240Z Inverter, dim after 60 seconds....what is it?

[solidpro]

Hey everyone

I have a 240Z and the LCD goes dim after about 1 minute. I generally think it could be one of two things - bad CCFL tube or bad inverter (in some way).

Anyway, following a lead off a fellow member, I was quickly able to acquire a second LCD and swap it - the problem remains, meaning I either have two LCDs with bad CCFL tubes (unlikely) or I have a faulty inverter.

Now the inverter on the 240Z is different to that on the 240, most likely because the LCD is different (higher resolution), so they're much harder to find. This is the inverter, front and rear - it has no electrolytic caps to replace and the fuse is good (if it wasn't, I don't think it would work at all). The coil is soldered well and the solid-state caps don't have any markings on them to even attempt replacing them. I think if they failed they would just short anyway, rather than cause a delayed issue.

https://i.imgur.com/iO0RVlS.jpg
https://i.imgur.com/gLaHUOo.jpg

The only other thing I can think of doing is multimeter testing the voltage coming out of the CCFL connector before and after the LCD goes dim. It won't really help me find a resolution though. Has anyone got any ideas on how I might fix it?

Thanks!

FORUM warning:
picture(s) WAY too big, tags removed. Please read the Forum Rules, especially Section 5: https://forum.thinkpads.com/viewtopic.php?f=16&t=14339

[SMA]

It may sound strange and admittedly it is just a guess, but I'd suggest you try another AC Adapter.

[RealBlackStuff]

The likely possibility: both have a broken CCFL tube, which only 'shows' when the tube heats up.

[solidpro]

Hmmmm. I think what I'll have to do is go and unpack a good 240Z from the attic to borrow it's inverter to swap over. I hate taking the LCD bezels off unless I really have to because the plastic is so brittle...

[RealBlackStuff]

The only other thing I can think of doing is multimeter testing the voltage coming out of the CCFL connector before and after the LCD goes dim. It won't really help me find a resolution though. Has anyone got any ideas on how I might fix it?

Careful!
The CCFL connector output voltage is anywhere between 750V and 1500V. I kid you not!
On the T60 inverters I use in my LED-mods I noticed that those trafo's get hot.
Since I don't need them for a LED-mod, I remove them, see also https://theboardroom.info/trafo.html

[Voodle]

Lemme help you out again too - I broke the inverter output on my 240Z last year while swapping the LCD's plastic, since the flex cable for the inverter was in at a funny angle and shorted stuff.
"I'm trying to figure out how the backlight turns itself on since I broke that signal on my otherwise-functional Thinkpad 240Z. All this because I wanted to swap over the screen plastic and put on a cable cover that was missing, this all has gone very wrong but I can call it a 'learning experience' "
"My Thinkpad 240Z is back up and running, traced the backlight enable signal back to a little AND gate, swapped that over from the donor 240X board and now the 240Z's backlight works again, yay."

This lead me to deriving the pinout of the inverter on both the LCD end and the inverter end, which I can share here:


On your one, check the enable pin on the test point of the inverter while the laptop's running, it should be at 3.3v the whole time. That should be pin 9 on this picture:


Mine was not giving 3.3v because what I shorted broke the AND gate that ran the enable signal, the yellow tape is pointing at the enable pin on the backside of the motherboard:


Hopefully it's related to the enable signal on yours and my info can be helpful, I think the inverter is pretty solid state since there's not really caps to go bad.
Another thing you can try if the enable signal is there, is to have the laptop run a different CCFL that's in another display to see how that behaves.

Or use an external CCFL inverter to test the screen's CCFL - I got a generic inverter that I hooked up to run from 12v and bypassed the enable signal, great for display testing. But there is a blue-box ccfl tester that you can get that does the same thing. My blue-box one died, hence the hacked up generic one.
Since you swapped the screen, it shouldn't be the CFL I guess?

BTW I also do not recommend testing or probing the inverter's output, best to stay on the low voltage side of it.

[solidpro]

Wow, thanks for the info. I'm going to test that 3.3v rail shortly.

However, if it does die on 'later' like I think we all predict it will, what would I do about it?

Thanks

[solidpro]

Hi

Ok, when testing on test point 9, before and after the LCD goes dim, the 3.3v rail does not drop - it's a solid 3.29v continuously.... I'm going to get another inverter....

**edit**

It's not the LCDs or the inverters. I dismantled a known good 240Z and switched everything around in all different combinations and the dimming follows the base of the 240Z and not the LCDs/inverters. They're all good. So it's something on the mainboard.... any ideas people?

[Voodle]

That's great that it's a different problem I don't know that that AND gate is easy to source without a donor board.
Next place to try is gently checking the 5v logic for the inverter (at least I think it's logic) and then the 15.7v line. Looks like there's a good test point on the PCB side that faces outwards - the thick trace with two groups of 4x vias.

Other things to consider are whether there's a fuse running the inverter specifically, or perhaps there's a cold solder joint on the mainboard's LCD connector.

If the 5v logic and 15.7v and 3.3v enable signal are all good, maybe check the ADJ line to see whether it's in the range of like 0v to 3v. Perhaps if the ADJ signal is at 5v that'd basically turn off the inverter. Does the brightness adjust work okay while the laptop's still running the backlight?

[solidpro]

Gawd, I was actually dreaming about fixing this machine this morning. It's like a monkey on my back.

So, quick question - do you think the circuitry involved in the first minute or so of powering the inverter is different from thereafter? Could it be a different circuit is involved and therefore different fuses?

I'm reaching the end of my abilities. For example, I have a really decent oscilloscope but never worked out how to use it. I would be able to find a dead fuse though....

[SMA]

So, quick question - do you think the circuitry involved in the first minute or so of powering the inverter is different from thereafter? Could it be a different circuit is involved and therefore different fuses?

As I understand it, power wise, the input to the inverter is the 16V DC from the AC Adapter.
Going by Voodle's drawing above, that is inverter pin 1+2, measured to be 15.7V.
If the Thinkpad is operated on battery power the voltage on inverter pin 1+2 will be lower, because the battery voltage itself is lower.
Do you have a battery installed when testing? If you have a good battery you could test it when operated on battery power only.

I seem to remember having heard of Thinkpads where the brightness varied when the Thinkpad was changed from being AC powered to being battery powered (and the reverse).

It is very unlikely that this is a fuse issue. Fuses are either completely on or completely off.

You should try to learn more about what is going on. Replacing components is one way of doing but the quick way is to take Voodle's suggestion about measuering voltages on the inverter pins.

[solidpro]

Ok, good ideas. Later today I'm going to test using different batteries, different PSUs.... will report back!