone give me an idea what's involved in replacing each?

It is difficult to tell if the CFL tubes or the power supply is bad unless you have the proper test equipment to determine this. Generally speaking, after the CFL tubes go weak, the power supply also has a lot of wear on it. When I used to change the CFL tubes I also changed the power supply at the same time. This gives better reliability.

Jerry G. wrote in message
...
On Friday, July 6, 2012 1:35:42 AM UTC-4, spamtrap1888 wrote:
So how can I tell if it is the display, the backlight, or the
inverter?

And could someone give me an idea what's involved in replacing each?

It is difficult to tell if the CFL tubes or the power supply is bad unless
you have the proper test equipment to determine this. Generally speaking,
after the CFL tubes go weak, the power supply also has a lot of wear on it.
When I used to change the CFL tubes I also changed the power supply at the
same time. This gives better reliability.

++++

Can something weaken in the inverter section , assuming a perfectly good
tube ? Or is it just increase in leakage through disc ceramics, wiring or Tx
winding insulation.

On Saturday, July 7, 2012 3:53:32 AM UTC-7, Sylvia Else wrote:
On 6/07/2012 3:35 PM, spamtrap1888 wrote:
So how can I tell if it is the display, the backlight, or the
inverter?

And could someone give me an idea what's involved in replacing each?


I once replaced both the inverter and the tube on a Thinkpad, only to
discover that the actual problem was a break in one of the conductors of
the ribbon cable joining them.


Was the symptom no video at all?

On Sunday, July 8, 2012 6:45:08 AM UTC-7, Jerry G. wrote:
On Friday, July 6, 2012 1:35:42 AM UTC-4, spamtrap1888 wrote:
So how can I tell if it is the display, the backlight, or the
inverter?

And could someone give me an idea what's involved in replacing each?

It is difficult to tell if the CFL tubes or the power supply is bad unless you have the proper test equipment to determine this. Generally speaking, after the CFL tubes go weak, the power supply also has a lot of wear on it.. When I used to change the CFL tubes I also changed the power supply at the same time. This gives better reliability.

Jeff L: How necessary is this?

Jeff Liebermann wrote in
:

On Sat, 07 Jul 2012 02:46:02 -0700, mike wrote:

You've got better probes than most people. Probes are significantly
derated at higher frequencies. What's the peak voltage out of the
inverter when the lamp is open?

I don't know, but I can measure one after I'm back from my
mini-vacation. I don't like running the inverters without a load as
I'm afraid the xformer will arc over. So, I've never bothered to
measure it. Also, some inverters will shut down if they see an open
or a shorted load.

My probes are nothing special. I bought them on eBay from some random
Chinese source. I got tired of constantly fixing my collection of HP
and Tek probes. So, I bought 10 new probes and am slowly blowing them
up as I blunder forward. By using two of these probes with
differential inputs to the scope, I essentially double the maximum
voltage rating of the probes.

What I really should use is a 100:1 probe that can handle 1200VDC:
http://www.dealextreme.com/p/250mhz-oscilloscope-probe-max-1200v-35610


Bottom line is that most people asking the question asked here
would be better off not probing the output of the inverter.

Maybe. However, I figured that anyone that owns an oscilloscope
should know something about how to use it and about its limitations.

Incidentally, I have several USB microscopes that are quite handy for
looking at SMT components that my rapidly fading eyesight can't see.

Got a recommendation for a cheap one of those?
I got my hands on one once with a killer zoom, but didn't think
I'd use it. Wish I'd bought it.
I do have a dental camera with fiber optic light source. Looks like
a colonoscopy machine. Gets real close, but takes up way to much space
to justify infrequent use.

Same problems. I bought several different styles on eBay for about
$20/ea. Something like this one:
http://www.ebay.com/itm/221064301923
All are rather crudely built, fat, etc. No adjustable illumination.
Very crude mounting hardware. The software universally sucks, but
fortunately there are substitutes. A major problem was that the
required drivers all interfere with each other, with my microscope
camera, and with my other USB video cameras. I don't know what to
recommend because nothing works perfectly.

The original plan was to strap the camera on the back of my hand to
aid in soldering. I'm getting old and my eyes are deteriorating. I
have trouble seeing SMT parts and thought the camera will help. It
does, but due to the lack of depth of field, the ability to keep the
camera in focus is limited. I've switched to surgeons eyeglasses,
which are much better.

The ideal solution would be a high resolution camera with autofocus
and zoom. Need to get it far enough away to make room for the
soldering iron under it.

I just put a layer of cellulose wrap over the camera lens to keep the
flux fumes from trashing the lens. Doesn't affect the image very
much, but does keep the lens clear. What I would really like is a
pseudo color far-IR camera, so I can see how hot things are getting
when I use my hot air SMT desoldering gun.

I still haven't mastered the coordination needed to look one direction
at a display and solder in another direction.

Yeah, that was another problem. I did OK, but was still somewhat
clumsy. The surgeons glasses were much easier.


one thing; there's usually a film cap between the two CFL filaments,that
with the tube OFF,has a relatively low reactance compared to the unlit
tube,and allows the filaments to conduct the voltage and heat up,and after
the tube strikes it's arc,the arc shorts around the cap and the filaments
turn off.

