Lots of possibilities, unfortunately. A FEW of them:
1. Somewhere there's a loose connection that breaks open and then resets itself, but the disruption causes the system to "freeze" in an unstable state and start a reboot, but fail to complete. This COULD be in any of the multiple plug-in connectors in your system. For a small example, even a poor connection at the CPU_FAN header on Pin #3 that carries the CPU cooling fan's speed signal back to the header can trigger the mobo to shut the system down to prevent a possible overheating of the CPU caused by fan failure. This "failure" may not be real, but if the header gets NO fan speed signal it believes that, anyway.
2. Similar concept, but this intermittent connection causes a momentary short circuit to Ground somewhere, resulting in similar situation.
3. A weak component on the edge of failure sometimes does fail, then recovers partially - same result. If this is the case, what MAY happen is that the situation is almost impossible to locate, but it gets worse and more frequent as time goes on. Hopefully if that happens its symptoms may become clearer and more complex until the likely cause can be recognized.
There are many other possibilities. One way to tackle this is to start replacing components until the trouble stops. BUT since you really have no way to know where to start this, it can get VERY expensive as a trial-and-error process, so I do not recommend that. The other way is to let this continue and start keeping a log of when it happens, what was going on at the time, and any other symptoms you notice. After a while if it continues, a pattern may emerge to guide you in identifying the likely weak or failing component.
Just a couple of personal stories related to my suggestions above. Do NOT take these are your best ways to handle, more like illustrations of SOME of the things I've experienced.
There is one cause I think is relatively common for some things and free to fix IF you are careful. Over time the metal contacts in connectors can develop a thin oxide coating leading to poor or intermittent connections at that point. If I suspect that, I shut down completely and disconnect from power. Open the case. VERY carefully go though the portions of your system you suspect are involved IF you have been able to narrow this down, and disconnect a connector, then plug it back in. Repeat several times. Be careful NOT to disconnect something else that you did not intend! Proceed to the next suspect connector. After you're finished, check again for disrupted connections and cables, or anything pushed out of place that might interfere with other stuff. Then close up and start up. If it is working OK, you may need to wait several days or weeks before you are confident this has actually helped. The disconnect / reconnect action can sort of "scrub" metal surfaces to remove the oxide coating causing the unreliable contact.
Between about 2003 and 2008 there emerged a rash of problems all traced to a particular type of electrolytic capacitor design that was used widely in PSU's and in mobos' Voltage Regulation sections. This problem was recognized as it became common and eliminated by 2010 or earlier by changing which type of capacitors were used in new machines, so it is NOT common now, but I will tell you my experiences simply as illustrations of intermittent troubleshooting. My first case was with this computer. Early events were sudden reboots in the middle of normal moderate uses. Then it would suddenly shut down and refuse to re-start with a push of the front On / Off button, BUT it could be re-started if I shut off the manual switch on the back of the PSU and left it off for a while (5 - 15 minutes) before turning back on. The pattern became more frequent over the next 6 - 12 months. I tried cleaning out dust, my contact cleaning process (above), etc, but to no avail. Finally, knowing of the capacitor problem, I removed the PSU and opened it up. Inspection showed one or two large capacitor cans with bulged tops, the typical symptom of that capacitor problem. Replacing that PSU completely solved it. A couple years later I had a similar problem with another machine in our retail store, but the symptoms were more severe and developed more rapidly. My son replaced its PSU, but the problem persisted. After a short time I opened up and inspected thoroughly again, and this time discovered bulged capacitors on the mobo in the Voltage Regulation area. Replacing the mobo solved that one. In each of these cases the problem began slowly and got more frequent over the space of many months before we had enough info to narrow down the search.
Facebook
Twitter
Instagram