Sign in with
Sign up | Sign in
Your question

Dell XPS 400 random shutdowns

Last response: in Components
Share
February 21, 2011 4:47:46 PM

Hello all, first post, been reading in the forum looking for a similar problem but so far no success. Here is my problem hope I posted in the right area:

Running XP, it is 4 years old stock, no recent hardware or software installs besides Norton Internet Securities 2011 and an Upgrade to Itunes. I was browsing the net and all of a sudden the computer just shut down giving a yellow light on the monitor and black screen. Restarted, then in about 2 minutes, did it again. No numbered diagnostics lights are showing and when it shuts off, the power button remains solid green, the fans are running, and the green LED is working on the MOBO. When it shuts down it sounds just like when I normally power down to turn it off from windows, a small click is heard and sounds like a drive is winding down, but when I used to shut it down using windows, before it started this, the whole system would power down and the power button green light would go off. I have already replaced the old Geforce 6800 graphics card as of today, and that aint it. If I leave it sit for an hour or so, it will run for about 5 - 10 minutes boot up normally, then without warning, shut down leaving a fan or the fans running and the solid green light on the power button. If I try to start it up immediately after it shuts off it will boot but then shut off within 2 minutes while its loading the deskop. Sometimes it will shut off completely by holding down the power button, but sometimes I have to unplug it from the wall to turn it all the way off. I have already blew all dust from the inside, which I normally perform every 6 months anyway, and it is still shutting down. It was not moved from one location or another and has sat in its currently location for several months. I really only power it up on weekends and turn it off when not in use, so as far as hours of operation its not that used.

I forgot to add that the reason I thought it was the graphics card was immediately when it started to do it, and I did not wait an hour, I would restart immediately and the monitor would not get anything showing "no digital input cannot display this mode", then by the time I got the new card, after an hour while I went to the store, I removed the old and put the new, it cranked up fine and installed the new card, then I used the install disk for the radeon card, and halfway through the install, it shutdown after running for about 10 minutes.

I have booted into only BIOS, and it will still shutdown while just sitting in BIOS.

Ram tested working fine.

I sent it to a tech and he cleaned out a few cookies saying that occasionally this may cause a system to lockup and shut down. I got it back, it worked fine for 24 hours, one day, then another 15 - 16 hours another day. I then started it the following weekend, it booted, ran for 30 minutes and shutdown again, same symptoms, green power light, fans running, but everthing else is dead. I sent it back to the tech. He plugged it in and its been going strong for four days straight. He has checked PSU, MOBO, Ram, everthing is working fine and CPU running cool. Now I am at a loss

Changed the Cmos Battery, still does it.

Plugged direct into wall bypass surge protector, still does it.

Tried another wall plug, different circuit, still does it.

Unplugged everything including monitor and tried another monitor, it still does it.

Tried another ac computer cord, it still does it.

None of the Capacitors appear brownish or leaking.

Did I say, it still does it, whew, anyway wondering if anyone else has had a similar problem and if so what they did to fix it. I was wondering if the PSU could be working intermittenly or even a problem with I/O cluster where the power switch is located which I read somewhere can cause weird stuff too. Any help greatly appreciated, I hate to start spending money on being a parts changer and it still not work, I have read people with similar problems who changed out their PSU and it did not work.

a c 713 ) Power supply
February 21, 2011 8:59:34 PM

If swapping PSU's does not work I would guess motherboard.
m
0
l
February 21, 2011 9:29:38 PM

Thats kinda what I was thinking too I just never had a computer do this, u think a PSU can do fine then sometimes cause problems, I would have assumed once it starts going bad it would just get worse, not work good then go on the blink every once and awhile. Btw it is the stock Dell 375 watt that came with it, and I have already been told the ATXs will work in the XPS400.
m
0
l
Related resources
a c 713 ) Power supply
February 21, 2011 9:35:03 PM

Looking at your original issues it sounds like temp issues but it does not make any sense that they just go away for a few days unless there is a fan (CPU fan, Chipset fan, PSU fan) somewhere cooling something that does not turn on all the time.
m
0
l
February 21, 2011 9:38:22 PM

