Boot Problems
in [Home Built PC]
|
From: Paul on 25 Jun 2010 03:13 Nospam wrote: > In message <f8FUn.25694$m87.7282(a)hurricane>, Andy > <andy(a)NOSPAMmanyplay.com> writes >> >> >> "Nospam" <nospam(a)nospam.demon.co.uk> wrote in message >> news:JhCs6cAQ6jIMFw5k(a)nospam.demon.co.uk... >>> Hi All, >>> >>> My son's world has just collapsed! >>> >>> His PC, home built has stopped working >>> After many year of use. >>> >>> He has not added any new hardware, software or updates >>> >>> It wont display anything on the screen at all. Not even the BIOS >>> name, nothing >>> >>> should it display anything if we boot it with out a hard disc and no >>> ram installed >>> . >>> I would have thought we would have a BIOS name then halt, I wouldn't >>> expect much with out ram but I would have thought we would have got >>> something. >>> >>> we have changed the graphics card as that was the only spare >>> component we have. >>> >>> Sometimes the unit switches itself off a few seconds after switch on >>> some times it stays powered but no display >>> >>> occasionally we have had a few successful entries into the BIOS but >>> mostly it is a totally blank screen >>> >>> I have measured all the levels out of the PSU and they are in spec. >>> >>> why would it turn itself off? >>> Or is the motherboard turning it off because it is not functioning. >>> >>> Any suggestions most welcome >>> >>> ASUS crosshair motherboard >>> 4GBytes DDR2 800MHz Ram >>> AMD2 CPU >>> 320 GB SATA hard disc >>> >>> >>> -- Nospam >>> >> >> Hi, >> >> The Asus Crosshair motherboard has a small lcd display mounted on the >> back plate. You can view it from the back of the PC case near the >> keyboard ps2 connector, it should give you some idea what the issue is. >> >> Andy > > The display just says > > Det DRAM > > so its detected the DRAM but thats all it says. > > I think the fact your computer switches itself off quickly, is just as informative as being stuck at "DET DRAM". On a cheap motherboard, the only reason for that to happen, would be an overheating CPU. But since your board is a $250 enthusiast board, they could easily have added other features to cause it to turn off. I took a look over here, and found one report of bad caps on a board. The board uses a mixture of polymer caps and electrolytic caps. The person reporting, didn't say exactly which ones were bulging, but there were two electrolytics. http://vip.asus.com/forum/topic.aspx?board_id=1&model=Crosshair&SLanguage=en-us There were a couple reports of dead PCI Express slots, but that is par for the course with Nvidia chipsets. I've heard of PCI Express slots dying on them before. That motherboard has an option to illuminate some blue LEDs, sort of like maintenance lights. Apparently, when you press the button, it disconnects +5VSB from the motherboard (which is normally monitored by a green LED on Asus motherboards). That suggests to me, if you were to press that while the system was running, the blue LEDs would come on, and power would be cut to everything else. Check to see if the blue LEDs are lighting up, when the power goes off. If so, it could be a defect in the LED lighting circuit, which removes +5VSB from the board. The manual doesn't describe operation of that feature in enough detail, and I'm going on an off-hand comment from a poster in the Asus forum. One problem with getting excited about the Asus LCD display, is the precision with which those codes are emitted. Looking at the page of codes listed in the manual, the codes would seem to be based in part on the Port 80 codes. But the listing isn't presented in the sequence you'd expect to find the codes, so it's hard to say how many other states it goes through, before getting to "DET DRAM". To detect DRAM, requires that the Northbridge and Southbridge are programmed, such that the SMBUS serial interface is operational. The processor uses the SMBUS (System Management Bus, 10KHz serial), to read the SPD chip on each DIMM. The BIOS gets RAM timings, to be used when the RAM is set to "Auto", from there. The BIOS also gets the claimed memory size. Most of the other resources needed by the processor, are inside the processor itself. Since this is an enthusiast board, the Vdimm regulator voltage may also need to be set. Your RAM may be 2.2V (nominal 1.8V), so chances are your son has that setting cranked to match what Crucial specified. The BIOS has an alternate means of sizing the memory, and it actually works. One person had a memory DIMM, where the SPD had incorrect information about the DIMM memory size, and the module still worked. And that means the backup sizing routine runs as well as the SPD based method, so that the BIOS will never commission memory that isn't really there (leading to a crash). So depending on how many other display