Starts up then powers down

Topic

New Reply

I have a Systemax BAX4200 that uses a AMD Athlon 64 X2-4400 processor and a K8N Neo4-F motherboard, 2 optical drives IDE, 3 HDD SATA, a Floppy drive and a multicard reader plus 3 cards. A while back, the unit would start up and in 3 seconds shut down. If I removed AC and waited a few seconds then repowered, it would start up and crash. Even if I lightened the supply load by removing the supply to the peripherals, no go. By going to Ebay, I obtained another motherboard (possible bad caps on board the reasoning), a new P/S (actually 2 of them) and another processor-hey, they were cheap enough being that they were for an older unit and not much in demand.
I then replaced the motherboard and the P/S (the cpu had not come at the time) and it worked. this happy state of affairs lasted for a month then once again the problem starts: a bit intermittent at first, quickly devolving into nonfunctioning. Since I do electronic audio repairs, I have some equipment to check things out. Normally, when a working unit is first plugged into AC, I would get a quick bump on my AC current meter then it is quiet. The START button is pushed and on it comes with variable current draw in it’s startup routine. Now if I apply AC to the unit, it starts up then does a shut down. If I wait a minute until the PS-on (green wire from supply) goes back to 0 V then apply AC, it will start up again and do its thing. A new P/S does the same thing.
Should I try the new CPU? It’s the only thing that is the same from the original install save the peripherals. I do not know if the CPU would send a signal to the P/S for some sort of protection as home theater receivers do. I hesitate because the install is a touchy, somewhat messy affair that hopefully would not need to remove the motherboard: once is enough.

Reply | Quote

Replies

Did the processor separate cleanly and smoothly from the motherboard and especially from the heat sink? Unless the thermal paste has been changed within the last few years , the processor and heat sink can be pretty much married to one another. That all needs to be carefully cleaned once separated and have a tiny fresh drop of thermal paste applied before assembly of heat sink to CPU. Also check that the connection is firm all around without the heat sink catching on the guides along the side. Also, I don’t remember for sure but if those processors have the four little standoffs out toward each corner in addition to the small center chip contact you have to be extremely careful not to put too much pressure on any one corner or it will snap.

If you have satisfied all the possible thermal challenges that could be causing quick shutdowns then it must be something else. I always assume thermal problems with quick shutdowns.

As long as you have the CPU it should be easy to change without changing the motherboard if you took all the maintenance steps described earlier.

Reply | Quote

Does it get through the ‘POST’? – Do you hear any ‘beeps’? Does it fail solid? Can you put a current meter to the AC line?
Take it down to motherboard – cpu – video – one memory stick – see if it comes up and says ‘no operating system found’ or just reboots.

Reply | Quote

By going to Ebay, I obtained another motherboard (possible bad caps on board the reasoning), a new P/S (actually 2 of them) and another processor-hey, they were cheap enough.

A “cheap” power supply could be the culprit. Some power supplies are literally junk, while others are very good.

Corsair makes a good power supply. Here’s a review on one of their power supplies:
http://www.hardwaresecrets.com/article/Corsair-HX650-Gold-Power-Supply-Review/1705/1

Here’s some good info about power supplies in general:
http://www.hardwaresecrets.com/article/Anatomy-of-Switching-Power-Supplies/327/1

Group "L" (Linux Mint)
with Windows 10 running in a remote session on my file server

Reply | Quote

As far as the CPU goes, I am using the original chip and heat sinking assembly and the unit had been working fine for a month after the transplant. I wanted to use the original chip as it is much peppier than the one that came with the (new) mobo. The power supplies are rated at 350 W both from the same manufacturer with the extra one being a 400 W unit. I tried plugging that one in on an external basis using just mobo and C drive powered-same problem. This particular unit has no POST beeps and yes, the unit was plugged into an AC meter to monitor current draw during boot. Using W=E x I, max power draw stays around 200 W max. Current draw pattern is the same either during a proper bootup or it’s current truncated pattern up to the shut down time.
I’ll try the really bare bones set up to see what happens and if that doesn’t work, I guess I’ll put in the extra CPU and cross my fingers.

Well, I’ll be. . . I took winsong02’s advice and disconnected all peripherals and went to one memory-still failed. Just for shiggles, I then removed the video card and bingo, it stayed on! I replaced the memory only; stayed on. Reconnected all drives and tried; stayed on. Put back video card; shut down. removed card; stayed on. That’s enough for me; I just put in an order for a new video card. The old one was a PNY GeForce 6600 GT and did well enough; I’m no gamer. I opted for a GeForce GT520 with 1GB RAM; that should do me. While I was at it, I checked the CMOS battery-0.8 V. I did not have a CR2032 to put in but used a CR2025-thinner but the same diameter.

Reply | Quote

Video cards can definitely cause problems if they are bad. My wife had a bad video card once which caused random shutdowns after being logged on for around 10 minutes. I ran a program called speedfan and found that there was an overheating problem, then determined that it was the video card (I don’t remember how I determined that). Replacing the video card solved the problem.

Group "L" (Linux Mint)
with Windows 10 running in a remote session on my file server

Reply | Quote

Ah, not working, then working for a month or so without a problem, then not working again tripped me up; that is not typical of a failed video card, unless it’s actively cooled and you might have noticed the fan was working while it was functional and acting up when it was failing again.

Reply | Quote

This is being typed on the (formerly) dead Systemax. The new video card was installed and drivers updated. Looks good, tastes great. I got an EVGA GeForce GT520 since I’m no gamer; reviews pan it for that use. I had to do some other types of updates first as they were very insistent and/or needed for security but it’s back on the air.

Actually, the system was acting just like F.U.N. Downtown’s unit in that it too worked for a month or so after all the hooraw of moving boards around and yes, the video card was actively cooled as is the current one. I couldn’t tell about the fan as the layout of the card(s) keep the fan pointed downwards disallowing sight without extreme repositioning of the box or using a mirror.

Reply | Quote