Jump to content

Help for a semi-dead PC


JorgeA

Recommended Posts

One of my computers is a Hewlett-Packard dx7500 Microtower that I bought on clearance at an unbelievable price about three years ago. I'm running Windows XP Pro, Vista Business x86, and Netrunner 12.12 on it. With an add-on graphics card, it serves mainly (and 24/7) as a participant in a distributed-computing (DC) project, but I've also used it to get a taste of both Linux and XP.

 

A couple of days ago, the computer -- at the time running on Vista -- finished a stage in the DC project and I rebooted into XP to see what updates (Windows Defender, MS Office) there might be waiting for it. I installed the updates and left it on overnight, as usual.

 

The next morning I went into my office and noticed that the computer was turned off. Nobody other than me goes into the office, so I knew that no living being had shut it down. It used to go off and not come back on after a power outage, but for several months now I've had it on a UPS battery so it should not have turned itself off even if the power had gone out (which it had not, in any event).

 

I hit the power switch as normal. Imagine my surprise and alarm when -- nothing happened!! No matter how hard, how many times, or for how long I pressed the switch, it would not come on. Time for Web research and diagnostics.

 

I plugged the power cord into a different outlet. Didn't help.

 

I opened the case and dusted it. (There WAS a lot of dust inside. :blushing: ) No dice.

 

I removed the graphics card. Still nothing.

 

Changed out the CMOS battery. Nada.

 

The only sign of life the PC is giving, is that the green LED at the back of the case, up where the power supply is located, is still glowing. The glow seems to be fainter than it used to be, but I can't be sure of that. However, nothing else seems to be doing anything: neither the fans nor the hard drive appear to be spinning up. There is no POST and no beeps.

 

My Web research seems to point to either the power supply or the motherboard. Two questions:

 

1. Is there any way to test these, to determine which one might be at fault? (Hoping it's the PSU that's having the problem, and not the mobo.)

 

2. Any other possible candidates for being the cause of the problem?

 

What do you think? Any suggestions that might lead to a solution are welcome!!!

 

--JorgeA

 

Link to comment
Share on other sites


Open the case and look for bad caps, just in case (you know, burst electrolytic capacitors).

If you don't see any, then buy or borrow a used but known-working ATX power supply (that computer seems to use a common, garden-variety, 300W el-cheapo power supply, so anything 300W or more should do) and replace the power supply. If the computer springs to life, then you may discard the old power supply and call it a day, or even consider giving your machine a higher-end, new, power supply. If replacing the power supply gets you nowhere, then you do have a problem. My 2¢ only, of course.

Link to comment
Share on other sites

Any other possible candidates for being the cause of the problem?

No, the usual suspect here is the PSU.

I'd think a MoBo with leaked capacitors would still show some signs of life, but it may depend on which ones are busted and how bad.

Link to comment
Share on other sites

If you don't have a power supply tester, you should be able to just hook up another PSU and see if it powers on. You might be able to do this by simply unplugging the current PSU from the board without removing any components... Although I do not know how that looks on the inside. When I do this, I also will unplug the HDDs from their power just so that if the PC does turn on, I can shut it off without having to worry about an OS being in mid-boot.

Link to comment
Share on other sites

Thanks, guys.

 

I inspected the capacitors (viewed some images on the Web so I'd know what signs to look for) and everything appears normal.

 

@dencorso, you specified I should try a "used but known-working" PSU. In our situation, is that preferable to a brand-new PSU bought at the store?

 

@Trip, just to make sure -- so I would be disconnecting the old PSU but leaving it in place, then on the operating table connecting the new PSU to the motherboard as a test, right? I would connect to it everything that's currently connected to the old PSU, except for the HDD?

 

@Ponch -- this is reassuring. I already had to replace a PSU on a different machine so I have a little experience with the process. (In that case, the LED wasn't even lighting up.) Plus the repair would be a lot less expensive than a new mobo or (God forbid) a new CPU.

 

--JorgeA

 

Link to comment
Share on other sites

@Trip, just to make sure -- so I would be disconnecting the old PSU but leaving it in place, then on the operating table connecting the new PSU to the motherboard as a test, right? I would connect to it everything that's currently connected to the old PSU, except for the HDD?

