PC Woes

[oops]

Hi everyone, I have some pea sea woes and I'm hoping someone can give me advice, suggestions, and general help.

About 3 years ago, I built my own system. The specs are as follows:

Intel P4 2.60C GHz
Asus P4P800 i865PE
OCZ Dual Channel PC3200 2x512MB EL DDR 400MHz CL2 Copper Heatspreader
Thermalright SLK-800U Socket -A/478 Copper Heatsink
Maxtor DiamondMax Plus 9 6y080M0 80Gb SATA150 7200RPM
SH 465W PSU
ATI Radeon 9600XT AGP 8x/4x 128b

Building this system cost me a fortune. However, about 6 months ago, I noticed that my computer acted strangely. Occassionally, my computer wouldn't exactly turn on. The keyboard lights would go on, the fan starts running, etc. But nothing showed on the monitor. Whenever this happened, I would wait a while and then press the reset button. After that, everything works ok. Since this was during frigid winter season, I assumed that it wouldn't start because it was too cold in my room and warmer in the other rooms. But now it isn't so cold and it still acts like this. It's gotten to the point where I have to reset the machine everytime I try to turn it on. Now, I'm not entirely sure if this is related but I ran a Seagate/Maxtor diagnostic and got a failed test result. The diagnostic said I should replace the drive. It's a brand new hard drive. It's barely 3 years old and I already have to replace it! I've gotten some suggestions from people but it's really all over the place. Some say that it's the PSU (I used a generic, unmarked, and possibly refurbished model). Something about unstable power fluctuations damaging my hard drive. Another said that I just got a HD with a shorter than usual lifespan. And yet another said that the PSU might have even damaged my mobo!

Regardless, I'm going to replace the HD with a Seagate Barracuda 7200.10 320Gb SATA 3Gb/s 16Mb buffer. I also ran some PSU power consumption tests (those online estimates) and I'm going to get an Antec TruePower 2.0 430W Dual 12V outputs. The question is, do I need to replace my mobo? I mean, how would I check to see if it sustained any damage? Right now, I don't even know what the real problem is! I could really use some help. If I'm going to fix my build, I'd like to do it at once and get it right at the same time. Getting the new HD and PSU will already cost me around $300 CDN. I really hope I don't have to replace the CPU and the mobo. Might as well build a new machine! I'd appreciate some suggestions/help on what to do in this case and as soon as possible. From the extreme slowness of my machine (the average disk queue length is going off the charts every few secs!), I think my HD will die on me any moment now.

 

[fueled_by_ramen]

I'll start off by saying (and this is based on personal experience and as such is probably biased) but I wouldn't be surprised in the least if a Maxtor crapped out after 3 years. In fact, 3 years is a long time for a Maxtor to work. I've had nothing but bad luck with all of the Maxtors I've owned plus the countless ones I had to replace for customers when working as a tech a couple years ago.

In terms of the cold causing problems, I've never heard of that before. Unless it was REALLY cold in the room (I mean near freezing cold) then it wouldn't cause problems. Computers don't like heat. Cold, if anything, is better for a computer (again, as long as it's not below freezing or something crazy).

Basically (and I know you don't want to hear this) all those people you talked to could be right. It's possible that the hard drive is just dying, possible that the generic PSU is dying, or that it's shorted out the hard drive or the MoBo. Luckily, the last two possibilities (while still feasible) are a lot more rare than a simple matter of a dying hard drive or PSU.

The simplest way to test the system would be to try different parts. If you had a spare PSU to pop in and run it, see if the same thing happens. And if you had a spare hard drive to do the same with that. If you were to try a new PSU and hard drive and the problems persisted, then unfortunately that leaves a problem with the system itself (ie. MoBo, CPU, RAM, etc). But if swapping one of the parts caused the problem to stop, well there you go.

Without the equipment to test PC parts, that's really the best way to troubleshoot any problem without more info or that isn't clear cut. They do have PSU testers (little devices with lights on them) that you can plug into the PSU to see if there is a short or a problem with one of the voltages. And they have PCI cards that you can plug into the MoBo that will run diagnostics and tell you if there is a problem (although I have less faith in these MoBo testers than I do in PSU testers). These devices are usually $10-$30 at your local computer store. But it's up to you, if you feel comfortable doing something like this or not.

