Acer AL2423W Recapped, backlight works but no display

[ndugu]

Re: Acer AL2423W Recapped, backlight works but no display

[QUOTE=Per Hansson;116676]Welcome to Badcaps and thank you very much for your post ndugu!
Did you have to use an oscilloscope to measure pins 2, 5 & 6 on U8?

[ndugu]

Re: Acer AL2423W Recapped, backlight works but no display

Yes, it is desirable an oscillograph

[Per Hansson]

Re: Acer AL2423W Recapped, backlight works but no display

I should mention I have ordered the X1 crystal and the Teapo capacitor for position C147 that I did not have in stock (it sits right on the power line to the panel FYI...)

I got my small caps order from Elfa today
Swapped out C147; a 100uF 25v Teapo cap for Rubycon ZL, no difference
Swapped out the 14.318Mhz X1 crystal next, also no difference...

I guess next test will have to be with a scope, thanks for the clarifications ndugu!
Btw, what sort of waveforms should I be expecting to see sent to the LVDS PCB via CN5 cable?
And also on the flash chip, what sort of waveform should be present here?
The scope I'll be using will be max 20Mhz, will this be enough?

[ndugu]

Re: Acer AL2423W Recapped, backlight works but no display

20 MHz it is enough. About waveform - a signal not periodic what can be the form - simply you observe activity of a signal in a big way no more Vcc for U8. If after short-term activity the signal disappears, the processor means has hung.

If the processor does not react to pressing of buttons it speaks that the processor does not work - or it is stopped because of program failure, or has broken. In this case business does not reach bus LVDS.

Look also at a reset signal on the processor, the reason in it can.

[Per Hansson]

Re: Acer AL2423W Recapped, backlight works but no display

Connected the monitor via VGA now to my Linux machine, in my first post I read the EDID data via DVI with GPU-Z, just wanted to see if it worked via VGA aswell...
And it does, no picture is displayed of course...

[root@localhost ~]# monitor-edid
Name: AL2423W
EISA ID: ACR0977
Screen size: 51.8 cm x 32.4 cm (24.05 inches, aspect ratio 16/10 = 1.60)
Gamma: 2.2
Analog signal
Max video bandwidth: 170 MHz

    HorizSync 31-81
    VertRefresh 56-75

    # Monitor preferred modeline (60.0 Hz vsync, 74.0 kHz hsync, ratio 16/10, 94 dpi)
    ModeLine "1920x1200" 154 1920 1968 2000 2080 1200 1203 1209 1235 -hsync +vsync

And the long verbose output if you want to see it...

[root@localhost ~]# monitor-edid -v
probing EDID using VBE (port 0)
 vbe: int 10h points to c000:0d18: 52 50 ba ca 03 ec fe c0
VBE version: 3.0, oem version = 4.52
Memory: 65536k
OEM name: NVIDIA
Vendor name: NVIDIA Corporation
Product name: NV34 Board - q231-0
Product revision: Chip Rev
Port 0:
 DDC1 not supported
 DDC2 supported
 Screen not blanked during data transfer
 Time to transfer one EDID block: 1 sec (rounded up)
parsing EDID from vbe
Name: AL2423W
EISA ID: ACR0977
Screen size: 51.8 cm x 32.4 cm (24.05 inches, aspect ratio 16/10 = 1.60)
Gamma: 2.2
Analog signal
Standard resolution: 640x480 @ 60 Hz (established timing)
Standard resolution: 640x480 @ 67 Hz (established timing)
Standard resolution: 640x480 @ 72 Hz (established timing)
Standard resolution: 640x480 @ 75 Hz (established timing)
Standard resolution: 720x400 @ 70 Hz (established timing)
Standard resolution: 800x600 @ 56 Hz (established timing)
Standard resolution: 800x600 @ 60 Hz (established timing)
Standard resolution: 800x600 @ 72 Hz (established timing)
Standard resolution: 800x600 @ 75 Hz (established timing)
Standard resolution: 832x624 @ 75 Hz (established timing)
Standard resolution: 1024x768 @ 60 Hz (established timing)
Standard resolution: 1024x768 @ 70 Hz (established timing)
Standard resolution: 1024x768 @ 75 Hz (established timing)
Standard resolution: 1280x1024 @ 75 Hz (established timing)
Standard resolution: 1920x1200 @ 60 Hz, ratio 16/10
Standard resolution: 1680x1050 @ 60 Hz, ratio 16/10
Standard resolution: 1600x1200 @ 60 Hz, ratio 4/3 (!)
Standard resolution: 1280x720 @ 60 Hz, ratio 16/9 (!)
Standard resolution: 1280x1024 @ 60 Hz, ratio 5/4 (!)
Standard resolution: 1280x960 @ 60 Hz, ratio 4/3 (!)
Standard resolution: 1152x864 @ 75 Hz, ratio 4/3 (!)
Standard resolution: 1440x900 @ 60 Hz, ratio 16/10
Max video bandwidth: 170 MHz