 

I just plug into the motherboard and hit the power button. I guess I flubbed with that part about the HDD, I do then test with everything but the HDD plugged in before going through the trouble of swapping the PSUs out. Basically, it is wasted effort to swap the PSUs only to find out that wasn't the problem.

Link to comment
Share on other sites

Supposing that I get a new PSU to test, HP has thrown me a curve ball by offering four (!) PSU models, the only difference being that two are "with power factor correction (PFC)" and two are "without" this PFC, whatever that means. Then one of each type (with/without PFC) is "RoHS: COMPLY_2.03" while the other one of each type is "RoHS: COMPLY_2.04".  :wacko:

 

I tried looking up this PFC concept but I couldn't make heads or tails of what they were talking about in the Wikipedia article. How's one supposed to decide from among them??

 

The part number for the PSU (by Bestec) that's currently installed doesn't match any of the numbers on the HP parts list for this computer, so I can't use it to try to match what I currently have. Looking up the PSU by the Bestec model number on Amazon (where BTW it's MUCH less expensive than if bought via HP) didn't solve the PFC/non-PFC mystery.

 

How to choose between PFC/non-PFC, and RoHS COMPLY "2.03" vs. "2.04"?

 

--JorgeA

 

Link to comment
Share on other sites

@dencorso, you specified I should try a "used but known-working" PSU. In our situation, is that preferable to a brand-new PSU bought at the store?

 

No. Not just for testing. ATX power sources are pretty much the same since at least 2008 and an used but working el-cheapo unit should be ridiculously cheap or maybe even gotten for free if you look around enough, and is great to have around for testing (I have two 450W no-brand for that). If you cannot get them locally, you'll still can find them at eBay or from recycledgoods (not so cheap, though, but probably more reliable).

Now, if you intend to keep the machine going after you ascertain it's OK and undamaged, after, say, having it work normally on the test power source for at least one week then you could consider spending more on a Cooler-Master, Seventeam, Huntkey, Corsair or even HP power source for the definitive replacement (I'd say any of those other I named ought to give you much more bang to the buck than HP, though).

Link to comment
Share on other sites

Supposing that I get a new PSU to test, HP has thrown me a curve ball by offering four (!) PSU models, the only difference being that two are "with power factor correction (PFC)" and two are "without" this PFC, whatever that means. Then one of each type (with/without PFC) is "RoHS: COMPLY_2.03" while the other one of each type is "RoHS: COMPLY_2.04".  :wacko:

Don't pay attention to anything HP offers you. Also don't worry about RoHS compliance, that isn't something the end-user has a care about. The only thing that matters to you is the wattage marked on the PSU you already have. If you determine the power supply is at fault, then you will need to get a new one that is at least the wattage of the one that is installed.

Now presuming this is what the inside of your PC looks like:

http://www.citilink.ru/_catalog_images/516568_v04_b.jpg

And this is what your board looks like:

http://www.manfungtech.com/image/cache/data/hp/pc/487741-001-800x600.jpg

Then your replacement PSU will be an ATX that has the 24 pin and the 4 pin motherboard connections. Some PSUs may have a 6 or 8 pin instead of the 4 so pay attention to pictures. You also want to take note of the amount of connections to other devices like drives. Make sure your replacement PSU has enough connectors to power your drives and whatever else.

Be aware that there may be an adapter plate on the PSU if the case is some custom thing.

Link to comment
Share on other sites

A lot of good information, @dencorso and @Tripredacus -- thank you!

 

I'm thinking of getting this unit from Amazon.com to try it out. (Funny, the price seems to have jumped from US$35 to $40 since yesterday. Maybe they know I'm in the market for one...)

 

The description for that PSU says:

 

Connectors: (2) SATA connectors, (5) peripheral power connectors, (1) Floppy connector, (1) 24-pin ATX connector, (1) ATX12V Connector

 

I see the 24-pin connector mentioned in there. Would the ATX12V connector be the same thing as the 4-pin connector? But none of the reviewers seemed to have a problem hooking it up, so we should be good to go.

