Dell E172fpb

[budm]

Re: Dell E172fpb

Did you recheck those transistors again to see if they are shorted out again and the tabs are not touching the chassis?
You can also remove the BEL FUSE which feed the B+ to the inverter circuits and see if the LED stays green with PC connected. Also check the value and leakage resistance of those 4 tuning caps by the 4 inverter transistors.

[veeone]

Re: Dell E172fpb

As i'm new to electronics which is the startup cap?

http://s1063.photobucket.com/albums/...rent=DV172.jpg

[retiredcaps]

Re: Dell E172fpb

a very slight timp noise coming from the board

If you mean "tick" noise, check the DC voltage across the startup cap. It can go bad without bloating.

[veeone]

Re: Dell E172fpb

Yup did the transformer legs and capacitors are original but no signs of overheating or bulgeing.
Green light just flashes never get the amber light unless you switch pc off.
Will double check all soldering next. Will try the hair dryer trick as well.vee

[jetadm123]

Re: Dell E172fpb

Hmm got different symptoms to others here. Turn my pc on and just get green light flashing and a very slight timp noise coming from the board but cannot identify where it is coming from.
Leave this way for approx 20 minutes and finally the monitor will come to life normally so seems to be something to do with getting warmed up......
Replaced:
(4) 2SC5707 Transistors (2) FU9024N Transistors (IRFR9024N) J598
(1) BEL 125V 3 Amp MS Fuse
No change in monitor performance.......
Resoldered many joins and still no change.
ANy ideas?? Cheers Vee

I take it you resoldered all the transformer joints?

Are the capacitors original? Regardless, try heating them with a hair dryer to see if that reduces/eliminates your 20 min wait time.

[veeone]

Re: Dell E172fpb

Hmm got different symptoms to others here. Turn my pc on and just get green light flashing and a very slight timp noise coming from the board but cannot identify where it is coming from.
Leave this way for approx 20 minutes and finally the monitor will come to life normally so seems to be something to do with getting warmed up......
Replaced:
(4) 2SC5707 Transistors (2) FU9024N Transistors (IRFR9024N) J598
(1) BEL 125V 3 Amp MS Fuse
No change in monitor performance.......
Resoldered many joins and still no change.
ANy ideas?? Cheers Vee

[budm]

Re: Dell E172fpb

That section of the chassis is slanted, now it makes sense why it stops working, I was surprised it did not blow the inverter fuse, but I am glad you figure it out, good job.

[tmwalsh]

Re: Dell E172fpb

In my case, I found that there were two places where grounds could have been connected by the case. Both were from the SMPS, one where AC line was coming in, the other further on.
Given that, after isolating those grounds, it still didn't work. Made no sense. There must be something that changes when the boards are mounted to the case.
It was one of the new C5707 hitting the shield apparently as it works now, displaying these words. I made sure to tilt it away from the slanted case, and it must have cleared as it is working fine.
Thanks for all the input and encouragement.
tom

[tmwalsh]

Re: Dell E172fpb

I think I understand now. I can check for long legs by keeping the board, mounted to the frame, insulated by cardboard as it sits on the backside of the display panel.

There are no non-ground screws. The two spots that could qualify for that are held in place with plastic posts that spread, and must be pinched to make them let go of the board.
The large ground plane is connected directly to the display logic board with four of the ten pins. The logic board is in turn connected to the frame with bared trace on both sides of the board at the two places retaining screws are used.
I can also test that it is not related at all to the CCFLs by insulating the frame from the panel completely in addition to trying it with the frame 'touching' but with all component legs insulated by cardboard. I have been staring at the schematic which appears to be very similar if not identical except for the sound parts, trying to figure out what would have failed and provided an extra load the power supply doesn't like. When it is out of the frame, it runs fine, and has no hot spots at all. I left it fired up in the horizontal, with no convection flow, and checked temps with a non-contact. The hottest temp on the board was 110F.
So two tests to run.

