Dell 2007FPB teardown

[budm]

After reading the http://www.badcaps.net/forum/showthread.php?t=10658
on Dell 2007. I found 2 of them today and took them apart to see how it is put together.
What I found are:
1) It only has one always on DC power supply output of 19V.

2) The 19V is fed to the Backlights inverter board.

3) The 19v is also fed (J4) the USB Hub/3 Buck converter regulators (3.3V (Q6) for the Logic board where the VGA/DVI, control panel are, 5V (Q14) (this 5V is also fed into 3.3V linear regulator for the USB interface IC) for the 4-port USB hub (these 4 USB ports are protected by 4 fuses), 12V (Q9) for the powered speaker bar DC output jack. This 19V is fed through 3A fuse. MOSFETs Q6, 9, and 14 are 2SJ598, P-Channel.

4) If the USB/Buck converter is not connected, the monitor will not work.

5) The 3.3V output from the Buck converter is fed back into the logic board which feeds U13 (1.8V Linear regulator) and U5 (2.5V Linear regulator) to run the LOGIC IC's on the logic board.

6) One bad design in the Buck regulator is that they parallel 2 Schottky Diodes together (http://electronics.stackexchange.com...des-a-bad-idea) to try to get more current handling, but if you do not precisely match the Vf of the Diodes, the one with lower Vf will be on, the one with higher Vf may not be on at all. They should have use single 2 or 3A rating Diode instead of 1A rating, Diodes are rated at 1A 40V. This may explain why this model has the same failure mode due to shorted out diodes.

[budm]

More pictures.

[spaceman]

Thanks for posting these. I had taken one of these apart months ago and somehow lost all of my breakdown photos. These certainly came in helpful.

[koru]

Very useful post - many thanks!

[ssheikh]

Sorry for reviving an old thread. And thank you for excellent pictures. I now have all three of my 2007FP dead. Surprisingly, the paralleled diodes on the USB board in all three monitors are ok. But ZD701 on all three is shorted out. Do you by any chance know the rating of this zener?

thx.

[budm]

ZD701 on which board? Its been awhile so I do not remember all the parts are, I will have to look for the board I have again.

[kalmara]

Hello, friends electronics & electricians.
I have recently aquired a Dell 2007FPB monitor and it's in near mint condition, but sadly - not functioning.
I work in the electronics area and I'm aware of the Ohm's law and etc., but it's difficult to repair the monitor without a service manual, even worse - without a single schematic. The pics and info in this thread have been very useful and I made a couple of conclusions but need help in order to continue with the repair.
My questions are mainly to the thread starter, budm, seems like you are a lot into this stuff, so hope you can help.

Things done so far :
Tested mains supply - 19VDC is OK.
USB/Buck converter board - blown 3amp fuse (F19), all the schottky's and fets tested OK, so I powered the board like so :

The result - 5V are OK, but 3.3V and 12V are gone. After I saw the 3.3V are gone I looked at the post of the fellow above me and found out that my ZD701 has also gone short and by the looks of it - seems like some sort of feedback to the TL1451 PWM driver IC, presumably 3.3V zener?? (not sure) I tried to desolder it to see if the 3.3V will pop up, but it broke apart as soon as I touched one of it's leads with the soldering iron. Another test - still no 3.3V and 12V, then I decided to check the gates of the mosfets and sadly they both have +VCC on their gates (19VDC) - the totem pole driver transistors tested OK, so I suspect the death of one of the PWM driver IC's (they are dual, so one is used fully, and the other is used 50%, coz only 3 voltages are needed).
Moving on to the main CPU board - the 3.3V rail measures a dead short to GND. I thought it was one of the linear regulators, so I desoldered one of them and the input pin of the other one, like so :

But sadly enough again - no change. There are a lot of caps to be suspected, especially electrolityc, but my main concirn is how to check if the graphical CPU os okay or not, because it would be a big FAIL if it hase gone bad and started up the whole butterfly effect.

The man from whom I bought the monitor told me it has worked flawlessly for 5 (five) years, and one day suddenly had an intermittent power issue - started turning on and off by itself, and then died just like that. All in the matter of 2-3 minutes.

p.s. Sorry if I have mistakes in the text - my mother's language is Bulgarian, not English.

Cheers!

[kalmara]

Looking at my post, and desperately trying to find the S/C, measuring here and there I glanced at the multimeter and a lightbulb went off in my head.
Thinking in Ohm's law - my "dead short" is actually somwhere in the 3ohm region, on a 3.3V rail, thats just over an Amp. Given the "2A rating" of the supply rail (2x1A schottky diodes in the power supply) I'm starting to think that there's nothing wrong with my digital board and will try to focus getting the buck converter board to work and give it a shot with the *load* connected. Now all I have to do is figure out the Zener's voltage.