 

One general question: are PSUs pretty standard in their shape and the location of mounting screws, etc.? I guess my question is how easy it is to get a PSU that will physically fit into its allotted place inside a PC case.

 

--JorgeA

Link to comment
Share on other sites

RoHS (just for the record) is IMNSHO one among the stupidest standards EVER made:

http://www.rohscompliancedefinition.com/

http://en.wikipedia.org/wiki/Restriction_of_Hazardous_Substances_Directive

http://en.wikipedia.org/wiki/Restriction_of_Hazardous_Substances_Directive#Criticism

 

Basically with the "noble" scope of freeing us from lead poisoning (something that may happen if you suck on a PCB :w00t: 24/7 for several months/years) they managed (while reducing the amount of lead used in industry of a very negligible quantity, something like maybe 1.50% or 2.00%) to raise by 15 to 30 degrees the temperature of soldering alloys (which causes a number of serious issues, particularly shortening lifetime of electronic components, increase NOTICEABLY cases of cracking AND making repairs a little bit more difficult):

http://en.wikipedia.org/wiki/Solder

curiously, the military, the medical and the aerospace/data transmission industries are largely exempted, but the differences between RoHS 2.03 and 2.04 are seemingly two version that HP "invented", it is possible that the former is an old version, and the latter is "latest" 2011 version or that the former is the 2011 "current" version and the latter is one of the "amendments". :unsure:

These amendments, particularly, are interesting:

http://eur-lex.europa.eu/legal-content/EN/TXT/?uri=celex:32014L0013

Substitution or elimination of lead in the populated printed circuit boards of mobile medical devices is currently technically impracticable. A temporary exemption to allow the continued use of lead solders is needed until further research has been carried out to identify alloys that are reliable for the normal life of mobile medical devices.

 

http://eur-lex.europa.eu/legal-content/EN/TXT/?uri=celex:32014L0012

 

Strong vibrations occur in Positron Emission Tomographs which are integrated into Magnetic Resonance Imaging. Research has shown that lead-free solders that have been investigated for vibration susceptibility are more vulnerable to early failure under the severe vibration conditions than bonds made with tin/lead solder. The specific conditions and geometrical restrictions of the equipment restrict the applicability of mechanical measures that could eliminate or sufficiently mitigate the effects of the intense vibration.

(3)
Substitution or elimination of lead is currently scientifically and technically impracticable. A temporary exemption is needed to allow manufacturers sufficient time for research to identify suitable lead-free materials and designs

 

.

As I see it, it is a very good thing that "important" things are made with reliable, proven to be working fine, materials, but I completely fail to understand why my laptop actually *needs* (by Law) to be made with less reliable (and actually more costly) materials.

 

jaclaz

Link to comment
Share on other sites

One general question: are PSUs pretty standard in their shape and the location of mounting screws, etc.? I guess my question is how easy it is to get a PSU that will physically fit into its allotted place inside a PC case.

 

In general, yes. Very rarely there may be some low profile PSU, but if you are unconfortable with that, you can measure and post the dimensions of your PSU. As for the mounting screws, they are always in the same places, AFAIK, so don't worry about that.

 

They're both el-cheapo, really, but since it's 6 of one or half a dozen of the other I'd go for this one, because it's cheaper and 400W. In case your issue really is the PSU, you could then mount the 400W instead for the duration, and you'd be better off than you started, but if not, you'd have a better test PSU at hand, IMO, and would have spent $15 less. @Trip & @jaclaz: please do chime in about this.

Link to comment
Share on other sites

Connectors: (2) SATA connectors, (5) peripheral power connectors, (1) Floppy connector, (1) 24-pin ATX connector, (1) ATX12V Connector

 

I see the 24-pin connector mentioned in there. Would the ATX12V connector be the same thing as the 4-pin connector? But none of the reviewers seemed to have a problem hooking it up, so we should be good to go.

 

One general question: are PSUs pretty standard in their shape and the location of mounting screws, etc.? I guess my question is how easy it is to get a PSU that will physically fit into its allotted place inside a PC case.

