Dell E172fpb

[GLORK]

Re: Dell E172fpb

Final summation: FIXED

i too had this DELL E172fpb with BenQ board inside.

triple THANKS to all who contributed to the 24 pages of input, i read most of it, i think that took longer than the actual repair. Surprisingly not a "bad-caps" issue ! (they all looked and tested fine)

i found the usual suspects 2SC5707 transistors near the output (backlight - driving) transformers, replaced the pair, and reflow'd all the solder in these 2 sections. Its been on for an hour and YES... (cue music) I FEEL GOOD, like i know that i should !

GLORK

[LLLlllou]

Re: Dell E172fpb

I must add - the transistors in question get BLOODY HOT - hotter than I would have thought would be usual, even given the large number of watts they are designed to handle. I've thought about adding a small fan etc - but that really begs the question - WHY ON EARTH should you need to go to these lengths to keep it cool? I reckon Benq has simply done a pretty shoddy job when it comes to design and reliability testing on this model, and I hope they've ironed the problems out now for their newer models.

Doc

The reason why the transistors run so hot is because the MJE3055 is not even close to an acceptable replacement for the 2SC5707 in this application.

Although the MJE3055 meets or exceeds Voltage, collector current, and power dissipation ratings, it is sorely lacking in gain bandwidth product, and in gain (hFE).

The 2SC5707 has a gain of 200 to 560, while the the MJE3055 only has a gain of 20 to 100.

The 2SC5707 has a gain bandwidth product of 330Mhz, while the MJE3055 is only 2MHz.

In short, this means the MJE3055 would need more current driven into the base to turn on fully, and would take much longer to turn on even if it was getting enough current. Either would mean the transistor would spend much more time in a higher resistance range somewhere between on and off, not where you want to be if you're the switch in a SMPS. Unlike a "close to zero ohm" state, which would cause very little power dissipation, they're spending WAY too much time "getting on" and "getting off" and cooking themselves in the process.

The 2SC5707 is a VERY specialized beast, not many transistors can match it for gain at high current, and when you add a 330MHz GBP, it pretty much stands alone.

[LLLlllou]

Re: Dell E172fpb

Let The Truth Be Known!! My Wife Took My Screen To A Repair Shop Whilst I Was Away On Business!! That Explains It All!!!!!!! It Didn't Make Sense Why Some Of The Components Are Different But The Circuits Are The Same! Now I Need To Find A Pair Of Fu9024n's Somewhere Online!! Q805 And Q812 Are Both Fed After The Diodes D801 And D803.

Damn women!

[Jack Crow]

Re: Dell E172fpb

Hi all,
An amazing amount of data on this unit.

Long story made short.
I got one of these dead from a client who did not simply want to toss it out.

So it's a learning experience that won't cost much.

I have considerable experience in CRT monitors, but this is my very first flat screen.

Looked it up on the web and found a quick fix article that seemed to apply to this monitor.
One of the four transistors was short city.

Since Im broke and did not want to wait for a mail order, I subed in four TIP31's.

Had to remove some metal work to make it all fit.
Remember this is a learning tool, not a real client repair.
So if I kill it no loss.

The big caps looked ok, the 1uF's were garbage but did not change them as of yet.

Anyhow, it's been running for 4.5 hours so far on the test pattern generator.
That smell when electronics has been over taxed, I am NOT getting.
This monitor is generating a normal amount of warm.
No hot spots over where the transistors went in.

The second biggest problem was removing the plastic case. What a pain.
Need to make some tools for that.

If it's still running on saturday morning, I will call it done.

Feel free to comment, I learn from this kind of thing.

Be well all.
Jack Crow

[retiredcaps]

Re: Dell E172fpb

The second biggest problem was removing the plastic case. What a pain.
Need to make some tools for that.

I just use plain old butter knives.

Feel free to comment, I learn from this kind of thing.

For Benq boards with c5707s, it is mandatory to reflow the inverter transformer pins and all the other points that people have mentioned.