I know its wierd, and I left the case open several times, and all three fans are running when it shuts down.
m
0
l
a c 713 ) Power supply
February 21, 2011 9:44:49 PM

Maybe applying new thermal compound on the CPU would be in order. Have you monitored your temps?
m
0
l
February 21, 2011 9:57:16 PM

The tech said he did and he said it was running cool. I did read something in the Dell forums about those stock PSUs acting strange and causing a power shut down on at least one XPS400, he changed it and it started staying on.
m
0
l
a c 713 ) Power supply
February 21, 2011 10:04:38 PM

PSU's can have a hard time delivering their rated capacities when they age, it could very well be the issue in your case especially if it is confirmed everything is running cool.
m
0
l
February 21, 2011 10:09:19 PM

The tech wanted to recreate my setup and see if he could get it to shut off so I had to bring my monitor and the power card for the computer to him. It would be nice if it would shut off while he's got it.

If I go ahead and decide to try a new PSU does anyone have any suggestions on a good ATX replacement, I have never really looked for power supplies before, I am not adding anything else to the system so I guess anything over 400 watts would be fine, trying to watch the price factor of course also.
m
0
l
a b ) Power supply
February 22, 2011 4:39:07 AM

copotay said:
The tech wanted to recreate my setup and see if he could get it to shut off so I had to bring my monitor and the power card for the computer to him. It would be nice if it would shut off while he's got it.

If the supply failed months ago, then a tech can see that problem immediately using a multimeter even if the computer does not crash all day. Yes, a failed supply can boot and run a computer for months.

What causes power off? A power controller. It has numerous inputs. Suddenly power off can be due to any of those inputs. Or a defective power controller.

A tech views simple numbers with a meter to see the problem immediately. About 1 minute of labor. Or one just replaces good parts until it starts working again. Or works again and the same failure appears next week.

Never select a supply on watts. In part because one 350 watt supply is also rated at 500 watts by some other vendor. Each is reporting numbers honestly. Watts is why so many 300 watt computers have 800 and 1000 watt supplies.

Your concern is amperage for each voltage. A new supply must have similar numbers. Or have amperage numbers consistently higher for all voltages. You can do this because the supply was selected using engineering reasons. So amperage numbers on the original supply reflect real world concerns.

Select based upon amperage. Watts and dollars are how others without any electrical knowledge recommend a supply.

m
0
l
February 22, 2011 11:52:46 AM

Tks very much for the detailed information, so basically, regardless if the system is starting and or running without kicking off, the tech should be able to trace down where the power is creating the problem whether it be the PSU or power controller on the mother board.

Also that is very good to know about the amps instead of volts or watts, so I will be looking for the output amperage as listed like this below which is behind the voltage amount and of course I will need to find out the output amps of the current dell PSU to match them up

+3.3V@20A, +5V@20A, +12V@30A, -12V@0.8A, +5VSB@2.5A

Very helpful nfo tks alot but I guess I will wait to see if the tech can track down the problem, again he is telling me everything checks out so I might have to question him for more detail to see if he is meter checking all of it even though the system is not shutting down.
m
0
l
a c 243 ) Power supply
February 22, 2011 12:10:38 PM

copotay said:
I will need to find out the output amps of the current dell PSU to match them up

+3.3V@20A, +5V@20A, +12V@30A, -12V@0.8A, +5VSB@2.5A

That's pretty close

+3.3V@17A, +5V@22A, +12V1@18A, +12v2@18A, -12V@1A, +5VSB@2A
30 amps combined +12v output
You don't have to match that, the Antec380 makes a fine replacement
http://www.newegg.com/Product/Product.aspx?Item=N82E168...