    HorizSync 31-81
    VertRefresh 56-75

    # Monitor preferred modeline (60.0 Hz vsync, 74.0 kHz hsync, ratio 16/10, 94 dpi)
    ModeLine "1920x1200" 154 1920 1968 2000 2080 1200 1203 1209 1235 -hsync +vsync

[ndugu]

Re: Acer AL2423W Recapped, backlight works but no display

EDID-information can be read directly from U1 2402 on bus I2C without involvement of the processor.

Parable:
- What do you search under a lantern?
- The key has lost.
Where you have lost it?
- Yes on that road.
Why here you search?
-zdes is more light...

[ndugu]

Re: Acer AL2423W Recapped, backlight works but no display

Parable:
- What do you search under a lantern?
- The key has lost.
Where you have lost it?
- Yes on that road.
Why here you search?
- Here is more light...

[Per Hansson]

Re: Acer AL2423W Recapped, backlight works but no display

ndugu; Well I don't actually own a Oscilloscope, so therefore I am searching under the lantern light meanwhile
Hopefully a friend will help me out during the weekend...

[Per Hansson]

Re: Acer AL2423W Recapped, backlight works but no display

A friend came over with a 10Mhz oscilloscope, we have the same basic understanding of how a scope works; bloody none

Anyway after poking at the buttons and dials for a while we kinda made it make something which almost looked like something
When the monitor is started there is a very small (0.5sec or so) blip on pins #2 & #5, then nothing
And on Pin #6 there does seem to be a signal present... (Pin #6 = "Serial Clock To provide the timing of the serial interface.")
I guess this kinda jives well with the fact that I only got one small blip when I connected up to the serial port of the monitor that I soldered on...
I will try to read the EEPROM with my Willem programmer in some way now, I don't have an adapter that fits but maybe I'll manage to jig something up with a couple of wires and some ducttape

[Per Hansson]

Re: Acer AL2423W Recapped, backlight works but no display

IT'S ALIIIIVE!

Hi guys, I'm sooo happy now
Me and a friend tried all day to piggyback my Willem USB programmer ontop of the EEPROM while it was still on the logic board, we even managed to read it about half way or so once
Some more trials of duct tape etc later we where no closer to the solution though...

Anyway, finally he convinced me to take the bull by the horns;
So I desoldered the chip, and then we connected some wires to it and read it with the programmer, turned out the firmware did look corrupt
So next we flashed it with the firmware that I linked to on that Russian board above
Soldered the EEPROM back in place and the damn thing booted right up, all seems to be working well!

There are some Russian error menus on it now but I guess I can live with that, me having such a love for Russian software like nginx and the likes
Also it is always good to know what "no signal" means in Russian
(And after navigating the menus for a while maybe I'll manage to find what "English" means in Russian too )
Thank you allot all who helped me in my trials with this monitor!!!

Btw, I should add that I tested the serial port again, both with Hyperterminal and the gProbe software, however I got nothing at all, so probably some kind of special adapter is needed here...

[retiredcaps]

Re: Acer AL2423W Recapped, backlight works but no display

IT'S ALIIIIVE!

Soldered the EEPROM back in place and the damn thing booted right up, all seems to be working well!

Congratulations! Well done and great perseverance.

[PlainBill]

Re: Acer AL2423W Recapped, backlight works but no display

That's great news. I've got a Dell monitor around here somewhere with the same symptoms.