states occur before "DET DRAM", would tell you whether the chipset is working and so on. But seeing "DET DRAM", means the BIOS has just entered that code. It doesn't mean the SMBUS is working, for example. Perhaps some previous routine has set up the chipset, which would be a positive sign. It would mean the board isn't entirely dead. The board is complicated, by virtue of the fact it seems to have two hardware monitor chips, many adjustable voltages and so on. A defect in one of those adjustable voltage sources, would be enough to prevent POST. The board seems to have its share of RAM compatibility issues, and depending on available RAM on hand, you might try swapping in one module of something else, to see if it'll get past DET DRAM. Even doing a test, without any DRAM present for one testing cycle, followed by plugging in one DIMM and trying again, may give different test results. Paul
From: Paul on 25 Jun 2010 04:37 Nospam wrote: > In message <hvtkg0$cf$1(a)news.eternal-september.org>, Paul > <nospam(a)needed.com> writes >> Nospam wrote: >>> Hi All, >>> My son's world has just collapsed! >>> His PC, home built has stopped working >>> After many year of use. >>> He has not added any new hardware, software or updates >>> It wont display anything on the screen at all. Not even the BIOS >>> name, nothing >>> should it display anything if we boot it with out a hard disc and no >>> ram installed >>> . >>> I would have thought we would have a BIOS name then halt, I wouldn't >>> expect much with out ram but I would have thought we would have got >>> something. >>> we have changed the graphics card as that was the only spare >>> component we have. >>> Sometimes the unit switches itself off a few seconds after switch on >>> some times it stays powered but no display >>> occasionally we have had a few successful entries into the BIOS but >>> mostly it is a totally blank screen >>> I have measured all the levels out of the PSU and they are in spec. >>> why would it turn itself off? >>> Or is the motherboard turning it off because it is not functioning. >>> Any suggestions most welcome >>> ASUS crosshair motherboard >>> 4GBytes DDR2 800MHz Ram >>> AMD2 CPU >>> 320 GB SATA hard disc >>> >> > > > Thanks for the very detailed response. > Much appreciated. > > I have answered your questions in the text but as a quick summary; > > I have tried the clr cmos button first with the cmos battery in place > and then without a batter in place but no joy. > > the battery reads 3.15 V on my Fluke DMM > > Also the beep test give the following > I fitted a speaker and get the following results > > no memory > one long beep > followed by two short beeps. > > With memory installed we just get one short beep. > > After some measurements and checking of plugs we got it to boot, we > haven't done anything really, unless we had a suspect connection. > > Also we only have two memory sticks in it, may be the other ram is dodgy > > we ran it for 10 minutes and switched it off then we had trouble getting > it to turn on. The power supply would work for a few seconds then turn off. > > I tried this for a few dozen tries > > I then unplugged 2 of his 4 fans from his 800 W PSU and was able to get > it to power up again. > > I am starting to think it might be PSU related. > > Some times it turns itself off after a few minutes of use (with the two > fans disconnected as described above). > > I don't know if the PSU is being told to power down by the CPU/Mobo or > by internal PSU circuitry. > > How do these atx PSU work > > > >> With the side off the computer, do you see the green LED glowing on the >> motherboard ? > > The LEDs that are glowing on the mobo are the 3 illuminated switches on > the mobo and one green LED > > Power > reset > clr cmos > > The green LED is near one of the slots, see next answer also. > > > Also there is a button on the rear of the unit that illuminates blue > LEDs around the mobo next to connectors, for easy of installing cards etc. > But we dont have these switched on as it inhibits booting but they do > work so power is getting to the mobo. > >> Asus motherboards have a green LED, to show the +5VSB rail >> is working. That green LED should never flash, and should remain on >> steady as long as the rear switch on the power supply is ON. > > This green LED is illuminates and is not flashing, solid green. > >> >> When you press the front (soft) power button, the power supply fans >> start to turn ? > > yes > > >> Are the system cooling fans working ? > > yes > >> That would show >> you've got some system power, but it doesn't tell you exactly whether >> all rails are in spec or not. >> > > I have used my Fluke DMM to check the voltages on the 24 way connector > and they are all in spec +- 5% > >> The next thing to check, is whether the