Is my second picture useless as well? I took it upstairs into the sunlight, and tried to get the image to fill the frame. Any hints?
tom

Results of insulating the trace side of the circuit board with cardboard: same, blinking power light and no CCFL fire.
Results of insulating the frame from the backside of the display panel, same, blinking power light, no CCFL fire.
Ugh. That means trying to figure out which bits are dysfunctional. I think I will stare at the schematic and attempt to 'isolate' one part of the board at a time if possible. I don't think the logic board would cause this, but it does diddle with the brightness control, I think.
tom

[budm]

Re: Dell E172fpb

"Budm is following what I said earlier that a long leg may be touching and the cardboard would prevent that and rule it out as a problem- I do not think he was advocating total isolation of the board from the frame" that is correct.

[selldoor]

Re: Dell E172fpb

Oh boy. One suggests to only use holes where a ground connection is guaranteed{with rings}
The other suggests using cardboard to insulate the board from the mountings.

Well I could say at least we can provide decent pictures! but I'll just think it out loud.
I read it that we are suggesting different ideas- not opposing ideas.
Budm is following what I said earlier that a long leg may be touching and the cardboard would prevent that and rule it out as a problem- I do not think he was advocating total isolation of the board from the frame.
I was thinking perhaps the non ground screws had worn away some insulation and were grounding traces that should not be grounded.
It is unchartered territory and most people prefer some logical suggestions
rather than silence.
I happen to like your idea of isolating with plastic washers but dont know
if it would cause a problem - how about trying washers on all but one then
if it didnt work try on all just to see.

[tmwalsh]

Re: Dell E172fpb

selldoor said: Try mounting it in the frame with only screws in the holes with solder rings.

budm said: Just to see if it is the component's leads may be touching, put the card board under the boards than moun the boards back in, if it still goes dead, then more likely to be bad connections/solder joints, if it works with card board underneath, then you have short circuits caused by the component leads some where.

Oh boy. One suggests to only use holes where a ground connection is guaranteed{with rings}
The other suggests using cardboard to insulate the board from the mountings.


What to do? The component side of the board faces 'into' the cavity of the case, with the printed circuit side facing towards the lcd panel. The component side of the board where the mounting screw holes are located is non-conducting where the board would rest on the case supports. The 'trace' side of the board has several metallic 'pads' or rings to connect the ground plane of the board to the case using screws with built-in 'washer heads' that would definitely contact the metallic pads. I was going to insulate the screws from the board by using snippets of plastic milk carton as insulation. That would leave the board disconnected electrically from the case. I am betting it would work, but know that something somewhere is broken, shorted, or ???.

Is it safe to leave the board un-connected to the perforated metal case/shield?

rtech: I will re-flow the solder on the large blue ballast caps. Can't hurt. But I don't understand how that would affect the operation of the circuit in or out of the metal case. If the solder connection was bad, then the cap couldn't function, and I don't know how the board is able to work.

[Rtech]

Re: Dell E172fpb

There are four BLUE ballast caps near to the Inverter transformers.I have a a couple of cases where these also need resoldering, even if they look ok.The first time it happened to me,I was amazed as I thought they all looked OK.

[budm]

Re: Dell E172fpb

Just to see if it is the component's leads may be touching, put the card board under the boards than moun the boards back in, if it still goes dead, then more likely to be bad connections/solder joints, if it works with card board underneath, then you have short circuits caused by the component leads some where.

[selldoor]

Re: Dell E172fpb

Pics are not really big/clear enough for troubleshooting however looking at another pic I see that some of the holes have a solder ring and other dont. Try mounting it in the frame with only screws in the holes with solder rings.

[tmwalsh]

Re: Dell E172fpb

Two pics attached?
Preview shows them, we'll see, maybe.

I'm sure I can do better, but don't have a clue as to what they'll turn out when I take them. Any hints?