 

--JorgeA

The ATX12V likely means the 4 pin BUT could mean other size. I, personally, would only purchase one with a photo to make sure it is the right one. Even in their description of "peripheral power connectors" is not clear. You can see in the picture, a blurry white connector, which is the standard molex connector and it is probably what they mean.

I wouldn't use Amazon, I'd use Newegg. They have better pictures:

http://www.newegg.com/Product/Product.aspx?Item=N82E16817371033

That is an example.

ATX power supplies are all the same size. As noted before, some cases are not standard and require an adapter plate. If this is the case with yours, when you remove the PSU, the plate would be attached to the PSU already. There are other form factors like Dencorso mentioned. Example, see this one's size and shape: http://www.newegg.com/Product/Product.aspx?Item=N82E16817151090

As far as recommendations on which to buy for testing, I couldn't say. I just have PSUs laying around I can test with. If I were to buy a PSU for replacement, I have my own favorites (Antec) but as long as it works, I'm fine with it. I'd probably say you should get an 80 PLUS (Efficiency) model rather than one without it.

Link to comment
Share on other sites

@Trip & @jaclaz: please do chime in about this.

 

 

The ATX is actually (as "form factor") an "industry standard", it's dimensions (and location of screws, etc.) are documented:

http://www.pcguide.com/ref/power/sup/form_ATX.htm

 

so, instead of posting the dimensions of the (stupid) HP "original" PSU, it would be easier if JorgeA could check it against the above specs.

 

It would be NOT the first time that HP uses a non-standard case/PSU, but most are actually "standard" units, and in this specific case, the unit seems really "plain enough":

http://h10032.www1.hp.com/ctg/Manual/c01570902.pdf

http://www.impactcomputers.com/463318-001.html

http://www.thebestpartinc.com/463318-001.html

 

What would I take more care about is the actual power rating on the single lines, over the years I have seen PSU's with every kind of "queer" combination of volts/amperes that although "summing up" to the "label" 300 W were (sometimes) very different on the single line/voltage. 

 

 

Of course if you "step up" and go for a 400 W unit the above potential problem should "vanish".

 

As well (but again this is not the case -pardon me the pun - since it is a "micro-tower" unit) there may be some issues with actual length of cables.

 

It seems to me (but *needs* to be double checked):

http://www.ebay.ie/itm/Hewlett-packard-463318-001-Power-supply-300-Watts-/371156353028?pt=PCA_UPS&hash=item566aa48004

like the HP 463317-001/463318-001 (PFC/No PFC) is exactly the same as the HP 5188-2625, i.e. any among:

  • LiteON PS-5301-08HA
  • Delta DPS-300AB
  • Bestec ATX0300D5WC
  • HiPro HP-D3057F3R

were "original OEM's" for it, but *any* no-name (and el-cheapo) unit should do nicely.

 

jaclaz

 

P.S.: As a side note, the way most brands/firms name their parts/spare parts deserve a special mention in the annals of stupidity :yes:.

Link to comment
Share on other sites

Hi guys,

 

Swapping out the HP PSU will do the trick most likely indeed. The PC has a normal ATX PSU that was produced between 2008 and 2009. It's totally normal for PCs bought in the box-pusher-store to only last for a few years so it most likely died of age.

 

I would swap it for something in the 40USD range, like a Corsair CX430 80+ Bronze Cert. PSU (and 20USD mail in rebate if you can get it) with "free" shipping. This PSU will lower the noise level your PC is generating. It will vent your PC a lot better with the large fan (I think a 130mm one) reducing heat of the CPU as well. The best part is, your PC will use about 25-30% less energy, resulting in 25-30% less heat. Anything not 80+ cert. I would not even touch any more! This PSU most likely will outlive the rest of PC. Just make sure when you build it in to keep the fan pointing downwards for the correct airflow.

 

Let me know how it goes ;).

 

 

... no-name (and el-cheapo) unit should do nicely.

 

...

 

P.S.: As a side note, the way most brands/firms name their parts/spare parts deserve a special mention in the annals of stupidity  :yes:.

 

 

LOL! :yes:

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...