If at all possible, try to swap out the PSU and/or hard drive. And see if the problem still happens.

 

[oops]

Hi, thanks for the reply.

Ok, so first things first, I need to get the new HD anyway since the Maxtor one is dying. Now, does Seagate have a long life expectancy? I'm thinking of the Barracuda one I mentioned in the previous post. But I'm a bit apprehensive after what you've told me about Maxtor HDs. I mean, Seagate just bought Maxtor... lol :?

Second thing, I do have (thankfully) a spare PSU. However, I think it's a generic one. I took a photo of its label as shown below. Do you think this will suffice if I disconnect my DVD-RW and CD-RW drives? Does it look like something that doesn't experience unusual power fluctuations? Should I use it to test to see if my current PSU is the problem or not or should I just go ahead and get that PSU monitoring stick thing?

Finally, what's your opinion on the OCZ mem and the Asus mobo that I currently have? I'd appreciate some insight as I've been out of touch with this hardware stuff for almost 3 years.

 

[alcattle]

It is good enough to test the MB. Start with the MB and one stick of RAM. Onboard GFX? switch those on. Plug the psu to the MB and see what happens

 

[fueled_by_ramen]

I mean, Seagate just bought Maxtor... lol :?

D'OH!

Well, I personally would still feel much more at ease with a Seagate.
Seagate or Western Digital...those are the two drives that I've seen the least amount of problems with. I'm sure others might have different experiences, but I think most would agree that they are more reputable than Maxtor.

Second thing, I do have (thankfully) a spare PSU. However, I think it's a generic one. I took a photo of its label as shown below. Do you think this will suffice if I disconnect my DVD-RW and CD-RW drives? Does it look like something that doesn't experience unusual power fluctuations? Should I use it to test to see if my current PSU is the problem or not or should I just go ahead and get that PSU monitoring stick thing?

That's a pretty weak PSU. The overall wattage is ok but with only 13amps on the +12v rail...that wouldn't be a good choice to use for anything other than testing, if that much.

If you wanted to, try it out. But make sure you have the bare minimum of devices drawing power from it. Basically, as alcattle said, use onboard video (take your video card out) but from what I am seeing, that board doesn't have onboard video (unless I'm looking at the wrong board). Unplug any other unnecessary parts (optical drives, sound cards, etc.) and use only one stick of RAM at a time. See if the problem persists.

If you can find one of those PSU testers to check your main PSU, that's always worth a shot. CompUSA has one for $13.
CompUSA ATX 12V Version 2.0 Power Supply Tester

Any chance you could use another hard drive to test that first? It'd be much easier and it could very well be the problem.

In terms of the Asus MoBo and OCZ RAM, they are both good companies. I don't see any problems with either in terms of manufacturer and specs.

 

[Drez143]

I never thought I would find someone with the same problem as me. I have a p4 3.4ghz, an asus p5p800 board, a raden 9700 pro video card and 4 seagate HD's. I have the same problem as you with the start up. I built my comp about 3 years ago aswell, and every time I start up my comp in the cold, I need to restart it right away to get video out. During the warmer weather though this problem becomes more 50/50. I contacted Radeon right away about this, and they told me it was my psu which was under powered. So around 2 yrs ago I bought a new antec 450 watt psu, and the same problem occured. Ive also gone through 3 MB's through Asus and warranty issues because the p5p800 MB is all around a cheapie that blows.

The reason I am writing though, is because I honestly dont think its a HD issue. Think about it. You would get video with your HD or not. If your HD was crapping out, after your boot screen you would get an error. The HD would not prevent your MB from sending display out. I found it weird that we basically have the same MB only mine is for 775 pin and we have the same problem. I would suggest looking elsewhere then the HD, even though you scanned it and got an error.

 

[will14]

So it occasionally boots sometimes it doesn't...
With a bad HDD I think that it would be more likely that you'd get past mobo post. PSU would make sense. As would a few of the others. Reminds me of when my computer went up in flames. I lost faith in all the parts.
It ended up turning out.
My psu was unstable fried my board, HDD's and Video card.
My ram/cd-roms/cpu were fine tho.
So you might have multiple problems.
Best way is if you have extra parts or rather know someone who does or who has a similar computer and might be willing to mix and match.

 

[fueled_by_ramen]

How cold is it in your houses? Yikes 8O

Just teasing. You're right though, normally it shouldn't be the hard drive, but with electrical shorts, you never know unless you test as much as possible.