If that cap shorts,the tube will never strike an arc,and the filaments stay
lit.

--
Jim Yanik
jyanik
at
localnet
dot com

On Sun, 08 Jul 2012 11:41:24 -0500, Jim Yanik
wrote:

one thing; there's usually a film cap between the two CFL filaments,that
with the tube OFF,has a relatively low reactance compared to the unlit
tube,and allows the filaments to conduct the voltage and heat up,and after
the tube strikes it's arc,the arc shorts around the cap and the filaments
turn off.

If that cap shorts,the tube will never strike an arc,and the filaments stay
lit.

Nope. There are no hot filaments on CCFL tubes. That's why they're
called "cold cathode".

I've taken apart many LCD panels and have never seen a capacitor at
the CCFL lamp terminals. There's the stiff wire coming out of the
glass tube, a flexible wire soldered to it, and a rubber insulating
cover. No capacitor. Something like this:
http://www.laptoprepair101.com/wp-images/ccfl/22-cable-soldered-to-backlight.jpg


--
Jeff Liebermann
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558

On Sun, 8 Jul 2012 09:19:08 -0700 (PDT), wrote:

On Sunday, July 8, 2012 6:45:08 AM UTC-7, Jerry G. wrote:
On Friday, July 6, 2012 1:35:42 AM UTC-4, spamtrap1888 wrote:
So how can I tell if it is the display, the backlight, or the
inverter?

And could someone give me an idea what's involved in replacing each?

It is difficult to tell if the CFL tubes or the power supply is bad unless
you have the proper test equipment to determine this. Generally
speaking, after the CFL tubes go weak, the power supply also has a lot
of wear on it. When I used to change the CFL tubes I also changed the
power supply at the same time. This gives better reliability.

Jeff L: How necessary is this?

Probably a good idea especially if the LCD inverter is cheap. In the
T43, the LCD inverters on eBay are quite cheap at $4 to $6/ea. If you
recall, I suggested that you purchase two:
www.ebay.com/sch/i.html?_sacat=0&_nkw=thinkpad+t43+lcd+inverter

It's true that trying to drive a failing CCFL tube is going to cause
problems with the LCD inverter. What happens is that as the tube
becomes gassy, it ceases to conduct, which causes the LCD inverter to
drive an open circuit. That raises the voltage at the transfomer
output. Eventually, the insulation on the transformer breaks down and
produces tiny arcs. This does take a while, but most users will try
to use the laptop until the display completely quits. Eventually, the
transformer will short, but in the mean time, the xformer will have
holes in the insulation. These could arc over later for other reasons
(e.g. moisture)[1]. There's no guarantee that whatever you buy on
eBay is any better than what you're replacing, but the odds are in
your favor. That's also why I suggested buying a spare inverter.

Hmmmm... ebay lamp prices vary from $4 to $18.
www.ebay.com/sch/i.html?_nkw=thinkpad+t43+lamp


[1] I've only done 4 autopsies on LCD inverters to see what has
failed. In 3 of the 4, the xformer had a shorted turn. I'm not sure
if this is a common characteristic as all 4 came out of identical HP
something laptops. I never could determine the cause of failure on
the 4th inverter.

--
Jeff Liebermann
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558

On 7/8/2012 9:41 AM, Jim Yanik wrote:
What I would really like is a
pseudo color far-IR camera, so I can see how hot things are getting
when I use my hot air SMT desoldering gun.

I'm not optimistic about that.
I have a crude far-IR camera used by firemen to find people
in smokey buildings. It's not calibrated, but I find that
the emissivity
of stuff varies so widely that it's useless for determining
relative temps.
If you aim a non-contact IR thermometer at a motorcycle
radiator, it reads reasonably accurately. Point it at the
aluminum cylinder and it's WAY, WAY off.

I cobbled together a crude setup for reflowing nVidia chips
on laptops. Gave up completely on IR and went with thermocouples
for temp measurement.

On Sun, 08 Jul 2012 12:17:17 -0700, mike wrote:

On 7/8/2012 9:41 AM, Jim Yanik wrote:
What I would really like is a
pseudo color far-IR camera, so I can see how hot things are getting
when I use my hot air SMT desoldering gun.

Ummm... watch your attributions. I wrote the above.

I'm not optimistic about that.
I have a crude far-IR camera used by firemen to find people
in smokey buildings. It's not calibrated, but I find that
the emissivity
of stuff varies so widely that it's useless for determining
relative temps.

True. All of them have a calibration knob, mostly to compensate for
changes in ambient temperature.

If you aim a non-contact IR thermometer at a motorcycle
radiator, it reads reasonably accurately. Point it at the
aluminum cylinder and it's WAY, WAY off.

Point it at a mirror and you get the temperature of whatever you see
in the mirror, not the temperature of the mirror. Yes, emissivity
matters.

I cobbled together a crude setup for reflowing nVidia chips
on laptops. Gave up completely on IR and went with thermocouples
for temp measurement.