[Jack Crow]

Re: Dell E172fpb

RC,
That is clever. Was thinking of tools that guys who fix dry wall use but a bit more flex.

Thanks for the tip.
The unit is still cranking next to the elbow.

I am now kicking my self for not taking some photos.

If I crack it open again, I will take some pix.

So let's see how this shakes out.

Jack Crow

[retiredcaps]

Re: Dell E172fpb

RC,
That is clever. Was thinking of tools that guys who fix dry wall use but a bit more flex.

Putty knives will work too. I think PlainBill uses guitar picks, some people have carved out wooden tools, and there are iphone pry tools on ebay for 99 cents + free shipping that might work too.

For my first monitor repair, I tried those annoying junk mail credit cards. Terrible. Then I switched to butter knives and had more success. Once you get the hang of it and know how much pressure to apply, it is pretty quick going.

It is almost certain that you will break at least one of the tabs opening up these snap-on monitors. I think I might have opened up one without breaking a tab.

[jetadm123]

Re: Dell E172fpb

I think there was some lengthy thread on another site where someone was using a TIP3055/MJE3055 transistor as a substitute for the 2sc5707. As mentioned earlier on this thread, the hfe was not satisfactory and the 3055 would overheat and short out. Nice to know that the TIP31 seems to be working and is available at Radio Shack. It's a T-220 case, which still makes it difficult to fit in without doing some mods to the metal enclosure. Yes, a photo of your mod would have been nice.

As for opening the enclosure, I used a stiff putty knife with a beveled edge to pry open a corner enough to fit in an old gift or credit card (preferred because they are thicker plastic). As retiredcaps mentioned, those free plastic cards are too thin and bend too easily. However, I have found that if you fold them in half, it makes them stiffer and they're good for a one time use. The butterknife will work, but make sure you take it easy when you get near the pushbutton board because the wiring for that board and the board itself are easily damaged.... speaking from experience.

[pedro]

Re: Dell E172fpb

One of the four transistors was short city.

Since Im broke and did not want to wait for a mail order, I subed in four TIP31's.

Mike, check the specs of the 5707. As others may point out, the 2SC5707 has a high Hfe and decent switching speed, and generally TIP31 or (shudder) x3055 type substitutions don't last all that long.

Your repair does prove that the transistor was the failed part, but unless you want to revisit in reasonably short order I would recommend always using the 2SC5707 in these repairs.

[PlainBill]

Re: Dell E172fpb

As retiredcaps said, I use guitar picks. You can get a dozen for about $5 at wally world. That way you can leave them in place so the case doesn't latch together as you are splitting it somewhere else.

One thing to be careful of is making substitutions for components. If you aren't sure of what parameters are critical you can waste a lot of time. In the case of this monitor the failure was not due to a part that was not capable of handling the load, it was due to bad solder joints on the transformer. The correct parts are available quite cheaply on eBay.

PlainBill

[Jack Crow]

Re: Dell E172fpb

All,
I know subing in parts is not always an ideal solution.

On the other hand, if the OEM units were any good this monitor would not be here.

We see so much stuff that is designed to fail after the warrenty expires.

Anyhow, the unit is still cranking after 20 hours and has not failed yet.

Most likely I won't get back to it until some time in the late afternoon.

Do remember this one is learning tool for me.
I do not usually work on flat screens.
Serial number one if you get my drift.
So expect errors.
Fortunatly they don't cost much.

Lets see what happens.

If I crack it open again, this time with better tools, I will photograph the parts transplant and the metal work cuts I made.

It's warm, the non contact IR thermometer says 107 at the hottest spot.

It's a few degrees hotter than this HP display in my computer rack. (95)

Keep it safe
Jack Crow

[PlainBill]

Re: Dell E172fpb

All,
I know subing in parts is not always an ideal solution.

On the other hand, if the OEM units were any good this monitor would not be here.


