crwdns2933423:0crwdne2933423:0

Repair guides for displays (or monitors) for computers or other devices with video output.

crwdns2934591:0501crwdne2934591:0 crwdns2934593:0crwdne2934593:0

Dell U2413 turning off

I recently bought a used Dell U2413 monitor. All were working fine within my initial use, though a few days after, the monitor started to turn itself off randomly within half an hour to an hour of use-- no power led, and would turn back on after a couple of minutes. After some troubleshooting, it appears that the power supply, or something within the power supply circuit is overheating, thus causing the monitor to turn off.

I'd like to know what specific components to look into for this kind of problem?

Thank you!

edit:

Currently tinkering with the board and I'm looking at two mosfets that might be overheating. I checked with the multimeter and they're both around .450-.470v.

K15A60U (with its own heatsink)

LD7904jgp7 (no heatsink, and near the K15A60U's heatsink)

Still not sure though which one needs to be replaced.

crwdns2934081:0crwdne2934081:0 crwdns2934083:0crwdne2934083:0 crwdns2934093:0crwdne2934093:0

crwdns2934109:0crwdne2934109:0

crwdns2889612:0crwdne2889612:0 8
crwdns2944067:012crwdne2944067:0

Hi @dancarlocanlas ,

Have you checked the capacitors in the power supply section? Just in case they are bulging etc.

Just a thought.

crwdns2934271:0crwdnd2934271:0crwdne2934271:0

yep. all capacitors and resistors seem to be fine -- no bulging, no burned parts, etc. I also noticed that if I provide ample air blowing at the back of the monitor, the issue doesn't persist. I do realize, this could work, but it might not be an ideal long-term solution.

crwdns2934271:0crwdnd2934271:0crwdne2934271:0

Hi @dancarlocanlas ,

Can you replace / renew the thermal paste on the heatsink that you mentioned?

Cheers.

crwdns2934271:0crwdnd2934271:0crwdne2934271:0

That's probably my next step, though I don't have an idea about thermal paste for electronic components. I only know about them for computer cpus. Do they have specific types for each application, or any thermal paste will do? Also, I'm looking at sticking a heatsink on top of the LD7904 chip (DIP), would that be advisable?

crwdns2934271:0crwdnd2934271:0crwdne2934271:0

Hi @dancarlocanlas ,

My untrained guess is that any new paste will be better than dried old paste.

As to putting on a heatsink. It may prolong the life of the component but if it is running hotter than it should due to whatever reason then it will fail eventually anyway, but definitely worth a try.

crwdns2934271:0crwdnd2934271:0crwdne2934271:0

crwdns2934275:07crwdne2934275:0

crwdns2934285:0crwdne2934285:0

crwdns2933315:05crwdne2933315:0

crwdns2934051:0crwdne2934051:0

After some time of trouble shooting i found that the shut down issue was related to a crow bar circuit in a voltage control circuit, this circuit is used to set the output voltage of the 5.3v supply. With the out put at pins 3 and 4 using the red ground wire a pin one. There is a sensitive gate SCR Q971 PN. MCR101L http://www.unisonic.com.tw/datasheet/MCR... ( near the top left corner of the pcb) that was turning on when heat was applied to it, some place around 150 deg. F. There is also another SCR Q972 that is used for the 24 / 11.9 supply that may cause the same issue, Q972 is near the lower center of the pcb, near a heat sink. i found a similar part at digi-key PN BT169-L, there is an issue and the -parts leads are not in the correct order, with some care one can rearrange them to work.

As also ways i will not take any responsibility for the repair, or any damage to you or the equipment, (KMA)

I have am having success with my fix , happy fixing

FBG

Hope this helps,

crwdns2934105:0crwdne2934105:0

crwdns2889612:0crwdne2889612:0 8

crwdns2944067:03crwdne2944067:0:

Hi Floyd. Thanks for your hint. I tried the thermal paste and radiator fix first but the problem still persists and the monitor shuts off.

Now for better understanding: You fixed it by completely replacing the electronic parts with a new one (MCR101)? And the three metal pins (part leads?) have to be rearranged in what exact order then?

I want to solder them in but don't want to risc to short the whole pcb.

Thanks for any suggestion!