Sounds more like caps going bad on the MB, than psu though.
They don't need to be brownish or leaking to be bad, simply bulging is enough.
m
0
l
February 22, 2011 12:22:15 PM

Tks for the suggestion, oh and I know, I hate to start buying stuff and replacing and it goes bad anyway because of a tiny little capacitor on the mobo
m
0
l
February 22, 2011 4:24:22 PM

What does anyone think about Active PFC PSU supplies being plugged into surge protectors or UPS devices having trouble with step approximated sine waves sometimes causing the system to shut down or reboot which could be why it is still running at the tech's shop. I do beleive the dell PSU in it right now does have Active PFC. I wonder if having it plugged into the Belkin BE108230-06 surge protector, which does not say if it supports PFC, caused an eventual problem since it has been plugged for a few months. Although I tried plugging it into wall outlets after it kills, I did not wait for several hours or days in doing so like once the tech goes to plug it into his ac outlet, which overtime may have corrected the problem until I again plugged it back into the Belkin. I know I may be talking long shot here but I just cant figure out why he cannot get it to shut down or find any voltage/amp probs.

I will tell everyone this about this Belkin product, I bought one for my Sony Wega 50 inch at the same time as the one for the computer, and from when it was new, occasionally I will have to unplug the entire Belkin surge protector because it will not turn on the television. Hitting the power buttons will not turn it on until I unplug the Belkin and then replug it back in. Before unplugging it, everything else plugged into it works, ps3, DVR, DVD player, just not the Television. Belkin could not explain why so they sent me a new one and it does it too. Oh and unplugging the Tv from the Belkin, then back into the Belkin did not work either, I had to completely unplug the Belkin.

So I wonder if the television was doing this what was it doing to the computer. If my tech still can't figure it out, I will try direct to wall when I get it back with the Belkin and see what happens.
m
0
l
a c 243 ) Power supply
February 22, 2011 4:28:28 PM

copotay said:
I do beleive the dell PSU in it right now does have Active PFC.

It doesn't
m
0
l
a b ) Power supply
February 22, 2011 5:36:12 PM

belkin is usually pretty junky quality...


btw, i once had a customer bring in a dell xps 400 to the computer repair shop i work at.
they said that nothing shows up on the screen. turns out the onboard graphics chip was shot, and adding in an old ati radeon x600 fixed the issue.

also seen dell xps 400 with intermmitent usb ports, graphics output getting grainy randomly, but no visibly bad caps...
m
0
l
February 22, 2011 5:43:15 PM

Awwwwe, I thought I had found something that could have caused it symptoms, I think I saw that it was while I was looking at the stock replacements, but must have been mistaken.
m
0
l
a b ) Power supply
February 22, 2011 11:43:03 PM

copotay said:
What does anyone think about Active PFC PSU supplies being plugged into surge protectors or UPS devices having trouble with step approximated sine waves sometimes causing the system to shut down or reboot which could be why it is still running at the tech's shop.

A surge protector and a UPS have as much in common as a motorcycle and an 18 wheeler. Which is that Belkin?

Anything a protector or UPS would do on a computer's power cord is already done better inside a computer. A typical surge protector: take a $3 power strip. Add some ten cent protector parts. Hype it with myths (ie warranty). And sell for $25 or $60. An obscene profit center.

Read its numeric specs. Does not even claim surge protection. No problem. An overwhelming majority knew Saddam had WMDs and that a Belkin does surge protection ... because others told them what to believe.

Most UPS output power so dirty as to even confuse some PFC circuits. Power from some UPS is so 'dirty' as to maybe harm power strip protectors and small electric motors. So where is that hardware protection? Find it in hearsay; not in the hardware.

Electronics even in 1970, as defined by international design standards, makes 'dirty' electricity irrelevant. Power that can harm those appliances is also perfectly ideal to all electronics. Because electronics are so robust as to not be harmed by typical UPSes.

Normally a 120 volt UPS (such as this one that outputs spikes up to 270 volts) does its function without problems. Its function: temporary and dirty power during a blackout. Apparently your Dell active PFC is confused by what most here did not know exists. UPS output is often the 'dirtiest' electricity an appliance will see.

Belkin hopes you do not learn any of this. Obscene profits would be at risk.

Is that Belkin a power strip protector or a UPS?

Back to your computer. All computers work perfectly normal even when voltage drops so low that incandescent bulbs dim to 40% intensity. AC power must vary so great as to be potentially harmful to the refrigerator and furnace. And is perfectly ideal power to all computers due to what is inside every power supply.

Almost no electronic failures can be visually located (ie capacitors). Your solutions begin with numbers. Such as from specifications or from test equipment (ie a multimeter that any electrically informed tech would have). Without critical numbers and other facts, then posted is wild speculation or a discussion or what should exist. What you should have. Not what you do have.