There are no tall caps, all are as when I opened it up. Perhaps, likely, moved to a different spot on the board, as I removed and tested all. The 'keys' board is in a bezel that surrounds the screen, and was insulated, etc, when testing. Once I put the two boards in the case, and attach screws to hold the boards in place, it quits working. I don't think it is a torque situation where the board is twisted or loaded funny. The CCFLs will not light up when the boards are mounted in the outer metal shell with the perforated panels, and the power switch green led flashes constantly until the power switch is held to depower the board.
These two pairs of lamps ride in a U shaped metal channel that then slides into openings at the bottom and top of the display assembly. They operate fine when waving in the breeze, and also when they are inserted into the display cavity. Well, I think fine, as I have the screen face down, and can see the 'floating rainbow' moving around. Cannot really observe the brightness as I can't get a direct angle, but they sure burn bright. Testing that was my 1st thing after finding it didn't work when assembled. Laid cardboard on the back of the display, connected the loose CCFLs, connected logic board, key button board, and power. Push button, lights after ~2 seconds. Install CCFLs and do the same test, lights work. I have not tried having the case separated from the panel by an insulator, and think it wouldn't make a difference. I think there is a component that is loading the inverter when the logic board and the power/inverter board are connected to the case.
The AC ground pin on the board does not connect to board itself, so it is not a power ground situation. I *think* the 'cold side?' ground is isolated from the switching power supply ground, but may get connected when the board is put into the case. I haven't really wrapped my mind about that except in the case of a TV chassis with a switch mode power supply. That I can visualize and understand. This thing, staring at the PDF file I downloaded, Benq_Inverter_Schematic.pdf, is confusing. In that schematic, the 2.2uf50V near the lower right is the only thing I see that is connecting to 'ground' directly other than the C5707 transistors.
May be I am getting too old to do this stuff any more.
The last thing I was thinking about was insulating the mounting screws from the ground pads on the inverter board, so that it would not connect through the screws to anything as the other side of the board is bare insulate. I was fearful of some sort of floating ground sending power backwards through something and loosening the magic smoke to the world, so am asking if that is a valid test?
tom

[selldoor]

Re: Dell E172fpb

As it all worked out of the case but not in, I should concentrate on what has changed.
I dont think we have any pictures of your boards - that might help us spot something.
Any extra long legs on any components that might be touching something, any tall caps that might be touching something. Is the "Keys" board fitting in ok back not touching anything. All connectors firmly connected and a working pc source connected?
Can the lamps or wiring be arcing out anywhere to the case.

[tmwalsh]

Re: Dell E172fpb

I tested the transistors using a transistor test meter that I built from a kit. You hook to BCE, and press the 'test' button. A led will light up indicating "OK". It has a pot to adjust current flow, and 3 of the 4 lit the led brightly, the 4th lit the led dimly. I was assuming that its gain was not on a par with the others. The led was barely lit as I remember. It is an Elenco Electronics DT-100k transistor & diode tester.
RTECH, I flowed all the transformer pins on the board, and also where the windings are soldered to the other end of the pins. From the descriptions I read, I wasn't positive that it was the pins that needed re-flow, as they all looked good. I had previously looked at the coils you mentioned, as I have had experience with that style coil, rather large windings, kind of 'coarse', in a Sony TV, that broke a winding(!), so I paid a bit more attention than normal.
This came from my brother, and the shield being 'loose in the box' when I got it made me think someone had been into the monitor before me. He stated he didn't have time nor tools to venture into the set. Given that, I wonder if all the solder had been re-flowed already. It just does not look like a low quality solder job, anywhere.
I have been thinking that one of the caps may be not so good, perhaps one of the 2.2uf on the transformers. It seems to be the only one that has a 'direct' connection to ground, and could mess up the load on the hv transformers. I think.
tom

[retiredcaps]

Re: Dell E172fpb

there are 2 inductors on these boards one associated with each pair of transistors, these are marked Lx, these have quite thick wire. and would resolder them as well, as it sound like, you have a intermittent track or dry joint,and when tightened down this could cause your problem.

For any of these Benq type boards with c5707s, I end up almost reflowing the solder on the entire board just to be safe. I neglected to reflow the inverter transformers in my first every c5707 repair and paid the price with a blown picofuse 24 hours later.

[Rtech]

Re: Dell E172fpb

there are 2 inductors on these boards one associated with each pair of transistors, these are marked Lx, these have quite thick wire. and would resolder them as well, as it sound like, you have a intermittent track or dry joint,and when tightened down this could cause your problem.