computer case has a small >> speaker, >> and whether that speaker is connected via a connector labeled "SPKR" >> to the PANEL header on the lower right hand corner of the motherboard. >> You need that connection, to listen for beep error codes. Some expensive >> computer cases, no longer have a speaker inside the case, which is a >> nuisance in this situation. You really need a speaker for the next test. >> > > No speaker unfortunately. But I improvised with a pair of headphones, > wow it was loud, think I might be deaf now ;-) > > without any RAM fitted it is looping with > one long beep > followed by two short beeps. > > The BIOS mobo manual does not list what the error beeps mean. > > The LCD display on the rear of the unit says DET DRAM > > no other clues. > >> To remove the RAM from the system, you turn off the power at the >> back of the computer. The green LED on the motherboard should be >> off, before working on the RAM. Otherwise, there could be standby power >> still in the RAM socket. Once the RAM is out, turn on the power >> at the back again. > >> >> When you push the button in this case, if you hear two or three beeps >> from the computer case speaker (not the 5.1 speakers connected >> externally to >> the computer), then that is a relatively good sign. For the computer >> to beep, the processor needs to execute some BIOS code. >> > one long beep > followed by two short beeps. > > With memory installed we just get one short beep. > >> If, on the other hand, the "no RAM" test gives no beeps at all, >> then you've got more serious trouble on your hands. The first thing >> you'd check, is that the ATX12V 2x2 square power connector is still in >> place. > > These are in place and read 12.54V, using a Fluke DMM > measure on the four green/yellow wires > > The 8 pins inside the two 4 way connectors on the mobo all look nice > clean and very shiny > > All the crimp terminals inside both 4 way square Molex type connectors. > They look tight not loose in any way. > > No scorching or melted plastic. >> >> For power connectors, you've got the main power connector (20 or 24 pin) >> and the 2x2 ATX12V power connector. The proper ATX12V one has two yellow >> wires and two black wires. Remove the connectors and check to see if >> the pins are >> shiny. If the connectors were installed a bit loosely, without fastening >> the retention latch on each connector, the loose connection causes the >> pins to heat up and burn. Sometimes there is enough heat to melt plastic >> and deform the connector. (I've only had one melt/burn here, but it was >> a Molex 1x4 that had a contact problem, due to the manufacturing >> quality.) >> Connector damage requires replacing connectors on either end, and >> Mini-Fit Jr. pins might not be available at RadioShack for example. I >> got some at my only good electronics store in town, but they don't >> stock everything I need to do those kinds of repairs. But I did get >> enough stuff to build my load box. >> >> If you're not getting any beeps, and the two power cables are intact, >> you have a couple ways you can go, based purely on frequency of failure. >> If you had a spare power supply, you could install it and retest the >> system. Due to the "capacitor plague" a few years back, where bad >> electrolyte was used in capacitors, there have been a number of >> power supply failures. I had an Antec PSU fail here, due to bad caps. >> >> You can take a multimeter and verify the voltages on the main connectors. >> These documents will give you some wire colors, pin names, and the like >> to work with. But many people are not familiar with electronics, in which >> case, replacing the power supply and retesting is another option. >> >> http://web.archive.org/web/20030424061333/http://www.formfactors.org/dev >> eloper/specs/atx/ATX_ATX12V_PS_1_1.pdf >> >> http://www.formfactors.org/developer/specs/atx/ATX12V_1_3dg.pdf >> >> http://www.formfactors.org/developer/specs/ATX12V_PSDG_2_2_public_br2.pdf >> >> Do motherboards fail "out of the blue" ? Sure. You could do a visual >> inspection, for a burned regulator component, such as in the Vcore >> circuits around the CPU socket. Look for bulging caps, burned MOSFETs, >> or burned toroidal or square inductor packages. You can get any number >> of pre-warnings, such as unstable operation (more crashes), refusal >> to start reliably, a different "smell", a different background >> noise coming from the computer case, perhaps a slight modulation >> of what would normally be constant fan speeds. Any little hints like >> that, help when trying to figure out what's broke. >> >> If a different power supply isn't helping, then the motherboard would >> be next. Install the CPU and heatsink on the new motherboard, and >> do the beep test again, with no RAM or video