Appreciate why active PFC gets confused. Some of the dirtiest power (ie that spike of up to 270 volts) comes from a UPS. Power supply routinely makes that and other anomalies irrelevant. Your computer problem would be on the other side of a power supply. Probably involving a power controller that many do not even know exists. Facts and numbers necesary to say more.
m
0
l
February 23, 2011 12:26:28 PM

Westom it is this one:

http://www.belkin.com/IWCatProductPage.process?Product_...

I will agree about the dirtyness of the output on these devices from everything that I have read about the UPS devices. Also I cannot confirm my PSU has active PFC because I cannot find any specs on it, according to Delluser1, it does not, however I had read that some people are having problems with the new dells and their UPS causing the same issues, shutdown, due to them definitely having active PFC PSUs.

I really appreciate your area of expertise, explanations, and answers, they definitely are enlightening. It is obvious you have advanced knowledge of electronics and computer technology. Although I do have some comprehension of the same, I am far from an expert in the field. My main area of skill is in criminal homicide investigations and in over 25 years of service I have come to learn that what is obvious may not always be the case which is why I look at all alternatives such as in this situation trying to find other possiblites after being frustrated that the problem is not showing itself to the tech. So please excuse me if I am asking questions that seem foolish and speculative in your mind. I now comprehend that in electronics and computer technology there are no other possibilites, there can only be one direct solution to the problem once a proper diagnosis is made with the aid of mathematics and testing.

My first computer was a Tandy, (Radio Shack for some of you that remember), that I threw away after it lasted me for years. Since, I have been through about six other computers, mostly dells, that have never given me any kinds of problems. I just sold them to upgrade to faster processors. It seems now, just like you are saying Westom, that companies are mass producing electronics in foregin countries, especially computer products, to hypnotize consumers into believing anything. I am not naive enough to believe that a company such as Dell can have quality control problems with its mass production and relatively low costs due to mass production, however you cannot leave out the factor that, not only Dell, but any company, computer part manufacturer or not, can make a concerted effort to have their products last a limited amount of time to ensure that eventually you will return and "buy another one". I have always been conservative when purchasing items expecting them to last for some time and I personally never saw the need to upgrade every few months or year just to get "the bigger, faster, or better" when what I did purchase was serving its purpose. This goes not only for computers but any products from grass trimmers to televisions. Ever heard the saying "they don't make them like they used to".

Then we come to my computer which was used occasionally only on weekends over the last couple of years because I was too busy or tired during the week from doing my real job to even turn it on. The tech even said he had never seen one that clean or well kept for its age. The limited use I did get out of it, over the period of time I had it, in my mind, should have still had no problems for the price I paid for it. We have over 400 Dells in our agency and according to our IT guys, we have had no major problems with the boards or PSUs. Again I am not ignorant to believe that electronics will and do have problems, it is just aggravating and annoying that the problem has not been found to this point which is why, Westom, I am posting to this community which does contain the experts that can explain, as you did, the very detailed information and what could be the issue. I appreciate and encourage you to continue your responses and hope you will afford me some patience until I am able to learn from your knowledge.

Tks again
m
0
l
a b ) Power supply
February 23, 2011 2:24:16 PM

Westom knows what he is doing, (in my opinion), but he is known to join random tech forums and post pretty much a huge rant about surge protectors. Oftentimes his posts lead to arguments about surge protectors.

However, there is only one flaw i find with what he writes. The part about that computer PSUs filter the dirty incoming power.

That is not always true, as a really crappy PSU like this one:
http://psurepair.freeforums.org/nscom-p4-500w-500w-revi...
has little to no input filtering components, which are responsible for cleaning up the power coming in, and also making sure that no interference goes out. a really crappy psu will not do either of those, so besides not blocking crappy incoming power, it also outputs crappy power back into the power line
m
0
l
a b ) Power supply
February 23, 2011 9:19:50 PM

shovenose said:
Westom knows what he is doing, (in my opinion), but he is known to join random tech forums and post pretty much a huge rant about surge protectors. Oftentimes his posts lead to arguments about surge protectors.