Keep it safe
Jack Crow

That is the fallacy that results in botched repairs. The most common cause of the failure on this monitor was bad solder joints on the transformers. The timing caps (gray 'boxes' between the transistors) also sometimes fail and cause it.

I just did the economics of your repair. The proper 2SC5707 transistors are available on eBay. The beat deal was 20 for $4.00 with free shipping. No driving to the store, spending $3.38 + tax for the wrong parts, no cutting, no listening to someone rant about sloppy technique.

PlainBill

[CaptHwdy]

Re: Dell E172fpb

on mine the screen will flash a sec then the green power light will blink green and the screen will stay black. is this the same as other's?

[PlainBill]

Re: Dell E172fpb

on mine the screen will flash a sec then the green power light will blink green and the screen will stay black. is this the same as other's?

It could be either shorted transistors or a broken CCFL or wiring to the CCFL.

PlainBill

[tmwalsh]

Re: Dell E172fpb

I can't believe I read the whole thing. But I did. My brother sent me his E172FPb, which died unexpectedly one morning.
It has an almost invisible quick flash, a blink if you will, of the CCFL, and the on/off button was flashing.
I confidently replaced the C5707's, and flipped the board expecting to see ring around the transformer leads. No such luck. I read the 'trouble shooting' guide, and printed out a copy of version 3, IIRC. The fuse checked ok, and the two 9024's when measured in circuit showed nothing when checked, reading OL in the display for every combination of lead checking.
I removed all electrolytics, and tested ESR using a Blue. All were better than specified by the resistance table for capacitance and voltage. I decided to replace the 9024's and did so. I laid the gizzards flat on cardboard, and connected the CCFL's to the outputs, and held my breath and plugged it in. The CCFL's lit up as if they were happy to be alive. I turned the machine off using the button, and then re-powered it and they came on again. Viola, all is well.
NOT.
Put the bulbs back in the display, re-connected everything, put all the covers on, connected to a PC, connected to power, and now I get a quick flash when I press the power button, and then back to black.
Waaaahhhhh. What now? I've taken the covers off, and am going to look for a connection that gets made/broken when the unit is assembled, or for contact with the metal case, or ..? First test will be to see if things get hot, and to see if it will light up with the bulbs again removed from the display, and laid out on cardboard.
Any suggestions? I've burned my finger on the solder-sucker soldering iron enough times that I don't want to go through that any more. Where do I look?
tom

[jetadm123]

Re: Dell E172fpb

You say you looked at the transformer leads, but did you actually resolder them?

Also, try reseating all the connectors.

[budm]

Re: Dell E172fpb

Also check the tuning caps (0.22~.33uF 250V) by the inverter transformers, if the tunung caps are bad or off value, they will take out your new transistors. Please post the pictures of the board.

[tom66]

Re: Dell E172fpb

Service manual for 173FP - maybe same?

http://www.tgohome.com/ServiceManual...&Search=Submit

[tmwalsh]

Re: Dell E172fpb

I re-soldered all the leads on the solder side even though they looked fine. Given that, I also re-flowed all the leads on the transformers where the windings terminated.
I don't have a cap tester, but did check ESR and for a tiny reading for a flash of time on a DMM, and slight deflection on an analog VOM. All seemed good.
Everything is now out of the case, and lying on cardboard. Of course, it lit right up, and is glowing even as I type. Moving my hand over the whole megilla, feeling for hot spots, nothing is particularly hot. The smoke has not escaped as of yet...
The ten pin connector between the power/inverter board and the logic board is not tight, but doesn't complain when wiggled. pics to follow if I can figure out how.
tom

p.s. Of the 4 c507's, one seemed weak, the others OK, fuse OK. The sheet metal slide cover was not installed as I found it. It shields the wires from the inverter to the CCFLs. Could it be causing the inverter to balk?

[budm]

Re: Dell E172fpb

http://s807.photobucket.com/albums/y...t=DSC02365.jpg

http://s807.photobucket.com/albums/y...t=DSC02355.jpg

Did you check those 4 caps (0.15uF)?