R7 240 intermittent black screen, can't solve

[Guilhermevarella]

Hey everyone. I just bought a R7 240 2GB GDDR5, and since it arrived i got the black screen issue since the first day. It happens at totally random moments, sometimes i could use it for 8 hours, gaming, browsing and all, temps reached a max of 65º, no issues detected, but when i was just browsing in the next day, within 15 minutes the black screen came, the pc never shutdown or restart, all the lights still on, just a black screen and all freezed. I tried everything, recent and old drivers, installing it without the radeon software (for every try i used the DDU in safe mode), created the TdrDelay at regedit, tried configuring power and fan settings at the radeon software, reducing the core clock, RadeonSoftwareSlimmer, disabling windows auto updating, factory settings reset, it seems nothing can solve this, somedays it got worse cause the issue keeps happening when i try to install the driver, after it, it happens at windows logon. On the onboard, pc run all fine. All other parts are newly acquired:
mobo Biostar 1155 H61 MHV2;
I5-3470 2.9 GHz;
8gb RAM DDR3;
PSU Thermaltake 430w 80+white;
SSD 120GB Adata (lost a HD recently);
monitor Samsung T24D310LH 24"
all the voltages are ok, the PSU was recently tested and it ran a GT730 2GB for more than a year, never had a problem with power supply. Someone have any suggestion? I don't want to resort to warranty now, because it seems pretty clear it's a software issue, since sometimes i can use the gpu for hours, gaming and all, without the issue, and sometimes it won't last 15 minutes browsing, and the problem showing at driver installation... prints i took while the card was on, actually using onboard.

 

[FreedomEclipse]

Download GPU-Z and give us a screen shot.

 

[Guilhermevarella]

Done.

Download GPU-Z and give us a screen shot.

 

[FreedomEclipse]

Im at a loss then. Have you tried running older drivers?

If you have and the card still crashes then it could be the card itself thats on its way out. If the card is under any sort of warranty, send it back before its too late. some second hand shops do offer 12month warranty from date of purchase so it depends where you got it from, you might be SoL if it was an ebay purchase unless you can convince ebay that the seller sold you a dying/defective card.

 

[Guilhermevarella]

Yes it is on warranty, i just want to certify it is not a software issue (due to it happening at random moments and not limited to high usage moments, installing drivers crashes and all), cause if is not a hardware issue, i would use the warranty, get it back and still not able to use it, anyway, thank you, if i can't find a answer in forums i will send it back this week!
Edit: yes, i tried almost every driver on AMD site for r7 240, no major differences, some of them showed the black screen right on installation moment

 

[FreedomEclipse]

Yeah send it back.

I had a system that was randomly BSODing because the motherboard was slowly on its way out. I could be hours or days between BSODs but it would still BSOD. I didnt replace the board as it was an X79 board and people were charging stupid money for secondhand boards but everything else either went into a new build or got sold off for parts and I didnt hear any complaints from the buyers so I assume the parts were still good.

You could try testing the card in a different PC just to be double sure but if your 730 worked without any issues then there is no real reason to suspect that the problem is anything else because both cards draw roughly the same power.


---

did you do a clean driver install with DDU??

if you did and the black screens still kept happening then send the card back for another one.

 

[Nabarun]

I know this may sound stupid, but lose cables can do miracles. Particularly DP/HDMI. Sometimes dust can do the same, but since the card is new, I'm assuming that's not the case here. Also, driver issues can be checked by trying a couple of older "recommended" WHQL ones, doing "clean" installs.

 

[Guilhermevarella]

I know this may sound stupid, but lose cables can do miracles. Particularly DP/HDMI. Sometimes dust can do the same, but since the card is new, I'm assuming that's not the case here. Also, driver issues can be checked by trying a couple of older "recommended" WHQL ones, doing "clean" installs.

Yes i checked all the cables and tested it using HDMI and DVI cables, no changes...

Yeah send it back.

I had a system that was randomly BSODing because the motherboard was slowly on its way out. I could be hours or days between BSODs but it would still BSOD. I didnt replace the board as it was an X79 board and people were charging stupid money for secondhand boards but everything else either went into a new build or got sold off for parts and I didnt hear any complaints from the buyers so I assume the parts were still good.

You could try testing the card in a different PC just to be double sure but if your 730 worked without any issues then there is no real reason to suspect that the problem is anything else because both cards draw roughly the same power.


---

did you do a clean driver install with DDU??

if you did and the black screens still kept happening then send the card back for another one.

My mobo was bought in january, the one that was running with the GT 730 was sold and i got a new one, so the PCI slot was untouched, also, at a tecnical assistance, it was told me they tested a GTX 1050 with this mobo and it was fine, the r7 240 worked on another pc, but since the black screen appears in random moments, it was inconclusive. Yes i've done many DDU clean installs, using it in safe mode, preventing windows from auto updating, even with the older, but recommended drivers... already talked with the store to treat about the warranty, until then, any ideias still welcome haha

 

[FreedomEclipse]

My mobo was bought in january, the one that was running with the GT 730 was sold and i got a new one,

Ah you see. This is why you should tell us as much as you can during your first post because up until now i was under the impression that you had been running the same board all this time. Everything ive told you up until now is probably a waste of time when the issue could have been a bad PCI-E slot on your board.

do you possess the fingers to unplug the card and inspect the pci-e slot? Just look for any bent or discoloured pins. It might be also worth giving the pins on your card a wipe down too with some isopropyl alcohol or contact cleaner before plugging it back in and seeing if its fixed.