card present. If it >> beeps, then you know the old processor is working in the new motherboard. >> If there is silence, then the processor could be dead. Processors >> are relatively reliable, at least compared to power supplies, and >> about the only kind of failures you'd normally run into, is a >> processor throwing errors, soon after installation. Occasionally >> a bad batch leaves the factory, but not that often. If you've had >> the processor for a while, and it was fine, it isn't likely to >> just "drop dead". (Exceptions, are applying extreme overvoltages >> to your processor, while overclocking. But the overclockers know, >> for particular processors, what is bad. For example, my E8400 is >> not supposed to be run over 1.400 volts.) >> >> At one time, AMD S462 socket processors were a bit flaky, due to >> the lack of thermal protection. Some of those would "cook", and >> when you disassembled the CPU and heatsink, you'd see signs of >> overheating. But modern processors from both companies are protected >> from overheat now, and the power supply will be shut off if the processor >> gets too hot. >> >> Good luck, >> Paul > This post was slow showing up here, so I'm answering this out of sequence with respect to the other posting. Your memory test results are interesting. With the memory removed, one long and two short, must be a "real RAM failure" code. The manual doesn't always list the beep codes properly. When you stick in one DIMM and get a single beep, that beep is the normal "speaker test" beep that motherboards give on successful POST. My motherboard here, beeps once at POST, to signify the initial testing is working. After the single beep, it still has to boot, and you can still have a functional failure after getting the single beep. But hearing the single beep is a good sign. That makes it seem like you got well past "DET DRAM", to get the single beep. Power supplies have internal protection. They may choose to switch off on overcurrent detected, overvoltage detected (such as when the regulation is failing on the power supply), or if overheating is detected. There are heatsinks, with things like rectifiers bolted to them, and one of those assemblies may have a thermistor on it, to sense operating temperature. If the supply gets too hot, it can switch off. Normally, you'd expect to have to toggle the switch on the back, to recover from a fault like that. It is even possible for a power supply to be compromised, by the voltage level on the PS_ON# signal being "in-between". PS_ON# is supposed to be a logic signal. The power supply should really threshold the signal, to clean it up. I've heard of at least one power supply, where if the PS_ON# signal is at a 1.4 volt level, the supply operates "half on". What that means is, all voltages are present, and measure correctly at the connectors. But the supply will shut off, well before reaching full load. You can take a look at your PS_ON# signal, and see if it is properly below 0.8V when ON, and properly above 2.0V when OFF. The motherboard should be able to drive it to 0.4V (well below 0.8V). The logic high level is defined by a pullup resistor tied to +5VSB. So when the supply is off, PS_ON# should be very close to being the same voltage as +5VSB is. In other words, proper drive is 0.4 and 5V (ON and OFF), versus the thresholds defined as 0.8 and 2.0 on the receiving end in the PSU. But if the power supply is "sick", then the PS_ON# fed from the motherboard could be fine, and yet the power supply still isn't stable. ---> Swapping another power supply may tell you more. <--- Since one poster in the Asus forums, managed to burn something on his Crossfire motherboard by using a 2x4 EATX power connector, if you connect some other supply to your board, continue to use the 2x2 ATX12V connector as your processor power source. The motherboard may have a plastic cap covering the other holes, and I'd leave it that way (since we don't know what really happened to that person). The plastic cap does not imply it doesn't work, but just to be safe, stick with the 2x2 kind for now. A person replying to him, suggested he plugged a PCI Express 2x4 in there, which is a possibility. Usually the shapes on the ends of the power connectors, prevent accidents like that. Paul