If I am posting, then others are posting popular lies and urban myths. When others are providing technical facts, then I read it and move on. No reason to post. They are posting honestly. A benchmark. Did he post without facts and numbers? Then he is probably lying.

Protectors are one of the biggest scams going. Most will post advertising lies as if that were science. Somehow that protector will magically stop what three miles of sky could not. Then get angry because they were so easily brainwashed. I grew up watching advertisers enjoy manipulating people with lies. Yes, enjoy. So easy was so many to 'know' smoking improved health. And yes, the most easily brainwashed then got angry and nasty when the Surgeon General provided reality. A reality that contradicted advertising.

Back to the discussion. Anything that a magic box might do on a power cord is already done better inside every supply. However, those so many functions routinely found in Taiwan supplies can be missing if a supply is sold to naive consumers who select only on dollar and watts.

Many functions required to be in computers exist because a computer manufacturer (Dell, HP, or a computer assembler) is responsible for meeting those requirements. Dell and HP do. But a computer assembler often does not know only he is responsible. Very profitable is to dump inferior supplies into the market. So many computer assemblers have no idea what a power supply must do. So many computer assemblers have insufficient electrical knowledge.

Ignorance creates a market ripe for dumping supplies that are missing essential functions. That sell for $20 or $45. A supply manufacturer need not meet UL, FCC, ISO, CEC, or other standards. That obligation is 100% on the computer assembler who too often has no electrical knowledge.

Anything a protector and UPS would do must already be inside every computer. An adjacent protector does what its specs say it does - virtually nothing. See that Belkin as a perfect example. A $7 grocery store protector has specs similar to what Belkin is selling for $50. Where does the Belkin even claim protection from each type of surge? It doesn’t. And does not have to. They are marketing only the most easily brainwashed. Those most easily educated only by advertising.

Active PFC circuits in newer supplies can be confused by a UPS with power that 'dirty'. A reality that exposes a widely believed myths about UPS 'clean' power. Myths created by advertising people because deceiving people is both legal and so fun. I was also entertained.

Learn realities. 'Dirtiest' power from a UPS is made completely irrelevant why what must already be inside every electronic appliance. Even most 'inferior' supplies will make crappy power irrelevant because even inferior supplies are that robust.

And finally, this rationalization of making a computer that it must be replaced every three years is why companies such as Packard Bell no longer exist. Only the dumbest among us kept buying domestic cars that were also built on that philosophy. And yes, cars in the early seventies had rust holes after three years because profits were important – the product was irrelevant. A concept based in what business schools teach. Those auto manufacturers stopped painting inside the fenders to cut costs – to increase profits.

My cars and computers easily last and are still used ten years later. I do not endorse the destruction of free market economies by buying crap such as GM or Gateway. I still have and sometimes use my oldest HP and Dell computers. Last night, I used an HP Vectra (a 486 machine) to read the NY Times and other tasks. Drove an older of my Accords (12 years old) yesterday that routinely gets more than 30 MPG in local driving and never fails. And I both laugh and cry at fools who bought Chevys or pre-2001 Fords. Laugh because they are fools. Cry because they advocated harm to the American economy and the destruction of American jobs.

I am not conservative about anything. Instead, I am informed. Learned long ago that the honest answer also always says why and with numbers. Yes, I grew up watching advertising easily manipulate the most easily deceived (brainwashed) using intentional lies. Legal is to lie in advertising. Watched advertisers so enjoy manipulating so many.

All properly constructed computers work perfectly fine in a 100 degree F room with balls of dust inside. Too many feel balls of dust are destructive because the crappy machine failed with dust inside. As long as dust filled heatsinks still have holes for airflow, then a properly designed computer works OK. That will make some angry. They cleaned dust from computers to make it work. Therefore know the above must be wrong. Another example of junk science reasoning. Dust balls identify an inferior design or a defective semiconductor. Another reason why I strongly discouraged others from buying Packard Bells, Gateways, and power supplies sold on dollars and watts.

