Dell E172fpb

[Bobdee]

Re: Dell E172fpb

hello again

https://www.badcaps.net/forum/showpo...0&postcount=85

re the link above
this posting and the following 3 by terry, krankshaft, and davmax should confirm it for you

[freddyzdead]

Re: Dell E172fpb

Ok, the referenced "C5707" means "2SC5707", they are the same. All Japanese transistors start with "2S", so it is common to omit it to save space. Unfortunately, European bipolar transistors all start with "BC", and I have seen cases where the manufacturer had left off the "B" so that a BC337 was marked as "C337" which could have easily been interpreted as 2SC337.

I repeat the warning about the quality of this board and the thinness of the copper tracks. Using a high-temperature iron to try and get the solder to flow better will most likely result in the tracks lifting from the board. Because they are so thin, they will just tear off if this happens. You should do the reverse, i.e. you should use the lowest heat that will do the job. A solder sucker is a must for removing components as this will traumatize the board the least.

Now I will tell you how my adventure is going.

I replaced the 2SC5707s with 2SC495Y transistors that I happened to have a supply of. These didn't even get warm, even though they are only rated at 1 amp collector current. They are small enough to not require any mods to make them fit. The "Y" means that they are the high-gain version, which is what is needed here.

What happened then is that the monitor worked for about ten seconds, then the backlight went out. I unplugged the CCFL plugs one at a time and verified that both inverters are indeed working. Then I tried lifting one end of R750 as suggested above. This resulted in the monitor staying on for about half a minute before the backlight turning off. So, lifting R750 obviously doesn't disable the feedback circuit, which I assume is what is at work here. As if the circuit is detecting that one or more lamps didn't fire.

What I can't find here is a decent circuit diagram of this board; all I can see is one that has been marked up with a modification that I don't think has anything to do with the problem, but the part numbers are illegible and don't improve when it is zoomed up.

Some people here have posted pictures using photobucket, which I can't understand, as such photos are liable to disappear at any time, which they have.

If I had a proper diagram of this board, I could probably get to the bottom of this trouble and we all could get more sleep. If somebody has one, could they please post a (hi-rez) copy of it? (Please, not photobucket)

I have worked in Medical Electronics and Digital Imaging for over 30 years and have come across a great many misbegotten pieces of gear that had to be redesigned on the fly, so this thing should be a piece of cake with good documentation.

[Bobdee]

Re: Dell E172fpb

Hello Freddyzdead,
Found your posting very interesting, a Question for you,
You say in your post,

I unplugged the CCFL plugs one at a time and verified that both inverters are indeed working.

I have to ask is that enough to eliminate the Inverters, also you did not eliminate bad connections to tubes or the tubes themselves, as they can be the problem, this posting by Chema, that makes a lot of sense to me, that I would use for elimination purposes, Its what I go by.

https://www.badcaps.net/forum/showpo...&postcount=296

My appologies in advance if I have misunderstood you, but I am novice trying to build the bigger picture.
Bob

[anyaddress]

Re: Dell E172fpb

Ok, the referenced "C5707" means "2SC5707", they are the same. All Japanese transistors start with "2S", so it is common to omit it to save space. Unfortunately, European bipolar transistors all start with "BC", and I have seen cases where the manufacturer had left off the "B" so that a BC337 was marked as "C337" which could have easily been interpreted as 2SC337.

I repeat the warning about the quality of this board and the thinness of the copper tracks. Using a high-temperature iron to try and get the solder to flow better will most likely result in the tracks lifting from the board. Because they are so thin, they will just tear off if this happens. You should do the reverse, i.e. you should use the lowest heat that will do the job. A solder sucker is a must for removing components as this will traumatize the board the least.

Now I will tell you how my adventure is going.

I replaced the 2SC5707s with 2SC495Y transistors that I happened to have a supply of. These didn't even get warm, even though they are only rated at 1 amp collector current. They are small enough to not require any mods to make them fit. The "Y" means that they are the high-gain version, which is what is needed here.