From: Jan Alter on 25 Jun 2010 08:38 "Paul" <nospam(a)needed.com> wrote in message news:i01kvm$j9k$1(a)news.eternal-september.org... > Nospam wrote: >> In message <f8FUn.25694$m87.7282(a)hurricane>, Andy >> <andy(a)NOSPAMmanyplay.com> writes >>> >>> >>> "Nospam" <nospam(a)nospam.demon.co.uk> wrote in message >>> news:JhCs6cAQ6jIMFw5k(a)nospam.demon.co.uk... >>>> Hi All, >>>> >>>> My son's world has just collapsed! >>>> >>>> His PC, home built has stopped working >>>> After many year of use. >>>> >>>> He has not added any new hardware, software or updates >>>> >>>> It wont display anything on the screen at all. Not even the BIOS name, >>>> nothing >>>> >>>> should it display anything if we boot it with out a hard disc and no >>>> ram installed >>>> . >>>> I would have thought we would have a BIOS name then halt, I wouldn't >>>> expect much with out ram but I would have thought we would have got >>>> something. >>>> >>>> we have changed the graphics card as that was the only spare component >>>> we have. >>>> >>>> Sometimes the unit switches itself off a few seconds after switch on >>>> some times it stays powered but no display >>>> >>>> occasionally we have had a few successful entries into the BIOS but >>>> mostly it is a totally blank screen >>>> >>>> I have measured all the levels out of the PSU and they are in spec. >>>> >>>> why would it turn itself off? >>>> Or is the motherboard turning it off because it is not functioning. >>>> >>>> Any suggestions most welcome >>>> >>>> ASUS crosshair motherboard >>>> 4GBytes DDR2 800MHz Ram >>>> AMD2 CPU >>>> 320 GB SATA hard disc >>>> >>>> >>>> -- Nospam >>>> >>> >>> Hi, >>> >>> The Asus Crosshair motherboard has a small lcd display mounted on the >>> back plate. You can view it from the back of the PC case near the >>> keyboard ps2 connector, it should give you some idea what the issue is. >>> >>> Andy >> >> The display just says >> >> Det DRAM >> >> so its detected the DRAM but thats all it says. >> >> > > I think the fact your computer switches itself off quickly, is just > as informative as being stuck at "DET DRAM". On a cheap motherboard, > the only reason for that to happen, would be an overheating CPU. > But since your board is a $250 enthusiast board, they could easily > have added other features to cause it to turn off. > > I took a look over here, and found one report of bad caps on a board. The > board uses a mixture of polymer caps and electrolytic caps. The person > reporting, > didn't say exactly which ones were bulging, but there were two > electrolytics. > > http://vip.asus.com/forum/topic.aspx?board_id=1&model=Crosshair&SLanguage=en-us > > There were a couple reports of dead PCI Express slots, but that is par for > the > course with Nvidia chipsets. I've heard of PCI Express slots dying on them > before. > > That motherboard has an option to illuminate some blue LEDs, sort of like > maintenance lights. Apparently, when you press the button, it disconnects > +5VSB from the motherboard (which is normally monitored by a green LED > on Asus motherboards). That suggests to me, if you were to press that > while > the system was running, the blue LEDs would come on, and power would be > cut to > everything else. Check to see if the blue LEDs are lighting up, when the > power goes off. If so, it could be a defect in the LED lighting circuit, > which removes +5VSB from the board. The manual doesn't describe operation > of > that feature in enough detail, and I'm going on an off-hand comment from > a poster in the Asus forum. > > One problem with getting excited about the Asus LCD display, is the > precision with which those codes are emitted. Looking at the page > of codes listed in the manual, the codes would seem to be based in > part on the Port 80 codes. But the listing isn't presented in the > sequence you'd expect to find the codes, so it's hard to say how many > other states it goes through, before getting to "DET DRAM". > > To detect DRAM, requires that the Northbridge and Southbridge are > programmed, > such that the SMBUS serial interface is operational. The processor uses > the > SMBUS (System Management Bus, 10KHz serial), to read the SPD chip on each > DIMM. > The BIOS gets RAM timings, to be used when the RAM is set to "Auto", from > there. > The BIOS also gets the claimed memory size. Most of the other resources > needed > by the processor, are inside the processor itself. Since this is an > enthusiast > board, the Vdimm regulator voltage may also need to be set. Your RAM > may be 2.2V (nominal 1.8V), so chances are your son has that setting > cranked to match what Crucial specified. > > The BIOS has an alternate means of sizing the memory, and it actually > works. > One person had a memory DIMM, where the SPD had incorrect information > about > the DIMM memory size, and the module still worked. And that means the > backup sizing routine runs as well as the SPD based method, so that the > BIOS will never commission memory that isn't really there (leading to > a crash). > > So depending on how many other display states occur before "DET DRAM", > would > tell you whether the chipset is working and so on. But seeing "DET DRAM", > means the BIOS has just entered that code. It doesn't mean the SMBUS is > working, for example. Perhaps some previous routine has set