Chris

crwdns2934271:0crwdnd2934271:0crwdne2934271:0

that works for me :) thank you ^_^

I used BT169D in place of Q971 and Q972.

just you need to bend some pins. BT169D has pin assignment as KGA that should be turned into GAK.

change the order and solder it.

wish you luck :)

crwdns2934271:0crwdnd2934271:0crwdne2934271:0

this worked for me for a few months but then the issue came back. Any ideas if there is something I maybe overlooked ?

Q971 and Q972, I modified those.

Followed this guide

https://niallbest.com/dell-u2413-monitor...

I bought these from Ebay.

12-Pack 1K Ohm 0505 1;8W 1% SMD Chip Resistors 1K0 11.0 kOhms 0.125 .125 W -USA

https://www.ebay.com/itm/234515403031?ha...

Any suggestions or help on this regard would be greatly appreciated!

crwdns2934271:0crwdnd2934271:0crwdne2934271:0

crwdns2934285:0crwdne2934285:0
crwdns2934057:0crwdne2934057:0

A number of my colleagues have dual U2413f monitors on their CAD workstations. One of them has developed the same fault as described in this thread, it shuts down when it heats up.

I saw the solution of changing the thyristors to ones with less sensitive gates. I couldn’t easily get suitable replacement parts locally, so I looked for a different solution.

I drew out the schematic of the crowbar circuit which has a standard configuration. I decided that a simpler method of fixing the problem was to stiffen the Zener bias resistor to keep the thyristor gate voltage lower during normal operation.

Block Image

The simplest method of achieving this is the put 1k in parallel with the existing 1k resistor resulting in halving the value.

The best part of this fix is that it can be done from the back of the PCB without unscrewing it and removing it from the metal frame.

For the benefit of others here is the process:

1.       Using a plastic guitar pick type tool along the seam between the front bezel and the back of the case, press in the clips and release the bezel.

2.       Place the monitor on its front and carefully lift off the back.

3.       Lift the metal shield and carefully peel the ribbon cable adhesive pad off the LCD.

4.       Press-in the clips and release the ribbon from the LCD.

5.       Fold over the PCAs to access the back of the PCBs.

6.       If necessary, the PSU PCA can be removed by squeezing the release clip on the inside of the PSU SIL connector on the signal PCA.

7.       The thyristors in question, Q971 & Q972 have 1k between their gate and cathode.

8.       Reduce the Zener bias resistance to 500R.

9.       Solder a 1k 0805 on the gate and scrape some resist off the ground plane and solder.

Block Image

Block Image

Since carrying out this modification the monitor has not had any further problems.

crwdns2934105:0crwdne2934105:0

crwdns2889612:0crwdne2889612:0 14

crwdns2944067:014crwdne2944067:0:

Seems legit. I have one more monitor, that I might try this on. Hopefully, no issues arise by the additional stress.

crwdns2934271:0crwdnd2934271:0crwdne2934271:0

Dołożyłem opornik zgodnie ze schematem. Wszystko działa jak należy.

Super !

crwdns2934271:0crwdnd2934271:0crwdne2934271:0

Works as a dream. Wszytko działa jak należy. Thanks.

crwdns2934271:0crwdnd2934271:0crwdne2934271:0

Sorry for my poor english but I don't understand the meaning of :

"scrape some resist off the ground plane and solder".

If someone could explain, it will be great...

crwdns2934271:0crwdnd2934271:0crwdne2934271:0

Can you provide a link to the 1k 0805 used?

crwdns2934271:0crwdnd2934271:0crwdne2934271:0

crwdns2934275:09crwdne2934275:0

crwdns2934285:0crwdne2934285:0

So the two thyristors are the answer for this problem. Q971 and Q972. Unforunately the original MCR part had a non-standard arrangement of Gate, Anode, Cathode (K) leads, and there are none I could find with the correct order, so you MUST bend the pins to have the right one go to the correct hole on the board, changing the Littlefuse Thyristor from KGA pin order to GAK order to match the original part - you CANNOT use the default arrangement.

The replacement part you want is Littelfuse S4X8ES2, and mouser.com has them for 45 cents each - the $7.99 shipping will be the most expensive part of this. I ordered 4 since they were so cheap so I had spares in case I accidentally broke a pin in the bending process.