Read spec numbers on that Belkin power strip. It acts no different from a $3 power strip until 120 VAC well exceeds 300 volts. It does no power conditioning. Even the filtering is near zero. It does not even claim surge protection. They are selling for $50 what can also sell for $7 at profit in grocery stores. Monster sells a similar product for $80 or $150 because so many are educated by advertising. Because Monster has a long history of identifying scam. Then selling a similar product at even higher profits. Most will ignore hard facts and numbers to recommend that Belkin because advertising (brainwashing) is that effective. And why advertising is so much fun.

The original problem implies a power controller powered off a machine. Why? Either a power controller is defective. Or one of its many inputs reported a failure. Solution starts by identifying a defect. Most facts can be obtained in but one minute with numbers from a multimeter. Is the power suply defective? Meter would have reported that and so many other 'options' immediately. A reason for my first post. Find the problem. Only then fix (replace) it. A UPS does nothing for the problem. Is most often recommended to do things it just does not do. That Belkin also does virtually nothing useful.

See those numbers for a power supply? Amps for each voltage? More honest answers provide numbers with the reasons why.
m
0
l
February 23, 2011 10:55:47 PM

Westom, I would like to have you and my tech hook up with each other so you can explain to him what he needs to look for or teach him a few things. I spoke to him today and passed on exactly what you said he needed to look for and he said he is getting all normal readings after the power supply across the power controller, and, that he knows how to read a meter and has repaired alot of them, he kinda got ticked off that someone was telling him how to do his job.

It still has not shut off, he has let it run for hours, turn off, rebooted, and still runs without turning off, he has had it for the last 12 days. He did say that he has seen power strips, like this Belkin, do crazy things to systems and cause problems like it is experiencing. I guess what I will do is use all those Belkins as door stops and go back to plugging directly into the wall when I had no problems, I would simply unplug for lighting storms, I did lose one like that one time, the PSU turned black and the mother board exploded inside the computer, there was no fixing that one.

Btw, I got the Belkins on sale for $11.00 about what they are worth just to have a few extra plugs for other things on one outlet.

So just to clarify, I should not use any type of power strip, surge protector, or UPS before the wall outlet?
m
0
l
a b ) Power supply
February 23, 2011 11:17:05 PM

a cheap power strip to simply give you 6 extra plugs is fine...but you're not supposed to expect amzing things from a crappy over-hyped piece of $h!t UPS...


and actually, westom, most PSUs used by dell, hp, gateway, etc, are usually pretty good. besides the design issue with the Bestec ATX-250-12E in many Gateway and other computers, the OEM PSUs always have good input filtering, large internal heatsinks, quiet fans, and decent weight. they just lack in capacitor quality and visual bling
m
0
l
a b ) Power supply
February 24, 2011 12:49:50 AM

copotay said:
Westom, I would like to have you and my tech hook up with each other so you can explain to him what he needs to look for or teach him a few things. I spoke to him today and passed on exactly what you said he needed to look for and he said he is getting all normal readings after the power supply across the power controller, and, that he knows how to read a meter and has repaired alot of them, he kinda got ticked off that someone was telling him how to do his job.

One needs no electrical knowledge to pass an A+ Certified Computer Tech test. Many techs get defensive rather than learn more. Especially when you are only repeating hearsay. And especially when a majority who recommend computer solutons most often have even less knowledge.

For example, Intel specs say 5 volts must be above 4.75 VDC. But only better techs would know why a multimeter therefore must read higher than 4.87.

Furthermore, readings must be taken with maximum load. Defects can remain hidden until a heavy load exists. Therefore a computer may work fine under all but worst loading. Many mistakenly confuse Prime95 with maximum load. Maximum load means a CPU is accessing hard drive, sound card, ethernet, USB, doing complex graphics, etc simultaneously (multitasking). Only then are voltage numbers on six wires useful.

Statement that voltages are fine filters out valuable information. Those number may contain additional knowledge. Another example of why any statement must always include why - the numbers.

Well, it is his job. He must fix it or release it. Until that happens, your information either will not be considered useful or will be regarded emotionally as meddling. At this point, what you learn here is only for future reference. Not just for fixing a machine. But for what is meant by the concept, "Don't work harder; work smarter". One reason we fix things is to better understand what is meant by, "Follow the evidence".

m
0
l
!