Before you open the case up just make sure you unplug your PC from the mains.

 

[Guilhermevarella]

Ah you see. This is why you should tell us as much as you can during your first post because up until now i was under the impression that you had been running the same board all this time. Everything ive told you up until now is probably a waste of time when the issue could have been a bad PCI-E slot on your board.

do you possess the fingers to unplug the card and inspect the pci-e slot? Just look for any bent or discoloured pins. It might be also worth giving the pins on your card a wipe down too with some isopropyl alcohol or contact cleaner before plugging it back in and seeing if its fixed.


Before you open the case up just make sure you unplug your PC from the mains.

Sorry about that. When i left in the assistance, it was all checked and cleaned to be tested with another gpu (gtx 1050). I'm thinking about calling another assistance in another store, one able to check hard and software also, too bad the delivery time / diagnosis takes like 4~5 days due to high demmands. The card pins at least looks like when it arrived, normal, tiny marks from being plugged and unplugged a couple of times

i'm always wondering why it happens at so random times, and specifically when trying to install some driver versions, also when changing settings like fan speed and core clock in adrenaline software, is the main reason i'm ''certain'' it should be a software problem.
(in any case, sry if i messed in talking english, not my main language)

 

[Guilhermevarella]

Update
I left it at a technical assistance, pc remained 16h on stress testing, no issues detected, when at home, after 1h, low usage, the issue showed up again....

 

[xrror]

Do you have a picture of the card? Is it like this one? https://www.techpowerup.com/gpu-specs/radeon-r7-240.c2463

tl;dr - I've worked with a lot of Dell ATi cards that are really similar to the one pictured above, and even with dead/stuck fans they usually still work. If you can, you really need to test that card in another machine if possible. But it's probably the card.

But after seeing many of these cards in some pretty hellish conditions with dead fans and covered in dirt and metallic dust yet still working, it takes a lot to make these not work.
The only other suspect is if your motherboard / PCIe slot is getting cranky. But that's another rabbit hole. And usually just cleaning the card edge will fix this.

You really need to test this in another machine and see if it still hangs.
-----
Some long-shots: this probably isn't it:
At my workplace we have a lot of these cards that came with Dell systems, especially the low profile. I'm sure you've checked the fan already, and I'd guess it would have shown up in stress testing but we end up oiling a lot of these. But even then normally the system doesn't crash, we just get complaints about the fan buzzing and making noise so that's probably not your issue.

Is your card a passively cooled version? On the slew of low profile dell cards we have, there is a red anodized heat pipe cooler that was used on quite a few ATi OEM low profile cards that ends up swelling, and killing the card! It literally crushes the core - but it's also super obvious when you look for it. But I'm not sure they used that on any 240's. I seriously doubt it's this... unless you just happen to have that heatsink.

 

[Guilhermevarella]

Do you have a picture of the card? Is it like this one? https://www.techpowerup.com/gpu-specs/radeon-r7-240.c2463

tl;dr - I've worked with a lot of Dell ATi cards that are really similar to the one pictured above, and even with dead/stuck fans they usually still work. If you can, you really need to test that card in another machine if possible. But it's probably the card.

But after seeing many of these cards in some pretty hellish conditions with dead fans and covered in dirt and metallic dust yet still working, it takes a lot to make these not work.
The only other suspect is if your motherboard / PCIe slot is getting cranky. But that's another rabbit hole. And usually just cleaning the card edge will fix this.

You really need to test this in another machine and see if it still hangs.
-----
Some long-shots: this probably isn't it:
At my workplace we have a lot of these cards that came with Dell systems, especially the low profile. I'm sure you've checked the fan already, and I'd guess it would have shown up in stress testing but we end up oiling a lot of these. But even then normally the system doesn't crash, we just get complaints about the fan buzzing and making noise so that's probably not your issue.

Is your card a passively cooled version? On the slew of low profile dell cards we have, there is a red anodized heat pipe cooler that was used on quite a few ATi OEM low profile cards that ends up swelling, and killing the card! It literally crushes the core - but it's also super obvious when you look for it. But I'm not sure they used that on any 240's. I seriously doubt it's this... unless you just happen to have that heatsink.

No, the model is this one, from PCYes:
After getting it back from technical assistance, i've been monitoring the issue withou removing the card. I noticed that it crashes almost everytime after i turn on the cpu for the first time, after like 5, 10 minutes, then it gets stable for a random number of hours. Another thing, it NEVER crashes during a game, 90% happens while i'm just browsing, just youtube etc, when it happens while gaming, ALWAYS is at loading screens, never at the lobby or during the game. On the weekend, after the first time after initializing, i could use it all day long without the crashes, constantly playing, next day when initialized, just browsing, it happened again. The issue don't happened at the assistance, they left in gpu/cpu stress test for 16 hours, and it didn't happened, at home, keeps happening. I don't see any way out anymore, tried changing cables, disabling drivers, changing power options, changing mouse/keyboard/audio device, still random moments to crash to a black/white screen and a buzzing sound...