What happened then is that the monitor worked for about ten seconds, then the backlight went out. I unplugged the CCFL plugs one at a time and verified that both inverters are indeed working. Then I tried lifting one end of R750 as suggested above. This resulted in the monitor staying on for about half a minute before the backlight turning off. So, lifting R750 obviously doesn't disable the feedback circuit, which I assume is what is at work here. As if the circuit is detecting that one or more lamps didn't fire.

What I can't find here is a decent circuit diagram of this board; all I can see is one that has been marked up with a modification that I don't think has anything to do with the problem, but the part numbers are illegible and don't improve when it is zoomed up.

Some people here have posted pictures using photobucket, which I can't understand, as such photos are liable to disappear at any time, which they have.

If I had a proper diagram of this board, I could probably get to the bottom of this trouble and we all could get more sleep. If somebody has one, could they please post a (hi-rez) copy of it? (Please, not photobucket)

I have worked in Medical Electronics and Digital Imaging for over 30 years and have come across a great many misbegotten pieces of gear that had to be redesigned on the fly, so this thing should be a piece of cake with good documentation.

working for short time then off!! check my post here:
https://www.badcaps.net/forum/showth...5099#post65099

[mswillia1]

Re: Dell E172fpb

Hi, thanks for your attentions and hereby I send the board of the picture and a detailed picture of the section with the faulty driver transistor.
The TFT is a HP1940 and the board number 48.L1G02.A00

There is a short on Q804 when I use my Digital Multitester. The short is between the upper middle leg and left under leg.

(as my mother language is Dutch I hope you understand my English).

It is on this side of the board where the 2sc5707 and FU9024N also were defect. I have replaced them allready but not tested the board because when I was measuring on the two FU9024N I founded difference. So I examined the board further and discoverd the defective Q804 but .... I do not have further information on Q804.

Regards Frans

I have a similar board, 48.L1G02.A01 and see some shorts in the C5707 chips and the STD10PF. Also a bottom side Diode, can't read it, D803 is shorted. I have not checked the transistors on the bottom side but I will after finding this thread. Did you have any more information on this board that would be helpful to me? I was just going to replace the whole Backlight Powersupply but I am not sure how to tie in the wiring of the inverter to this board.

Has anyone wired in an off the shelf inverter?

It is an HP1940 monitor with a BenQ assembly and a AUO display.

Thanks
Mark

[freddyzdead]

Re: Dell E172fpb

I have found a pair of 2sd1691 transistors as the core voltage regulator on an old motherboard. I have installed these in place of the blown-up 2sc5707s and the monitor has been running for days now, as cool as can be. Before that I had tried many different types all of which overheated and either died or were shut down by the detection circuit. I think these might be a better choice than the 5707, at least as far as robustness goes. It is foolhardy to try to put something like a 3055 or a tip42 in this circuit as they are just not even remotely suitable. One of the most important requirements in this application is a low VCEsat which for the 2sd1691 is very low at about 0.1 volt. Also, the inverter runs at more than 100 kHz which means they have to switch very fast.

Don't be looking for bad capacitors; that isn't the problem. This is not a motherboard. If you change the transistors and they get hot, you haven't fixed it. It will blow up again.

Also, I have attached the circuit diagram for this unit which hasn't been marked up and is readable.


For those who can't get the monitor apart, there are no screws other than those that hold on the foot. There are catches that just snap together. I slide my thumbnail along the crack on the sides and bottom to pop them open, but I have more heavy-duty fingernails than most. A screwdriver will do it, but you are likely to leave mangle-marks unless you are careful.

[freddyzdead]

Re: Dell E172fpb

@Bobdee:

All I wanted to do was check that both inverters are running, as the display will still function with only half the lamps on.

Bad connections to the lamps internal to the panel are extremely unlikely; changing the lamps without damaging anything would be difficult. These lamps have a very long life (> 50,000 hours). When they fail, they usually start up pink or orange and gradually change to white at low brightness.