PlainBill

[ndugu]

Re: Acer AL2423W Recapped, backlight works but no display

I congratulate. However take your time to rejoice, in my case breakage has repeated in 1-1,5 months. Tell you has programmed the same Flash chip or new?

It I to that can be because of the bad power supplies Flash chip have acquired some deficiency and in due course the data are distorted. Then would be correct to deliver a new chip.
In my city on sale there is no given chip, therefore I cannot check up my theory.
Please inform me on rus_69@mail.ru if breakage repeats. If will not complicate.

[Per Hansson]

Re: Acer AL2423W Recapped, backlight works but no display

PlainBill & retiredcaps, thanks for the kind words and encouragement, I hope you have luck fixing your monitor too PlainBill!

ndugu; thank you for the heads up, if my display fails again I'll send you a mail (you could also just subscribe to this thread)

I did reprogram the same flash chip, I did flash erase, erase verify, than flash & verify
Then I disconnected the programmer and reconnected it again, and read the chip, and manually HEX compared it with the file I used to flash... All matched...

The flash chip is for sale here on eBay for example; http://cgi.ebay.com/SST25LF040A-33-4...-/200470003143

Though I do think it is strange the flash chip would be damaged by poor capacitors, so maybe you are correct, it could be some other defect.
Because I found after I had flashed my chip the settings in the monitor seemed to be preserved, so probably the nvram for settings is in one of those EDID chips (24AA02)
(So the question is why the flash becomes corrupt, if the monitor is not writing nvram settings to it that is...)

[PlainBill]

Re: Acer AL2423W Recapped, backlight works but no display

PlainBill & retiredcaps, thanks for the kind words and encouragement, I hope you have luck fixing your monitor too PlainBill!

ndugu; thank you for the heads up, if my display fails again I'll send you a mail (you could also just subscribe to this thread)

I did reprogram the same flash chip, I did flash erase, erase verify, than flash & verify
Then I disconnected the programmer and reconnected it again, and read the chip, and manually HEX compared it with the file I used to flash... All matched...

The flash chip is for sale here on eBay for example; http://cgi.ebay.com/SST25LF040A-33-4...-/200470003143

Though I do think it is strange the flash chip would be damaged by poor capacitors, so maybe you are correct, it could be some other defect.
Because I found after I had flashed my chip the settings in the monitor seemed to be preserved, so probably the nvram for settings is in one of those EDID chips (24AA02)
(So the question is why the flash becomes corrupt, if the monitor is not writing nvram settings to it that is...)

While it is possible the flash chip was damaged by bad power, the phenomena of the contents becoming by an unstable power supply is well known. Perhaps the most notorious example was several RCA TV chassis with poor solder connections.

As I recall the preferred solution was to repair the solder joints, use the service menu to view the various settings, replace the eeprom, and reenter all the settings.

PlainBill

[Per Hansson]

Re: Acer AL2423W Recapped, backlight works but no display

That is interesting PlainBill
I can't of course know if the settings my monitor now has where saved in nvram in some separate chip on the board
Or if the settings infact are stored in the flash chip I reflashed (I guess it very well could be that way, since the default language became Russian after the flash, and I downloaded the firmware off a Russian board...)

Of course I do now have a backup of my original corrupted firmware too, so if I'm really crazy some day maybe I'll try to find a partition table or something in it with the nvram parameters and try to read what they where

Here are some other fun tidbits from the firmware;

Revision:  1.72
Time=%d.*RSP Auto OK.*RSP Auto Fail.IA

Auto corrupt the firmware when warranty time is out?

V1.72 Beta_01.0525 14:00

Typical Acer QC control^^

16:30:14 Dec 22 2006.AUO_WUXGA_M240UW01_V0.BOARD TYPE: EV_BD.TCLK = %dMHz

AU Optronics MVA panel, very nice panel

HDCP_Init: error reading BKSV.&&&&&&&&&&&&&&&&&TEST0 %x.HDCP_Init: error reading BKSV.aaa.HDCP_Init (gencrypt enabled): NVRAM addr 0x%x.HDCP_Init (gencrypt disabled): NVRAM addr 0x%x.bbb.*************TEST %x.HDCP_Init: gm_ReadNVRAMAbsolute error reading keys

I think this will always fail, because the cheap bastards didn't implement the HDCP cryptographic chip AFAIK!