The data sheet for the part is here: https://www.mouser.com/datasheet/2/240/L...

To open the case on your Dell U2413, I highly recommend the ifixit kit with the guitar pick things (or actual guitar picks). They get the case open without marring it.

I’ve attached a few pictures showing where the parts are on the power supply board in the monitor and also how to bend the pins of the replacement thyristor so it has them in the correct order when you solder it onto the board. This will stick up a lot higher than the original due to the bending (so you don’t have any of the legs touch - DO NOT HAVE THEM TOUCHING EACH OTHER AT ANY POINT), but that’s no problem.

Block Image

Block Image

Block Image

Thank you Dell for cheaping out on parts, cutting another corner that your customers have to figure out to avoid throwing a perfectly good monitor away.

crwdns2934105:0crwdne2934105:0

crwdns2889612:0crwdne2889612:0 7

crwdns2944067:06crwdne2944067:0:

Hi,

Perhaps suitably sized heatshrink tubing over the component legs will reduce any inadvertent short circuiting.

Just a thought.

crwdns2934271:0crwdnd2934271:0crwdne2934271:0

Yeah, that's a possibility. If you bend pin 1 three-dimensionally (I went back, over, down, forward, down) there's no chance of it touching the other two pins. Since the point of this part is temperature sensitivity, I wasn't sure if adding tubing might throw that off by insulating the pins, so I decided against it and just went with careful bending.

It was just nice to get this monitor working again for a little over $10 Finding the part was the hardest part since I originally searched for a 1:1 pin compatible replacement and came up short.

crwdns2934271:0crwdnd2934271:0crwdne2934271:0

Hi,

You would know better than I would but I thought that the temp sensing would be done through the component case and not the connections. In normal circumstances if it were mounted close to the board there wouldn't be much exposure of the legs.

Also you would only have to cover the one leg to insulate it from the other two

Anyway, still well done!

crwdns2934271:0crwdnd2934271:0crwdne2934271:0

Here is the data sheet for the original component.

https://www.datasheetarchive.com/pdf/dow...

You'll notice that the operating temperature maxes out at 110, where the component you are replacing maxes at 125. Makes sense why it doesn't work, as the monitor runs hot.

crwdns2934271:0crwdnd2934271:0crwdne2934271:0

So I just performed this repair on one of my monitors and it's working great. Thanks for your help on this. It's people like you that make this community great. Just wish that this repair was better known and could help keep more electronics out of our landfills. As an American, we are so wasteful.

crwdns2934271:0crwdnd2934271:0crwdne2934271:0

crwdns2934273:01crwdne2934273:0

crwdns2934285:0crwdne2934285:0

It has been running for two days without turning off. I put new thermal paste on the K15A60U and I attached a radiator to the LD7904jgp7

Block Image

crwdns2934105:0crwdne2934105:0

crwdns2889612:0crwdne2889612:0 2
crwdns2934285:0crwdne2934285:0

Hi. This is not exactly the same problem. My monitor doesn’t turn on anymore. I opened it and checked the boards. It seems there has been something going on in power input circuit and couple of components has blown. Is it possible someone could provide me original values for C909 and FB901. The rest I can still read. If one of you wonderful persons could help me out with these components, I could start fixing the monitor.

Thank you very much.

crwdns2934105:0crwdne2934105:0

crwdns2889612:0crwdne2889612:0 0

crwdns2942205:01crwdne2942205:0:

I don't have easy access to the circuit any more. FB is normally a ferrite bead. In the absence of any other assistance you could look to see if there are any other similar looking sections of circuitry in the area and measure the components.

Look out for failed semiconductors near the other failed components, I sounds like you may have had an over-voltage across the cap and excess current through the ferrite..

I would be very careful just replacing the cap and ferrite until you have checked the circuitry connected to them.

crwdns2934271:0crwdnd2934271:0crwdne2934271:0

crwdns2934285:0crwdne2934285:0

crwdns2934229:0crwdne2934229:0

Dan Carlo Canlas crwdns2934231:0crwdne2934231:0
crwdns2936625:0crwdne2936625:0:

crwdns2936751:024crwdne2936751:0 5

crwdns2936753:07crwdne2936753:0 23

crwdns2936753:030crwdne2936753:0 96

crwdns2942667:0crwdne2942667:0 11,957