[freddyzdead]

Re: Dell E172fpb

@Bobdee:

(Why the Administrators allow editing for only 10 minutes? It causes duplicate posts such as this one.)

All I wanted to do was check that both inverters are running, as the display will still function (sort of) with only half the lamps on. Also, swapping around inverter/lamp combinations gives some more diagnostic info.

Bad connections to the lamps internal to the panel are extremely unlikely; changing the lamps without damaging anything would be difficult. These lamps have a very long life (> 50,000 hours). When they fail, they usually start up pink or orange and gradually change to white at low brightness. Dismantling the panel to this extent may be unwise, as Sch72 revealed in his post where he removed all the diffusing sheets and didn't know in which order to replace them. And I liked the lightbulb inside the CRT monitor idea. Very redneck, indeed.

I have seen the post by Chema; I don't see how the unit could survive that much desoldering/resoldering and swapping of parts. I, too, have resoldered all connections on the inverter transformers, even though my practiced eye told me there was nothing wrong there. It would seem that that was what Chema's problem really was, as by that time he had resoldered everything.

Anyway, having a readable circuit diagram was a big help to me. And I really do feel that the 2sd1691 is a better choice than the original 5707.

[EGuevarae]

Re: Dell E172fpb

@Bobdee:

All I wanted to do was check that both inverters are running, as the display will still function with only half the lamps on.

Bad connections to the lamps internal to the panel are extremely unlikely; changing the lamps without damaging anything would be difficult. These lamps have a very long life (> 50,000 hours). When they fail, they usually start up pink or orange and gradually change to white at low brightness.

Not always the case. I have been working with Polaroid TLA-01511C units recently (5 so far, all except one with CCFL problems). It/they startup OK (some of them. Others startup with the pink/red hue and then die ...) and can be all white even for hours (13 hrs is the most I managed to get from them) and suddenly they shut off, with no red, pink, orange, blue, green, purple, magenta, cyan, or you name the color... they just die. The solution? Easy : there are two lamps at the bottom of the units. Ona has a white/blue cables connector, and the other one is white/pink (ironically, the one with the pink cable is the lamp that goes pink). I take two units, get the lamps out, separate them, and put the two blue cable lamps together in one unit. So, this covers two aspects : lamps dying without changing color, and changing the tubes without much complication. And caps are not always the problem, but they are the components that give (sometimes, other they die or kill in silence ...) the best visual indication that they gave up the ghost ... I guess that covers a third topic (the "Don't be looking for bad capacitors; that isn't the problem" line. more than 80% of the repairs I've done so far involved changing caps either as a corrective or as a preventive measure... the other 20% comprehends the "for parts" and the "my tubes are dead" units)

[EGuevarae]

Re: Dell E172fpb

Bad connections to the lamps internal to the panel are extremely unlikely;

Just typing on a Sony LCD that had a loose connection inside the plastic boot in one of the tubes in both the upper and lower pair. I've seen it before on Samsungs (172N.... I hate that model....). It happens.

[EGuevarae]

Re: Dell E172fpb

I have found a pair of 2sd1691 transistors as the core voltage regulator on an old motherboard. I have installed these in place of the blown-up 2sc5707s and the monitor has been running for days now, as cool as can be ........

Davmax, what do you think about this?

[EGuevarae]

Re: Dell E172fpb

I have a similar board, 48.L1G02.A01 and see some shorts in the C5707 chips and the STD10PF. Also a bottom side Diode, can't read it, D803 is shorted. I have not checked the transistors on the bottom side but I will after finding this thread. Did you have any more information on this board that would be helpful to me? I was just going to replace the whole Backlight Powersupply but I am not sure how to tie in the wiring of the inverter to this board.

Has anyone wired in an off the shelf inverter?

It is an HP1940 monitor with a BenQ assembly and a AUO display.

Thanks
Mark

There's a thread of a 1940 here

[freddyzdead]

Re: Dell E172fpb

Just typing on a Sony LCD that had a loose connection inside the plastic boot in one of the tubes in both the upper and lower pair. I've seen it before on Samsungs (172N.... I hate that model....). It happens.