Most of DVM's have thermocouple inputs. I got lucky and bought a
large box of pre-made thermocouples, in a variety of packages. I use
the tiny ones for spot temperatures. Still, it would be nice to have
a non-contact method that works.

In the past, I was working on converting a supermarket laser scanner
mechanism into a flying spot IR imager. I had something that sorta
worked but the response time was so slow, that is was nearly useless.
If I have time, I plan to continue as I now have a faster
microbolometer.

--
Jeff Liebermann
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558

On 7/8/2012 3:36 PM, Jeff Liebermann wrote:
On Sun, 08 Jul 2012 12:17:17 -0700, wrote:

On 7/8/2012 9:41 AM, Jim Yanik wrote:
What I would really like is a
pseudo color far-IR camera, so I can see how hot things are getting
when I use my hot air SMT desoldering gun.

Ummm... watch your attributions. I wrote the above.

I'm not optimistic about that.
I have a crude far-IR camera used by firemen to find people
in smokey buildings. It's not calibrated, but I find that
the emissivity
of stuff varies so widely that it's useless for determining
relative temps.

True. All of them have a calibration knob, mostly to compensate for
changes in ambient temperature.

The one's that I've seen have a calibration for emissivity.
Temperature compensation is internal and fixed.


If you aim a non-contact IR thermometer at a motorcycle
radiator, it reads reasonably accurately. Point it at the
aluminum cylinder and it's WAY, WAY off.

Point it at a mirror and you get the temperature of whatever you see
in the mirror, not the temperature of the mirror. Yes, emissivity
matters.

Depends on your definition of mirror.
If it's a front-surface mirror, yes.
IR doesn't go thru the glass very well.
I just did the experiment with a standard back-surface mirror
and a clear sheet of glass. Readings were similar and not anywhere
near the temp of the light bulb I was using as a a test source.

I cobbled together a crude setup for reflowing nVidia chips
on laptops. Gave up completely on IR and went with thermocouples
for temp measurement.

Most of DVM's have thermocouple inputs. I got lucky and bought a
large box of pre-made thermocouples, in a variety of packages. I use
the tiny ones for spot temperatures. Still, it would be nice to have
a non-contact method that works.

In the past, I was working on converting a supermarket laser scanner
mechanism into a flying spot IR imager. I had something that sorta
worked but the response time was so slow, that is was nearly useless.
If I have time, I plan to continue as I now have a faster
microbolometer.

On 9/07/2012 2:15 AM, wrote:
On Saturday, July 7, 2012 3:53:32 AM UTC-7, Sylvia Else wrote:
On 6/07/2012 3:35 PM, spamtrap1888 wrote:
So how can I tell if it is the display, the backlight, or the
inverter?

And could someone give me an idea what's involved in replacing each?


I once replaced both the inverter and the tube on a Thinkpad, only to
discover that the actual problem was a break in one of the conductors of
the ribbon cable joining them.


Was the symptom no video at all?


Intermittently. When the video vanished, the image was just visible when
looked at from the correct angle. Although I didn't bother to prove it,
the broken conductor was presumably the one leading to the lamp. Breaks
in other conductors would have different effects.

Sylvia.

On Sun, 08 Jul 2012 18:21:20 -0700, mike wrote:

True. All of them have a calibration knob, mostly to compensate for
changes in ambient temperature.

The one's that I've seen have a calibration for emissivity.
Temperature compensation is internal and fixed.

Y're right. The antique design that I threw together didn't have AGC.
Range was determined by knobs and trim pots.

The FLIR imagers that I sometimes borrow has manual adjustments for
image temp range and emissivity as well automatic DDE (digital detail
enhancement) also known as AGC. On the side or bottom of the screen,
is a crude color to temp calibration strip.
http://www.flir.com/cs/emea/en/view/?id=42844
One of the product features is that it's almost point and shoot, with
all adjustments being automatic.

Point it at a mirror and you get the temperature of whatever you see
in the mirror, not the temperature of the mirror. Yes, emissivity
matters.

Depends on your definition of mirror.
If it's a front-surface mirror, yes.
IR doesn't go thru the glass very well.
I just did the experiment with a standard back-surface mirror
and a clear sheet of glass. Readings were similar and not anywhere
near the temp of the light bulb I was using as a a test source.

Try it with low-E window glass. It reflects IR from the surface
coating.
http://www.wisegeek.com/what-is-low-e-glass.htm

I just tried the mirror and light bulb test. Your correct. It
doesn't work. I do have a borosilicate glass SIR "hot mirror" that
reflects IR, but that's cheating. I tried it on the bathroom mirror
and it didn't reflect much heat. However, a polished piece of
aluminum sheet metal reflects IR just fine.

Zero for three. I'm not doing too well today.

Argh... the rubber grips on my Craftsman IR thermometer are oozing
oil, getting sticky, and slowly disintegrating. What else can go
wrong?

--
Jeff Liebermann
150 Felker St #D http://www.LearnByDestroying.com
Santa Cruz CA 95060 http://802.11junk.com
Skype: JeffLiebermann AE6KS 831-336-2558