If he could at least just pop a different hard drive in (since it's really easy to do) it would tell him for sure either way. I learned that when troubleshooting PC problems, you wanna try as much as you can, just in case.

I still have a bad feeling about the PSU though. Something else he could try is to pop a different video card in, if at all possible. (Which would also be easier than swapping the PSU). I have an occasional problem where my screen will cut to black after switching in and out of a game and back into windows multiple times. Although I never have a problem at startup, as he is describing, it could be a similar issue. My issue is caused by a crappy old video card and multiple changes in resolution (from a game *usually 1024x768* to windows *1600x1200* and back again).

 

[Drez143]

I personally would fear its the video card unless it is a short. But, I dont understand why if it was a short that reseting the computer and the problem goes away. If capicators were being shorted then maybe, but that would be a crab shoot.

On my system personally I tried I believe two other PSU's. It actually wasnt that hard because I didnt actually remove my PSU. I just unpluged it to the MB and main components, and then rested the new one on top of the case and pluged it into the MB, Video Card, and HD. I still was getting the same problems with video on cold boot.

I dont get the video out problem though if my computer has been on for a while and then I turn it off and turn it back on. It only occurs off of a straight up room temp cold boot. I am a firm believer that its the video card, but ATI would have no belief on that.

 

[oops]

Hi Everyone,

Thanks for all the replies!

As this is urgent, I'm going to get the Barracuda since it seems reputable unless someone has had a really bad experience with it.

As for the PSU, how's the Antec one? Here are the specs for the Antec PSU http://www.antec.com/specs/TPII430_spe.html. Or should I get one that's a bit more powerful? I don't really know how to read the specs but I think it's 17A on the 12V? I'm going to get another PSU anyway because the one I'm currently using is unmarked. I don't even know the amperes! lol Better get it right this time grrrrrrr

And drez, to be honest, I wasn't going to post but I got sick and tired of a seemingly malfunctioning computer and I needed to choose a new HD. The Maxtor is really fried.

 

[alcattle]

The PSU has 2 +12 rails, both 17A. It will run a 9600 NP. Until you get to the latest GPUY like 8800 should you worry.
Both solid choices, GL on the repair

 

[oops]

Ok, I rushed off to the store after work to get the new Seagate Barracuda HD. I should add that my old PSU has the following specs:

465W
AC Input 115/230V ~50/.60Hz, 10A ~7A
Fuse Rating: 7A/250V

Max Values

+3.3V: 38.8A
+5V: 44.5A
+12V: 22A
-5V: 1.5A
-12V: 1.5A
+5VSB: 2.5A

Then I did the following tests:

Test 1:
- PSU (old), MB, 1 stick mem, video, HD (old Maxtor)
- result: nothing. Screen doesn't show anything.

Test 2:
- PSU (old), MB, 1 stick mem, video, HD (new Seagate)
- result: nothing. Screen doesn't show anything.

Test 3:
- PSU (backup), MB, 1 stick mem, video, HD (old Maxtor)
- result: nothing. Screen doesn't show anything.

Test 4:
- PSU (backup), MB, 1 stick mem, video, HD (new Seagate)
- result: could not perform test because the backup PSU does not have the SATA power connectors.

Test 5:
- PSU (old), MB, 1 stick mem, video
- result: nothing. Screen doesn't show anything.

Test 6:
- PSU (backup), MB, 1 stick mem, video
- result: nothing. Screen doesn't show anything.

Hypothesis:
- the computer's inability to startup on the first try has nothing to do with the HD or the PSU

Conclusion:
- the computer's inability to startup on the first try has everything to do with the MB? Please let this be extremely false! :?

Can someone help me with the conclusion? I mean, it's gotta be pretty hard to damage a MB while all the other parts escape without so much as a scratch, right? Help... I'm really melting... sniff I don't think I can afford a new MB!

 

[will14]

You can buy a cheap mobo for that processor.
Also it's ez for a mobo to go with nothing else going.
although it sounds more like video.
forgt the rest of my point.
Without looking I'll guess you can get a mobo with sata etc for around $50.
mebe look for one that can take ddr and ddr2 and core 2 compatible.
I dont know if this exists but I know amd had a similar phase between the 2.
then you could throw a core 2 and dd2 in when you wanted to upgrade.