OK, I believe you. But I don't advise an inexperienced person to dismantle one of these panels on the off chance that there might be a bad lamp connection inside. As Sch72 showed when he didn't know how to put back the diffusing sheets, you can brick the thing very easily.

I am wondering how many different brands of panel one might find in these things. Mine has a Chung Hwa panel in it; never heard of them until now.

[EGuevarae]

Re: Dell E172fpb

OK, I believe you. But I don't advise an inexperienced person to dismantle one of these panels on the off chance that there might be a bad lamp connection inside. As Sch72 showed when he didn't know how to put back the diffusing sheets, you can brick the thing very easily.

That's true. In that situation is where a known good lamp comes in handy, but if you just have one unit, it would be very difficult to do .....
As for the diffusing sheets, so far I haven't needed to pull these off all the way, as I always manage to pull out the lamps without having to do it.

I am wondering how many different brands of panel one might find in these things. Mine has a Chung Hwa panel in it; never heard of them until now.

That is also true. In the past, I bought two eMachines with only ONE thing being different among them externally : serial number. Even month/year of manufacture was the same. One had an AU Optronics LCD panel, and the other one a LG/Philips. As for Chung Hwa, I have seen CRT tubes of that brand (lots of them), but so far only one 15" and one 17" panels.

[kevinmorrisone]

Re: Dell E172fpb

Also completed repair with TIP3055 (MJE3055) from Radio Shack. Replaced all four transistors Q739/Q740 Q759/Q760. Originally checked short at 759/760 (in cct) but upon removing Q760, found only it to be blown. Replacing singly did not help, replaced pair Q759/760 and OK. Replaced all 4 for insurance and reflowed all solder on transformers and transistors in inverter circuit. I used a dremel to cut open the area around just the outer two transistors about 15x30mm so no need to insulate. Checked heat after only a few minutes but OK, monitor burned in now for almost 24 hours. No idea what belw it, hopefully it was just cold joint somewhere or aging of the component based on the design but we'll see. I checked Q743/Q751 and all other components per UMD.

Anyway, a friend's monitor so for $7 a good repair and easy too. Great forum and posters, esp. Davmax and UMD_EE both provided great info and UMD pdf is invaluable for anyone attempting this repair, see link below

https://www.badcaps.net/forum/showpo...&postcount=120

[Bobdee]

Re: Dell E172fpb

Hello KMORRIS67738
Congrats on your fix, nice one, but it puzzles me as to why you never took Davmax's advice on the transistors, his recomendations on the 2sc5707's are clear to me, that they are the Best ones for the job, and you can come by them easily,
None of the nonsense of cutting holes in casing
bob

[EGuevarae]

Re: Dell E172fpb

Hello KMORRIS67738
Congrats on your fix, nice one, but it puzzles me as to why you never took Davmax's advice on the transistors, his recomendations on the 2sc5707's are clear to me, that they are the Best ones for the job, and you can come by them easily,
None of the nonsense of cutting holes in casing
bob

Welcome to the forums KMORRIS
I've seen that repair before (I didn't know exactly what substitute was used, but I've seen the cut because the new ones won't fit).
Let's hope one of the in house gurus can bring some light to this matter ...

[gveranda]

Re: Dell E172fpb

Hi Guys,

Ive just fixed FP767 which uses the same inverter board as E172fpb with MJE3055 as on the photo,so circuit board sits nicely inside the metal cover never had to cut holes or anything.

[Bobdee]

Re: Dell E172fpb

Nice one Gveranda,
Can you tell us what make monitor it is, is it another Dell or BENQ ???
it will be helpful for future cross reference to others,
also why you never replaced 2sc5707 with same ???
as they are easily got from ebay
bob

[Wizard]

Re: Dell E172fpb

Interesting that 3055's still made after 30+ years!

It begun with this TO-3 package. Good old massive copper plate.

http://images.google.ca/images?hl=en...=Search+images

Cheers, Wizard