Samsung 2693HM - the common "2 seconds to dark" problem

[kenikh]

I got my hands on a Samsung 2693HM after the previous owner tossed it. I plugged it in and get the "2 seconds then dark" problem that seems to plague this monitor's smaller cousin, the 2493HM. I can't find any previous references of a 2693HM owner having the same problem, so appear to be treading in new waters.

Repro steps:

• Connect monitor to computer
• Turn on - blue light stays lit
• Desktop shows up for a few seconds then goes black
• I can see images on the screen after it goes dark with a bright light shining on the screen
• Turning the monitor on/off consistently lights up the screen for the same short period, then goes black

I've read as many 2493HM threads as I can find and there doesn't seem to be a definitive "gotcha" problem. From what I've read, it seems that the solid blue led indicates that the power board and logic board are likely to be working okay. That said, since this monitor's boards are significantly different enough from the 2493HM, I figured it warranted a new thread.

First, no bulged caps anywhere - I understand this isn't necessarily the only indicator of failure. I have a multimeter at home, so can bench test if I need to, but really dislike messing around in electronics with the power connected. I'll go there if I have to...

On the caps, the caps on the power board are the typical crappy Sam Youngs. Interestinly, the caps on the inverter board are signifcantly higher quality Rubycon ZL 220uF 35v and compared to the 24" monitor, has four as opposed to two caps (the 24" has 470v caps, BTW).

Given the symptoms, the assumption is that the inverter board caps have gone dry, but given that these are forum recommended replacement quality caps, I am less certain. I guess anything can fail.

I have also heard that the Slo-Blo fuses can cause issues, as well. I could buy new fuses or just jumper them to test, although not my first choice.

Question: Would (those in the know) just replace the 4 caps on the inverter board without testing or go through the testing battery first?

Pics below.

Thanks!

[jetadm123]

Re: Samsung 2693HM - the common "2 seconds to dark" problem

Rubycon Zl caps are pretty good caps and there's less chance of them failing than some of the other crap brands that have been used for this type of application. You can try performing some basic testing before replacing the caps. On the inverter, there appear to be two surface mount fuses right above the wiring. Have you checked them for continuity? Since the caps are rated at 35V, then the input voltage is probably 24V. Check for this voltage before and after the fuse(s) to verify the 24V is getting to the inverter and that the fuse is good. Also, check the 4 mosfets ( I read one as APM4052D) between each cap for shorts.

[kenikh]

Re: Samsung 2693HM - the common "2 seconds to dark" problem

Thanks for the advice - I'll give it a shot.

[PlainBill]

Re: Samsung 2693HM - the common "2 seconds to dark" problem

One other thing to test. Try checking the resistance between the two pins of each sockets (the part mounted on the PC board) for the CCFL leads. IF these are direct wired you will be checking the secondary of the inverter transformers.

PlainBill

[kenikh]

Re: Samsung 2693HM - the common "2 seconds to dark" problem

One other thing to test. Try checking the resistance between the two pins of each sockets (the part mounted on the PC board) for the CCFL leads. IF these are direct wired you will be checking the secondary of the inverter transformers.

PlainBill

I assume that you mean these:

[kenikh]

Re: Samsung 2693HM - the common "2 seconds to dark" problem

FWIW, I got a really good deal on five United Chemicon KY caps in the right spec ($3 for all 5). Would have bought them here, but there were none available in 35V. I'll test the screen and if all points to the caps, I'll have them in time for replacement.

[PlainBill]

Re: Samsung 2693HM - the common "2 seconds to dark" problem

I assume that you mean these:

Yes. I would have to see a close-up picture of the inverter to be sure. After looking at the pictures of the inverter transformers, You should be able to probe directly to the points where the transformer is soldered to the board. One of the reasons for 'two seconds to black' is an open or shorted transformer. Resistances should match to within 3% for all secondaries.

PlainBill

[kenikh]

Re: Samsung 2693HM - the common "2 seconds to dark" problem

I decided to execute a little practice on something more disposable before trying this monitor. My success is directly attributable to you guys giving me the confidence to try this out:

https://www.badcaps.net/forum/showthread.php?t=15949

Thanks!

I'll be crackin on this one soon, which I consider the next level as it will require using a multimeter and live current to test where the fault is.

[kenikh]

Re: Samsung 2693HM - the common "2 seconds to dark" problem

Finally got around to playing with this monitor last night. Th power board was full of Sam Young caps, so I went ahead and replaced them all with Nichicons, without testing. I figure it's insurance if not th eproblem...which it wasn't.

Fired the monitor back up and the same symptoms persisted. I started testing the inverter board, unplugging all leads to the CCFLs, then plugging them in one at a time.