Also I should add that I replaced the 3x capacitors on the backlight preemptively before I put the monitor back together
They are pictured below, I can't make out the brand name (NK Con?)
They are rated at 180uF and 30v, they are 8mm wide and 9mm tall, I measured them at 0.11ESR and 178>180 microfarad (yes, really!)
The colors are very very close to Sanyo/Suncon, however the bung not so...
I replaced them with Rubycon ZL 220uF 35v caps, they have the same width but are 11.5mm tall, it was a very nice snug fit under the backlights metal cover

[Per Hansson]

Re: Acer AL2423W Recapped, backlight works but no display

Just managed to get into the service menu, this is what my monitors settings are right now;

[ndugu]

Re: Acer AL2423W Recapped, backlight works but no display

Yesterday again reflash my chip, at power up - then the screen short-term seemed has gone out, to buttons did not react. Has checked up flash - the code is corrupt after 7D002 address. Has anew programmed, the screen monitor has ceased to switch on, to buttons does not react. Again the code is corrupt after 7D004 address.

It seems that the processor, флэш here not has died and. Indirect tags that the processor unsuccessful it not datasheets on a site of the manufacturer, also are not present support in GPobe.

[Per Hansson]

Re: Acer AL2423W Recapped, backlight works but no display

My monitor has an annoying problem, I can change the values in the service menu, except "Gain" and "Offset1" which I can not move the cursor to
However the option "Offset" can be changed, but it is reset to firmware defaults after a complete powerloss (standby works fine)
The issue is the same for the setting "user" in the normal menu (not service menu) It resets the RGB values to 100/100/100 after a power loss
The monitor does however save most of the other settings, like "7500" and "9300" in the service menu, and contrast/brightness in the user menu, language too (now English)

I have created a thread about calibrating the monitor on this forum here, it was then I ran into this issue...
http://www.avsforum.com/avs-vb/showt...php?p=19048839

ndugu; this retailer sells the logic boards, there is even one that says "HDCP" in the name, they should work since they are listed as spare parts for monitor type Acer AL2423W
Look under the section "main board"
http://www.zandparts.com/acer/index....=3a&language=1

[Per Hansson]

Re: Acer AL2423W Recapped, backlight works but no display

Posting an update on this.
The monitor did not fail but it started to behave erratically, would hang on startup with the Acer logo.
Would be impossible to turn off, and other misc problems like that, leaving it unplugged for 10 mins and then starting it always did bring it back tho...

Per the advice of ndugu I bought a new SST25LF040 Flash chip, the seller ecking767 on eBay has them for sale at $10....

So I again desoldered the original SST chip that I had flashed with the Russian firmware
Since I now had something to compare with I could find some things out
Address;
7D384 = settings in service menu for "7500" setting in HEX format
7D387 = settings in service menu for "9300" setting in HEX format

So this confirms that I am using the calibration data from the Russian guys monitor who originally dumped this firmware. (Originally I thought maybe the settings where saved in some other chip...)
I did have a look in the original firmware from my monitor that I had dumped 5 weeks ago before reflashing it. But the flash was corrupted at this address, infact almost the whole flash was corrupted, thousands of lines did not match...

Now the reflashed original SST chip I had been running with up untill today was corrupted after address 7D086, which matches quite well with what ndugu says above aswell...

Sadly the monitor still does not remember the settings changed with this new eeprom in the normal menu system, "warm", "cool" or the RGB values under "user"
It also does not remember the Offset settings in the service menu, but it does remember the 7500 and 9300 settings just like before.
Now since I managed to find these settings in the eeprom it would seem logical that the other settings are stored in some other chip
However that seems contradicting because when I was poking with the monitor before I tried the menu option "auto color" which reset the gain to R:2 G:2 B:2 and And Offset1: R:104 G:124 B:79, which then could not be changed, but it was saved...
However now that I replaced the eeprom these setting where restored to what you can see in the picture posted a few posts back...