up the > chipset, which would be a positive sign. It would mean the board > isn't entirely dead. > > The board is complicated, by virtue of the fact it seems to have two > hardware > monitor chips, many adjustable voltages and so on. A defect in one of > those adjustable voltage sources, would be enough to prevent POST. > > The board seems to have its share of RAM compatibility issues, and > depending on available RAM on hand, you might try swapping in one > module of something else, to see if it'll get past DET DRAM. > > Even doing a test, without any DRAM present for one testing cycle, > followed by plugging in one DIMM and trying again, may give > different test results. > > Paul My thought about turning the computer on and it briefly starting is the notion that it has bad caps. When the PS starts a charge builds in the cap (s) to begin it running, but the leak drops the voltage to initiate a failsafe shutdown. Paul suggests that it could also be a partial failure of the mb, which sounds like it could also be possible. Sorry I don't know more to help you. Having a spare PS to try would help a little here. -- Jan Alter bearpuf(a)verizon.net
From: Robin Bignall on 25 Jun 2010 10:34 On Thu, 24 Jun 2010 03:02:46 +0000 (UTC), "Smarty" <nobody(a)nobody.com> wrote: >Paul wrote: > >> Nospam wrote: >> > Hi All, >> > >> > My son's world has just collapsed! >> > >> > His PC, home built has stopped working >> > After many year of use. >> > >> > He has not added any new hardware, software or updates >> > >> > It wont display anything on the screen at all. Not even the BIOS >> > name, nothing >> > >> > should it display anything if we boot it with out a hard disc and >> > no ram installed . >> > I would have thought we would have a BIOS name then halt, I >> > wouldn't expect much with out ram but I would have thought we >> > would have got something. >> > >> > we have changed the graphics card as that was the only spare >> > component we have. >> > >> > Sometimes the unit switches itself off a few seconds after switch on >> > some times it stays powered but no display >> > >> > occasionally we have had a few successful entries into the BIOS but >> > mostly it is a totally blank screen >> > >> > I have measured all the levels out of the PSU and they are in spec. >> > >> > why would it turn itself off? >> > Or is the motherboard turning it off because it is not functioning. >> > >> > Any suggestions most welcome >> > >> > ASUS crosshair motherboard >> > 4GBytes DDR2 800MHz Ram >> > AMD2 CPU >> > 320 GB SATA hard disc >> > >> >> With the side off the computer, do you see the green LED glowing on >> the motherboard ? Asus motherboards have a green LED, to show the >> +5VSB rail is working. That green LED should never flash, and should >> remain on steady as long as the rear switch on the power supply is ON. >> >> When you press the front (soft) power button, the power supply fans >> start to turn ? Are the system cooling fans working ? That would show >> you've got some system power, but it doesn't tell you exactly whether >> all rails are in spec or not. >> >> The next thing to check, is whether the computer case has a small >> speaker, and whether that speaker is connected via a connector >> labeled "SPKR" to the PANEL header on the lower right hand corner of >> the motherboard. You need that connection, to listen for beep error >> codes. Some expensive computer cases, no longer have a speaker inside >> the case, which is a nuisance in this situation. You really need a >> speaker for the next test. >> >> To remove the RAM from the system, you turn off the power at the >> back of the computer. The green LED on the motherboard should be >> off, before working on the RAM. Otherwise, there could be standby >> power still in the RAM socket. Once the RAM is out, turn on the power >> at the back again. >> >> When you push the button in this case, if you hear two or three beeps >> from the computer case speaker (not the 5.1 speakers connected >> externally to the computer), then that is a relatively good sign. For >> the computer to beep, the processor needs to execute some BIOS code. >> >> If, on the other hand, the "no RAM" test gives no beeps at all, >> then you've got more serious trouble on your hands. The first thing >> you'd check, is that the ATX12V 2x2 square power connector is still in >> place. >> >> For power connectors, you've got the main power connector (20 or 24 >> pin) and the 2x2 ATX12V power connector. The proper ATX12V one has >> two yellow wires and two black wires. Remove the connectors and check >> to see if the pins are shiny. If the connectors were installed a bit >> loosely, without fastening the retention latch on each connector, the >> loose connection causes the pins to heat up and burn. Sometimes there >> is