Question PC Shuts off randomly without warning. Help diagnose cause?

[kesomon]

Hello! How are you all this fine evening, hope you are well.

Recently I decided to upgrade a few components in my PC to improve performance. Added more RAM, upgraded the graphics card, and put in a new PSU. However, since then, at times the whole computer will shut off at random without warning. At first, I thought it was the fault of my antivirus program, as it would shut off most often during the early period of bootup - I thought it was the antivirus doing an update in the background and rebooting to finish without giving a heads up. But I just uninstalled the antivirus, and it crashed on me again last night. So clearly the antivirus was a trigger, but not a cause.

I have the feeling the answer to this question will be 'your <Mod Edit> is old, replace it all' but I wanted to see if a specific cause could be diagnosed which I could fix without having to buy a whole new setup.

The specific specs of my PC are:
MB: MSI 7H77A-G43
RAM: 16GB
Processor: Intel Core i5-3350P CPU @ 3.10GHz (4 core)
OS: Windows 10 Pro
Graphics: AMD Radeon HD 6870 (bought used, fan makes terrible screeching/buzzing noise, many regrets)
PSU: 750w (also bought used)

Not sure what else I need to list. Anyone spot anything immediately and obviously wrong?

So, my potential hypotheses:

• I do know that the CPU hasn't been upgraded, and has never had new thermal paste applied, so overheating might be the issue, but I also don't run that many heavy things. At most, I usually play The Sims 4, or watch videos online while also browsing other sites.
• graphics card might be to blame. The fan switches on and off frequently, and buzzes/shrieks when it does, so maybe it isn't drawing enough or drawing too much power and not able to regulate it well.
• PSU may be borked, since I got it used from the RE-PC. Had a hard time finding a unit that could fit in the tower and had the right wire connections for the graphics card.

Any knowledgeable help would be appreciated! Many thanks in advance.

 

[Pistolpackin]

Did all this start to happen after the used parts were installed?
Not to beat you up or anything, but we never ever use used power supplies, ever. Power supplies lose up to 5%+ efficiency each year.
What was your previous video card; onboard (APU [cpu & GPU in one]) or was it another discrete video card (what make and model) ?
Looks your motherboard has onboard video. I would remove the video card and observe the behavior using onboard video only.

Looks like that card pulls 141w during idle and 317w under full load. The problem is most likely the power supply, but the card could be too. Remove it and see if your computer crashes with no video card in it.
https://www.techspot.com/review/325-amd-radeon-6870/page10.html

 

[kesomon]

Yeah it started after the new parts were put in. No issues with the previous video card, though I don't remember what it was it was much less powerful than the Radeon. It was a separate card though, not onboard. never tried the new PSU with the old card, I probably should've.

I should'a just gone with a full new build rather than try to hack it together from castoffs. lesson learned.

 

[kesomon]

Update: Last night I checked CPU temp in bios after a zoom meeting, and it was running at about 63*C. I changed Bios settings to Eco/power saving, in case it made a difference. I then had one window of chrome and the task manager open, and played two hours of The Sims 4, to see what the power draw load was. Sims pulled High Power on occasion but was on average moderate. After about two hours of play, the computer shut down abruptly (wiping my unsaved progress, naturally.) So still not sure if it's power flux or overheating, but given how long it ran without dying, I'm leaning towards the latter currently. I may try to get and apply some fresh thermal paste to start.

 

[DSzymborski]

What exact PSU are we talking about here?

 

[Pistolpackin]

Update: Last night I checked CPU temp in bios after a zoom meeting, and it was running at about 63*C. I changed Bios settings to Eco/power saving, in case it made a difference. I then had one window of chrome and the task manager open, and played two hours of The Sims 4, to see what the power draw load was. Sims pulled High Power on occasion but was on average moderate. After about two hours of play, the computer shut down abruptly (wiping my unsaved progress, naturally.) So still not sure if it's power flux or overheating, but given how long it ran without dying, I'm leaning towards the latter currently. I may try to get and apply some fresh thermal paste to start.

Did you try it without the video card installed? Does sims4 run poorly using APU?
Take it off eco power saving and check event viewer and view 'error' leading up to the timestamp of the crash.

 

[kesomon]

Apologies for the delay in responding, It's been a busy week.

@DSzymborski the exact PSU being used is the 'Fatal1ty 750w', #OCZ750FTY model. Not sure if that is the correct information to give.

@Pistolpackin I booted up to check event viewer today, and by happenstance the graphics cooling fan started shrieking at an unbearable pitch. I have now removed it and installed the old card, which I was able to find in the dusty junk box, thankfully. The old card is an AMD Radeon HD 5670, which ran Sims4 fine before.

I'll test out playing it for a while to see if it shuts down on me with the old card now. If it does, then it's likely something wrong with the PSU or CPU to overheat.

 

[kesomon]

After playing The Sims 4 for 5 hours without a single hitch, removing the newer graphics card has solved any issues I had, so the flaw was either in the card or card + power supply combo. On the negative, this both means I bought a card that's a dud I can't return for a refund, and that I don't have a usable card powerful enough to run any other modern games without quality loss. sigh. guess I'll close this thread as 'solved, but not satisfied.'