Author Topic: Lenovo Pro2840mD monitors power down after a while (Read 4070 times)

EvilDeece · « **on:** January 18, 2021, 04:05:01 am »

Hi folks,

I have a couple of Lenovo Pro2840mD monitors that I pulled out of an e-waste pile.

Both monitors work initially, but after a while, power down hard (no standby light, buttons unresponsive), and will not come back up unless power is cycled at the wall. Once power cycled, they last less time than when brought up cold.

I've hunted around but can't find replacement PSU boards for them.

I've buzzed out all the electro caps in the PSU, and they all appear to be good (capacitance in spec, no leaking, low ESR).

There are 2 control chips on the back of the PCB:
LD7790GS (https://datasheet4u.com/datasheet-pdf/Leadtrend/LD7790/pdf.php?id=770292 )
GB98AEGN - backlight inverter? (https://www.datasheetq.com/datasheet-download/142142/1/O2Micro/GB98A )

I'm guessing it's a temperature related issue based on the behaviour, but I need assistance in narrowing it down.

It would be tricky to bring the monitor up with the board exposed, due to how it's mounted (screwed into a metal shield which is then bolted to the back of the panel), otherwise I think using a thermal imager might reveal some candidates.

Does anyone have advice on how to investigate the problem further?

[EDIT] I forgot to mention, both these monitors are circa 2015

james_s · « **Reply #1 on:** January 18, 2021, 04:53:22 am »

Problems when cold that get better when warm is a classic symptom of bad electrolyic capacitors. My first suspect would be that small one near the middle, that's probably part of the startup circuit.

EvilDeece · « **Reply #2 on:** January 18, 2021, 05:06:08 am »

Thanks for the tip, it's the other way around though, the shutdowns become more frequent when warm. I'll re-check that cap anyway.

james_s · « **Reply #3 on:** January 18, 2021, 06:35:19 am »

In that case try poking around with an insulated stick and see if you can find a ticklish spot. Cracked solder joints are a common cause of intermittent problems. A can of air duster held upside down to freeze parts can also be helpful.

tonycrouch · « **Reply #4 on:** January 30, 2021, 05:01:14 am »

Hey there,
You are certainly not alone in your plight with these monitors.

I use a dual setup of these 2840md's @ work (1 works perfectly, the other has the same symptoms you describe).

I've recently purchased a 2nd unit (with the same auto-shutoff fault to hopefully work out the problem).

I had originally suspected LD7790GS IC to be the culprit as a result of high - temperature but, like your picture, mine also has no heat marks / damage.

My CAPs are also within spec and show no physical signs of problems.

My only thought is to replace LD7790GS IC and hope for the best. However, I cannot be certain that there is not something further down the line that is not causing an over-voltage which in turns causing the IC to shut off power.

tonycrouch · « **Reply #5 on:** January 30, 2021, 05:08:24 am »

Also, on closer inspection the only physical defect I could see was the solder joint between C822 & D851 - as shown in the attachment.

But given the fact this monitor initially powers on, and takes 10 odd minutes to shut off then I highly doubt this is a problem.

amyk · « **Reply #6 on:** January 30, 2021, 05:10:18 am »

Use freeze spray (a can of "duster air" upside-down will work) to find the heat-sensitive component.

tonycrouch · « **Reply #7 on:** January 30, 2021, 05:28:05 am »

Thanks for that :-)

I don't have any air in a can to use. However, I do have this from loctite - freeze & release - would that work? It feels just like an upside down air can on my arm at least.

https://www.autobarn.com.au/loctite-freeze-and-release-aerosol-310g-far

Apologies for the n00b questions. But, if the IC was over - temperature, would that not burn / mark the PCB?

Given, the PCB is so clean (in both our cases) would that not mean it's something else causing the IC to fail for another reason (ie. no heat - maybe over-voltage or something else?)

amyk · « **Reply #8 on:** January 30, 2021, 09:36:25 pm »

That contains R134a which is a fine refrigerant, but also flammable butane, propane, and also the oil which might be problematic.

A heat-sensitive component doesn't have to be itself overheating, but just contain something like a tiny crack that opens with heat.

tonycrouch · « **Reply #9 on:** January 30, 2021, 10:34:36 pm »

Hi All,
I’ve ordered a replacement LD7990GS chip last night - but sadly won’t be delivered until the beginning of March.

I’ve put this monitor back together this morning and it’s currently working (which is good means I didn’t break it getting it apart) - but I suspect it’s only a matter of time before it randomly powers off.

Will report back should there be some amazing improvement.