 

[fueled_by_ramen]

At least now we know a bit more. Like I said, trying different parts and combinations of those parts is the best way to figure out the problem if it isn't clear cut. Looks like you did a great job trying each combination.

Just in case, if you wanted to you should be able to bring the Seagate back and get a refund, since it doesn't seem like the old Maxtor had any shorts. But from what I was gathering you pretty much needed a new HDD anyway.

By the way, is the computer POSTing at all? Do you hear your usual beeps when you first turn it on? Or different beeps or nothing at all?
If not, or if the beeps are different, it might be the RAM.

I don't think that MoBo has onboard video. So if at all possible, could you find an AGP card to swap in place of your 9600? There's a chance that it's the video card (as I mentioned earlier). Do you know someone with an AGP system that would let you borrow their card? You can also try to buy a cheap low-end GFX card and just return it afterwards. I know, not the most honest method but for the sake of trying to figure out for sure what's causing the problem, it's worth a thought.

 

[oops]

Well, I was thinking of doing that. I called up a friend but we ran a few more tests over the phone. This time, I took out everything. The only things plugged in are the MB and the PSU. Nothing. Absolutely nothing. No beeps, nothing. I guess that pretty much solves the problem? Somehow, I'm still hoping for a miracle... 3 years... first the HD and now the MB too. I could really use a hug right now.

Anyway, anyone have any theories as to why this happened? I don't want to buy a new MB and have that one fail in 3 years. I can't believe I'm saying this but even a HP or Dell MB & HD would last longer than 3 years. Why??

 

[fueled_by_ramen]

If absolutely nothing happens, no beeping or anything, then it really is starting to look like the MoBo. I know you don't want to hear that. Still can't be 100% certain, but the way things are looking, it seems more and more likely.

But, as someone else mentioned, you can get a budget MoBo for about $50 on Newegg.com. Personally though (and I'm only saying this because it would suck to get a new MoBo and have the same thing happen) I would look into either testing that PSU or getting a new one. Like I mentioned in my first post, and as others have said, it is possible for a faulty PSU to short out a MoBo or GFX card or other component. It's not very common anymore, but can still happen. So one scenario would be that the PSU is faulty and caused a short in the MoBo. Or it could have just been the MoBo calling it quits. Without at least testing your current PSU, there is a risk that it would happen again. Just wanted to throw that out there.

Worse case scenario, if your budget is tight (mine sure is so I can sympathize) you can return that Seagate HDD and get a new MoBo instead. And if you can find that PSU tester I showed you earlier (usually about $10-12) you can check your PSU for any problems. Heck, you could even return the tester afterward too I guess. It might be good to test the PSU first, because with things like PSU's, some stores don't accept returns on opened PSU's. I know CompUSA doesn't.

As for why it happened, they are man-made parts...sometimes a component, like a MoBo can last for years and years...sometimes it craps out after 6 months.

Computers are just like anything else. Cars for example. Things break on cars all the time. It could be an inexpensive KIA or a luxury BMW...things break.

Sorry I couldn't be of more help.

 

[oops]

Ah swell. Well, I took your advice and went looking for a cheap MB that I can use. Only problem is, not many stores offer MBs for socket 478.

This is turning out to be a pretty bad week. I really don't know. I have an Intel P4 2.60C GHz 800MHz. Know any MBs that also have a 24 ATX-pin and SATA 3Gb/s? My RAM is also DDR400MHz PC3200. I feel... outdated... using an ancient P4. Computer technology sure changes quickly. Any ideas?

 

[alcattle]

This was the only one I came up with. Not cheap $80 but I think it meets all your needs. A little different the the MB you have now.
ASUS P4V800D-X Socket 478

 

[Pinoy]

Hmmm. I have the same problem as you guys, my box always had this "issue" since a few months after I built it years ago.

My computer:
p4 2.8e
asus p4p800 se
1 gb of kingston ddr400
6600 gt
antec 480w modular psu
1 80gb + 160 gb WD hard drives

Of course there's a little twist, most of it has been replaced since I've built it. The psu I had originally blew and took the original motherboard with it, I got the board rma'd and I bought the 480w psu. I've replaced the original 9800 card with a 6600 gt and added the 160 gb drive.