And about the two 1A schottky diodes in parallel - here's another point of view:
I doubt someone can match perfectly two diodes' Vf, especially in a massive manufactured think like a PC monitor, so ...
There are two diodes - one is necessarily with a lower Vf, so it will handle (if not the majority) most of the current, yes but that die gets hot with 1A flowing through it, so Vf gets higher and "the ball is passed" to the second diode, which initially had higher Vf, but now is the lower of the two. While the second diode is handling current and heating up, the first one is cooling down, and so on ...
But don't get me wrong - I think to change all of my double diodes with single SR34's (3A rated), that was just my 2 cents.

Cheers!

[budm]

Dell is too cheap to do it properly. Not good engineering practice to just parallel two discrete diodes, two on the same die may be.

http://books.google.com/books?id=nZz...rallel&f=false

http://www.electro-tech-online.com/t...apacity.95160/

Just Google proper way to parallel two diodes.

[kalmara]

Updates :
This morning I decided to go the trial & error way, since I don't have access to a scope. The first thing I did was swap the two PWM controller IC's in order to see will the problem power lines follow the IC's - I thought that the 3.3 and 12V will come up and the 5V will go down, but nothing actually changed. Conclusion - the problem is elsewhere.

And hour or so after that I found a resistor out of spec (think it was 4k7 or 47k, I forgot it already) under that white blob of dielectric-used-to-be stuff under the first PWM controller (upper right corner of my first pic), changed it and voila! The 3.3V rail came up, even without a zener.

Seems that the zener is going to be a 3.3V one, since it's cathode is directly on the 3.3V output (resp. anode on gnd), no current limiting resistors, no nothing. The PWM IC has it's resistor divider for the error amplifier input, just as those on the 5V and 12V lines, so I suppose something like thermal stability was in mind when they put that darn zener in there.
I decided not to search for the problem on the 12V line, since I don't have or plan on having dell speakers, so the hell with it.

Anyhow - now there's another problem. Without input signal the "NO VGA (or NO DVI-D) CABLE" message is absolutely perfect clear with perfect color and moving around the screen, but when VGA or DVI input is applied there is some sort of "picture", but it's quite noisy with white lines all over the place and it's jumping vertically. The other thing I noticed is when I left the monitor to display the "NO DVI-D CABLE" message for awhile, somethink like 10-15 minutes - after that it starts jumping vertically as well.

I ordered schottky diodes, zener diodes, some fuses and low ESR caps to beefen up the buck converter board and plan on changing ALL of the 10-22-47uF electrolytes in the digital board, and test again.
Meanwhile a friend of mine has prepared me an 2007WFP board, which is exactly the same as mine, with two differences :

First : The GPU is FLI5961-LF on my board, FLI5961H-LF on the widescreen board, which "H" suffix, according to some document I found on the internet means an "HDCP" (High-bandwith Digital Content Protection) feature on the DVI input of the GPU, so I suppose it'll work.

Second : the connector for the usb/buck board is on the opposite side of the pcb, but that can be redone in the matter of minutes. I think also swapping the FOUR (4) memory chips if i get to nowhere with my board and use his. Three are some atmel flash chips, and one is an 25x40 eprom.

[meguy]

Hello guys..!!!
I have a problem with my 2007FPb which is somewhat i dont now unique may be, firstly i was using it, after continous use of around 6-7 hours, the monitor began to flicker very much, fearing the damage i switched it off. after ample ahours , the monitor when i turned on it doesnot turn on. now i knew the flicker problem is mostly because of capacitors, so i opened it up. Then i found everything is perfectly ok visibly. NOw it is still open , connecting the display, power , odd thing happens, when i have colours on display i mean normal display it doesnot show up anything,lamp doesnot lit(i can see froma corner of the LCD) But when the display is black screen saver with little windows logo the lamp is flickering again very high and the LCD show the display although flickering.
What might be the Cause? any idea anybody?
what should i do? how to proceed with the testing?
Budm Specially your help would be of great benefit.
Regards

[ac21]

I had a 2007wpb that wouldn't turn on at all unless i let it sit plugged in for a few hours..then it would turn on long enough for the "no signal message to come on" then die again it would repeat this randomly.

After reading kalmara's post i checked the row of resisters under the "white blob of dielectric" on the usb board .. the first one is a 2.7k (printed on it 272) i first jumped a resister of unknown value across it and the monitor started working. then i desoldered the SMD and soldered in a regular 2.2k resister and put it back together ..everything is working great so far.

I read a bunch of post here and there but couldn't figure it out so i made this post so people in the same situation can do the same but maybe you should replace it with the right resister

here are some pictures of the location on the resister thanks to budm.

[IMG]http://imagizer.**************/v2/800x600q90/716/qgga.jpg[/IMG]
[IMG]http://imagizer.**************/v2/800x600q90/12/tp0c.jpg[/IMG]

[aortiz]

Originally Posted by petabyte

As this is a brand new forum, I thought I'd open up a thread to encourage new members to introduce themselves a bit.. now hey.. let's not get too personal.. ok ?