EvilDeece · « **Reply #10 on:** January 30, 2021, 10:50:44 pm »

I've set the machine up with the PCBs exposed, and tapped vigorously on every through-hole component, as well as the PCB wherever there was space (to the point of introducing minor deflection in the PCB), and couldn't get the problem to reproduce. I might try warming up areas with an SMD rework gun and see if I can find an area that is heat sensitive.

tonycrouch · « **Reply #11 on:** January 30, 2021, 10:57:11 pm »

As an aside, not sure if this will help at all - but I’m currently running the monitor without it’s additional peripherals (ie speaker / menu buttons etc) - no improvement as yet - and I highly doubt it will fix the problem but I figured it was worth a shot

UPDATE - no improvement - shut off after approx 3 mins

tonycrouch · « **Reply #12 on:** January 31, 2021, 12:07:14 am »

Decided to check out the GPU board - there’s only this pink pad between the GPUs and heat sinks. Is this normal?

tonycrouch · « **Reply #13 on:** January 31, 2021, 12:09:35 am »

what I’m assuming to be GPU chips ...

tonycrouch · « **Reply #14 on:** January 31, 2021, 12:29:46 am »

Ok - another update - I’m starting to think the GPU IC chips (or at least that’s what I’m calling them) are the cause of this problem.

I have disconnected all internal cables on the monitor except for the secondary board and the backlight (as it won’t turn on without the backlight).

I placed a fan I stole out of my young fellas bedroom aimed at the GPU heat sink on its lowest setting - the monitor worked flawlessly for 30 mins (25 mins longer than the screen has ever stayed on before). To clarify the monitor DID NOT turn off - I just got bored waiting for it to turn off

I turned the fan off and within 4 mins the monitor turned off - the GPU heat sinks were very hot to the touch.

I’m thinking of pulling off the pink pad underneath the heat sinks and applying thermal paste and seeing how we go.

Rasz · « **Reply #15 on:** January 31, 2021, 01:45:11 am »

The Scaler chip looks BGA, probably cracked solder joint underneath, warms up, pcb flexes, loses connection, crashes, turns off. This is how Xbox 360 RROD worked. Reflowing with lead balls will fix it.

EvilDeece · « **Reply #16 on:** January 31, 2021, 01:57:11 am »

Hmm, the thermal pads are pretty thick, they may need to be replaced with a shim instead. I measured mine at 2mm.

I'm thinking the easiest option would be 1.6mm aluminium flat bar from Bunnings: https://www.bunnings.com.au/metal-mate-20-x-1-6mm-1m-aluminium-flat-bar_p1064257

In the meantime, I'll try reflowing the video ICs - there's 3 of them on the board, each with their own 50x50mm heatsink.

tonycrouch · « **Reply #17 on:** January 31, 2021, 02:10:07 am »

Good idea on the BGA needing a reflow.

My thermal pads were very, very thin (far thinner than 2mm) - and turned into mush when touched. So I reckon they’ve had it.

I’m gonna order some new heat sink pads (I’ll get a few diff thicknesses) and see how we go.

But I suspect the reflow might be the ultimate fix. I don’t have access to a heat gun - so @evildeece if you are able to give it a whirl, I’d be much appreciated.

EvilDeece · « **Reply #18 on:** January 31, 2021, 03:10:51 am »

I'm just reassembling after reflow (it's entirely possible I botched it as I bumped one of the chips and had to nudge it back).

I did notice that some of the screw holes on the scaler board sit flush, while others don't. This may affect the heatsink contact. I'm going to dill those holes out to ensure better contact.

[Update] Ignore this, the screw mounts on those holes don't rise up to penetrate the board.

tonycrouch · « **Reply #19 on:** January 31, 2021, 03:22:50 am »

Fingers crossed for you mate - let me know how it goes 👍

Rasz · « **Reply #20 on:** January 31, 2021, 03:42:13 am »

Quote from: EvilDeece on January 31, 2021, 03:10:51 am

I'm just reassembling after reflow (it's entirely possible I botched it as I bumped one of the chips and had to nudge it back).

I wouldnt turn it on after that, have to remove, reball, and solder again

EvilDeece · « **Reply #21 on:** January 31, 2021, 03:46:54 am »

Yeah, definitely botched it... On to unit #2...

(I don't have the time & patience to learn how to reball it for the sake of a 5 year old monitor)

EvilDeece · « **Reply #22 on:** January 31, 2021, 04:52:30 am »

Botched the second screen too - nothing was disturbed, but maybe overheated? The screen comes up white, and the power LED does not light and none of the buttons respond.

I guess these are going back into the e-waste pile.

tonycrouch · « **Reply #23 on:** January 31, 2021, 05:38:53 am »

Ah mate that’s so sad - good on you for giving it a go 👍

I have bugger all BGA experience, and worried I will stuff it up - might be time to call in a pro - sadly though the nearest BGA place to me is 4+ hours away.

tonycrouch · « **Reply #24 on:** January 31, 2021, 05:40:02 am »

On second thoughts mate, those chips are about $30 each - I’m thinking of ordering some - I can ship you some if you want to give it another go 👍


EEVblog Main Site	EEVblog on Youtube	EEVblog on Twitter	EEVblog on Facebook	EEVblog on Odysee

Author Topic: Lenovo Pro2840mD monitors power down after a while (Read 4070 times)

Share me