It still doesn't boot when I hit the power button sometimes, the lights come on the fans spin but the monitor doesn't click on and no post beeps. Only thing to do is reset and it boots like it should, does this like 2/5 boots it seems.

The only thing that's original in my system is cpu and ram, so I'm gonna go out on a limb and guess it's either the ram or the motherboard design. After all, we all seem to have Asus Pxpx00 boards

 

[techgeek]

Well first off, you said about three years, is it less or more than 3 years. If you really think it the motherboard, and it's less than 3 years, you can RMA that board, ASUS boards are warrantied for 3 years (hope you have the receipt).

One thing you may want to check with your motherboard is all the electrolytic capacitors especially around the VRM. Check to see if any of them look swollen or have leaked any of the electrolytic fluid onto the motherboard. As these caps age, the electrolytic tends to dry out. The caps start to acting more like resistors (conduct electricity rather than store it ie. more leakage current), resistors generate heat when current passes through them. Since they are essentially sealed aluminum cans, the heat causes them to swell, sometimes they will breach and leak. Sometimes they will actually pop. I have seen this happen in the lab, when someone hooked it up in their breadboard backwards (yes electrolytic caps are polarized), backwards the cap acts like a small resistor, heating very quickly. The person I am thinking of caught the can right in the forehead (he was leaning over the board when he powered his circuit up). It was damn funny. There was a rash of bad electrolytic caps a few years back (maybe even still some in the channel when your board was manufactured), this was a result of industrial espionage. One company had a formula for an excellent electrolytic solution. Another company stole it, but didn't get the whole formula. They produced caps at a very cheap price (which of course made this catastrophe wide spread), but without the full formula these caps had a very short lifespan. IEEE magazine had a write-up on it a few years back. Many motherboard manufacturers along with many consumer electronics companies were affected by this.

Next, I wouldn't go by any results you obtained with that backup PSU, since it is pretty low powered (not much current output on the 12V rail). I am not saying that for sure you don't have a faulty motherboard. You said you called a friend, does he have a computer, and is he local. It shouldn't be much of an imposition for him to let you use it for testing. This way your are using a known good PSU for testing. I can't say for sure that your mobo would beep code without graphics card or RAM. I would think at the very least it would need a stick of RAM, but maybe not, as firmware should run directly from the EPROM. I assume you still had CPU and the 12V CPU power connector plugged into the motherboard, cause it definitely won't beep if there's no CPU (cause something has to execute the firmware).

Hope some of this helps, sorry for the history lesson, but it may prove relevant. Good luck.

 

[oops]

Well, I don't know... all ASUS mb problems lol. But I do know that the HD failure is partially my fault. Apparently, I'm not supposed to connect the SATA power coupling and the regular one at the same time. Hence my name.

Oh I did some checking on Intel processors. I didn't even know their dual cores are so varied now. I'm so behind... Apparently, the specs for each CPU is so detailed! I never checked this stuff before when I first built my current system and I think I'm lucky my machine is partially functioning at all. Mostly lucky guesses. So much researching, checking, comparing, and double checking before one builds a custom rig...

Speaking of researching and checking, I have a mini question. In my MB manual, it says I require a minimum of 350W. However, it doesn't say if this figure includes the CPU or not. I did a quick calculation as follows:

MB: 350W
Fan Heatsink (CPU): 26.4W
Network Card: 3.25W
Radeon 9600XT: max 24.76W

Total without CPU: 404.41W

CPU: 86W (Does this figure look right???)

With the CPU: 490.41W

I still haven't factored in the power consumption of my DVD-RW, CD-RW, memory, and hard drive!! I'm using a 430W PSU and it looks like I don't even have enough power to begin with! Unless, I calculated everything incorrectly :? Gee... just how much would a 500W+ PSU cost?? No newegg for me. Can anyone shed some light on power consumption stuff for me?

BTW, RMA was the first thing I considered.

PS Oh wow... Intel chips can use as much as 150W and even higher! Scroll down to read the SilentPC review (http://en.wikipedia.org/wiki/Thermal_Design_Power)

 

[oops]

And if you can find that PSU tester I showed you earlier (usually about $10-12) you can check your PSU for any problems.

Ok, I bought the PSU tester and everything lights up green. No problems. Now what do I do? It's like this one big roadblock. How am I supposed to replace the parts if I don't even know what caused the problem in the first place?