And if you wouldn't mind to keep this thread \"clean\" let's try to keep the the chitchat (i.e. even the welcome to the board replies) to a minimum or this thread may get to 100 pages fast

So I'll start things off :

My name is petabyte and I'm a forum junkie. :oops: I've been off the wagon for a number of years and countless times I've tried to stop but it's useless.. I need to feed my learning addiction.. :!:

Now just a bit about me.. I'm a fairly secretive guy in terms of privacy on the net.. especially when the things you write are visable to anyone with a net connection. But I love to share the info and knowledge I have, so that's why I'm here, to share and learn.

my formal education is that of an electronic engineer and I held a field service position with a firm for over 20 years.. that job is gone now along with company car and laptop <cry> oh well..
So needless to say (but i will) I've always tinkered with things.. yes, a hacker.. (too bad that term lost it's orignal meaning) a hardware hacker mostly.. hacking fixes in whatever I find that breaks.. (except for cars, I have friends for that) I got started a little late into the computers, my first box was an ibm pc-xt with a whopping 10meg HD,blazing fast 4.66mhz 8086 processor, 640k mem running dos 3.3.. man was I cool back then.. :roll: nevermind..

so these days, I'm forced to be a software guy as well.. as I fix pc's for friends and that usually means cleaning virus/trojan/adware and reinstalling OS's.. yeah hardware breaks but not that often :cry: so as a result, I keep up on my software knowledge by visting a few security forums.. and trolling around a few hardware forums as well.

now I found this place from a thread at motherboards.org/forums
where I've been hanging out lately.. it's a nice friendly place with some pretty smart people.. mostly, of course, it's about mobo's.. So this person mentioned badcaps.net and I recalled reading articles a year or so ago about the faulty caps and was curious as to what you guys were up to.. and I said \"w00t.. they have a forum\"..

and so now.. you're stuck with me :P

so what's your story ? come on.. give it up..

Originally Posted by budm

After reading the http://www.badcaps.net/forum/showthread.php?t=10658
on Dell 2007. I found 2 of them today and took them apart to see how it is put together.
What I found are:
1) It only has one always on DC power supply output of 19V.

2) The 19V is fed to the Backlights inverter board.

3) The 19v is also fed (J4) the USB Hub/3 Buck converter regulators (3.3V (Q6) for the Logic board where the VGA/DVI, control panel are, 5V (Q14) (this 5V is also fed into 3.3V linear regulator for the USB interface IC) for the 4-port USB hub (these 4 USB ports are protected by 4 fuses), 12V (Q9) for the powered speaker bar DC output jack. This 19V is fed through 3A fuse. MOSFETs Q6, 9, and 14 are 2SJ598, P-Channel.

4) If the USB/Buck converter is not connected, the monitor will not work.

5) The 3.3V output from the Buck converter is fed back into the logic board which feeds U13 (1.8V Linear regulator) and U5 (2.5V Linear regulator) to run the LOGIC IC's on the logic board.

6) One bad design in the Buck regulator is that they parallel 2 Schottky Diodes together (http://electronics.stackexchange.com...des-a-bad-idea) to try to get more current handling, but if you do not precisely match the Vf of the Diodes, the one with lower Vf will be on, the one with higher Vf may not be on at all. They should have use single 2 or 3A rating Diode instead of 1A rating, Diodes are rated at 1A 40V. This may explain why this model has the same failure mode due to shorted out diodes.

are buck regulators a circuitry on a board ?,or that is the name of the board? if so, which board is it of the 4 boards?

[aortiz]

are buck regulators a circuitry on a board ?,or that is the name of the board? if so, which board is it of the 4 boards?

[budm]

The buck regulator switching power supplies are on the board where the USB ports are. You can see 3 TOROID inductors on that board.

[aortiz]

i have a dell 2007WFPb that dont light up,the on light stays blue,and i can see what i think is the backlight(very dim,i think)you can only see it with the room lights off,i have some limited electronics knowledge,have a digital fluke 79 series2 multimeter,a soder sucker,solder wick,a 35 watt soldering iron,and very little knowledge on lcd monitors,but i do want to learn and can follow instructions,worked in california for some 20 years or so,as an electronics tech on a rapid transit system ,very mechanical oriented work more so than electronic,parts changer basically,i did get an as degree in electronics but that was some 30 years ago or so.anyway what do you recomend i start checking,and please poit out the name of the boards too.
please help.

[selldoor]

Aortiz - Please start your own thread. post good clear pictures of your boards front and back.

[aortiz]

pardon my ignorance but i dont know how to post a new thread,i looked all over the place for the "NEW THREAD" tab and cuouldnt find it,sorry,can you please help me with that? i am new at this forum thing .
thank you

[Davi.p]

browse through the forum tree, when you find this on top of the therads list there's a blue button to the left..

[aortiz]

thank you davi.p,i contacted the site and all of the sudden the blue "new thread" button was there.i think it had to do with my previleges as a new member,i was doing something or did something wrong,or did not do as requested.
i was able to start a new thread just a few minutes ago.
its so good there is good people like you willing to help a total stranger in trouble.
thank you