enough heat to melt plastic and deform the connector. (I've only >> had one melt/burn here, but it was a Molex 1x4 that had a contact >> problem, due to the manufacturing quality.) Connector damage requires >> replacing connectors on either end, and Mini-Fit Jr. pins might not >> be available at RadioShack for example. I got some at my only good >> electronics store in town, but they don't stock everything I need to >> do those kinds of repairs. But I did get enough stuff to build my >> load box. >> >> If you're not getting any beeps, and the two power cables are intact, >> you have a couple ways you can go, based purely on frequency of >> failure. If you had a spare power supply, you could install it and >> retest the system. Due to the "capacitor plague" a few years back, >> where bad electrolyte was used in capacitors, there have been a >> number of power supply failures. I had an Antec PSU fail here, due to >> bad caps. >> >> You can take a multimeter and verify the voltages on the main >> connectors. These documents will give you some wire colors, pin >> names, and the like to work with. But many people are not familiar >> with electronics, in which case, replacing the power supply and >> retesting is another option. >> >> http://web.archive.org/web/20030424061333/http://www.formfactors.org/d >> eveloper/specs/atx/ATX_ATX12V_PS_1_1.pdf >> >> http://www.formfactors.org/developer/specs/atx/ATX12V_1_3dg.pdf >> >> http://www.formfactors.org/developer/specs/ATX12V_PSDG_2_2_public_br2. >> >> Do motherboards fail "out of the blue" ? Sure. You could do a visual >> inspection, for a burned regulator component, such as in the Vcore >> circuits around the CPU socket. Look for bulging caps, burned MOSFETs, >> or burned toroidal or square inductor packages. You can get any number >> of pre-warnings, such as unstable operation (more crashes), refusal >> to start reliably, a different "smell", a different background >> noise coming from the computer case, perhaps a slight modulation >> of what would normally be constant fan speeds. Any little hints like >> that, help when trying to figure out what's broke. >> >> If a different power supply isn't helping, then the motherboard would >> be next. Install the CPU and heatsink on the new motherboard, and >> do the beep test again, with no RAM or video card present. If it >> beeps, then you know the old processor is working in the new >> motherboard. If there is silence, then the processor could be dead. >> Processors are relatively reliable, at least compared to power >> supplies, and about the only kind of failures you'd normally run >> into, is a processor throwing errors, soon after installation. >> Occasionally a bad batch leaves the factory, but not that often. If >> you've had the processor for a while, and it was fine, it isn't >> likely to just "drop dead". (Exceptions, are applying extreme >> overvoltages to your processor, while overclocking. But the >> overclockers know, for particular processors, what is bad. For >> example, my E8400 is not supposed to be run over 1.400 volts.) >> >> At one time, AMD S462 socket processors were a bit flaky, due to >> the lack of thermal protection. Some of those would "cook", and >> when you disassembled the CPU and heatsink, you'd see signs of >> overheating. But modern processors from both companies are protected >> from overheat now, and the power supply will be shut off if the >> processor gets too hot. >> >> Good luck, >> Paul > >Paul, >I am one of perhaps several if not many "lurkers" here who devour this >forum to learn more about PCs. On behalf of all of us, I want to thank >you for an absolutely remarkable contribution you have been making to >provide clear, detailed, and thoroughly described answers to the wide >range of questions which people bring here for assistance. > >Your range of knowledge is extremely impressive, and you have helped me >personally in the past with similar examples of extremely insightful >information, always focused directly on the pertinent points. > >I know it is a big "risky" to single you out, since many others bring a >tremendous amount of very useful information to this forum as well, but >your replies have a special clarity, completeness, and depth of >knowledge which is really extraordinary. I do indeed thank all of the >other contributors as well. > >I so very much wish that other forums and other contributors could use >this forum and your replies as a "model". It is rare to see such >courtesy, pertinent / relevant answers, and a lack of egocentricity, >and I, for one, applaud these efforts with great respect. > >Thanks again for all of us. > Hear hear. -- Robin (BrE) Herts, England
First
|
Prev
|
Pages: 1 2 3 Prev: Any recommendations on a AM2+ motherboard Next: high quality Adidas shoes wholesale |