Turns out the bank of three on the left happily fired up the CCFL for the usual 2 seconds, the bank on the right, no light from CCFL. I'll have to trace back the circuits, but figure that there is sort of binary split in the power routing and the issue is likle in a cap or fet on that side.

Thoughts?

[PlainBill]

Re: Samsung 2693HM - the common "2 seconds to dark" problem

Finally got around to playing with this monitor last night. Th power board was full of Sam Young caps, so I went ahead and replaced them all with Nichicons, without testing. I figure it's insurance if not th eproblem...which it wasn't.

Fired the monitor back up and the same symptoms persisted. I s tarted testing the inverter board, unplugging all leads to the CCFLs, then plugging them in one at a time.

Turns out the bank of three on the left happily fired up the CCFL for the usual 2 seconds, the bank on the right, no light from CCFL. I'll have to trace back the circuits, but figure that there is sort of binary split in the power routing and the issue is likle in a cap or fet on that side.

Thoughts?

1. For various reasons, it is preferred that pictures be hosted on this site. That way if someone looks at this thread in a year they wont get the 'this image is no longer available' graphic.

2. PLEASE don't post pictures inline. Use 'Manage attachments' - below the text entry area.

3. Have you checked the two fuses on the inverter? If one is open, that could cause this problem. Your idea of checking the FETs is a good one.

PlainBill

[kenikh]

Re: Samsung 2693HM - the common "2 seconds to dark" problem

Woops - forgot about that policy. Fixed.

Haven't checked anything on the inverter board beyond what I mentioned. Was planning on looking further today. I will assume that the below items are the fuses you are referring to?

[RJARRRPCGP]

Re: Samsung 2693HM - the common "2 seconds to dark" problem

1. For various reasons, it is preferred that pictures be hosted on this site. That way if someone looks at this thread in a year they wont get the 'this image is no longer available' graphic.

QFT. And *never* use Photobucket!

[PlainBill]

Re: Samsung 2693HM - the common "2 seconds to dark" problem

Woops - forgot about that policy. Fixed.

Haven't checked anything on the inverter board beyond what I mentioned. Was planning on looking further today. I will assume that the below items are the fuses you are referring to?

Got it on the first try!!!

It bears repeating - always assume a fuse blew for a reason. Some people prefer to insert a jumper and watch a hole burn in the board, but that practice is not recommended.

PlainBill

[kenikh]

Re: Samsung 2693HM - the common "2 seconds to dark" problem

Update: I'll keep noodling on the inverter for grins, but I found a replacement, new inverter board for $20. $20 is definitely on the good side of the time to money ratio.

[kenikh]

Re: Samsung 2693HM - the common "2 seconds to dark" problem

PlainBill guessed right: we have a current issue at the fuses (see below)

I've never played with a PCB like this one before. With no pins thru the board, all solder connections seem to terminate on the surface and the board seems heat sensitive to hot solder (it bubbles). Is there anything I should know before digging in further?

[PlainBill]

Re: Samsung 2693HM - the common "2 seconds to dark" problem

PlainBill guessed right: we have a current issue at the fuses (see below)

I've never played with a PCB like this one before. With no pins thru the board, all solder connections seem to terminate on the surface and the board seems heat sensitive to hot solder (it bubbles). Is there anything I should know before digging in further?

This is surface mount technology. Slightly different techniques are needed, but the differences are minor. Removing the components is easy with ChipQuik.

Since you have an inexpensive replacement available, it probably wouldn't be practical to spend too much money on this, but there are several possible tests you can perform. The first is simple - measure the resistance between the upper end of the open fuse and ground. If the resistance is less than 10 ohms, there is a short.

The next test is slightly more complex. The driver ICs appear to be in a 5 pin package. I'd check for a short between any of the 5 pins on all of the packages. If any pins are shorted on one IC, you have identified the problem.

The last test requires bypassing the fuse. I prefer to use a 1157 type automotive taillight to limit the current. If the lamp glows only dimly, it is safe to pit in a replacement fuse. If it glows at full brightness, there is still a problem.

PlainBill

[kenikh]

Re: Samsung 2693HM - the common "2 seconds to dark" problem

I like to learn new things, so may make the investment in a Chip Quik. It's pretty cool!

[kenikh]

Re: Samsung 2693HM - the common "2 seconds to dark" problem

Inverter board arrived the other day, so I slapped it in...success! The upshot is that my power supply now has a full set of brand new Nichicon caps, too. The monitor works great.

Most likely won't go ahead with the Chip Quik and try to fix the old inverter board. Too many other things on my plate. Thanks again for all of the help - my computer thanks you, as do I!

[Davodevo]

Re: Samsung 2693HM - the common "2 seconds to dark" problem

Hi Newby here from New Zealand. I see this is an old post but was hoping like hell there is still a place where I can get the inverter board for $20. I have this same issue with this model. Local Samsung parts want to sell me new panel for $672....