1 of 1 people found this helpful
If the motherboard is still working well in other regards, you could just disable the onboard LAN in the motherboards BIOS setup and install a compatible Network Interface Card (NIC) and its relevant drivers to give you back your network connectivity. That's what I did with a work colleagues computer some time ago and it worked a treat.
It'd definitely be worth buying a surge suppressor to power your computer system from now on. It's possible to buy them with ports for your phone line as well so if the mains don't get struck by lightning but the phone lines (with your internet connection) do get a spike from lightning, you're still protected to a degree. Some of them even come with a promise to replace equipment connected through the device that fails due to lightning (up to a certain value). Any comprehensive somputer store should have info on suitable surge suppressors for home computer use.
Oh I see your problem now. What motherboard is it that you have in this PC? If it doesn't post or boot up anymore then maybe the BIOS has fried as well as the LAN circuitry (and possibly even more). When the LAN stopped working, I assume the rest was working OK - ie the screen hadn't frozen and you could still shut it down normally (rather than having to force the power off with the case button)?
If it's an Intel branded board you may be able to revive it with a recovery procedure BIOS flash but that'd be about all you could do short of learning motherboard design theory and obtaining some heavy duty testing and soldering equipment. I've changed a serial port on a motherboard before (that was physically damaged) but it was clear where the faulty part was in that case - with your board who knows and what looks damaged may only be the tip of the iceberg. I say this because many 'sensitve' components such as memory (including that used to store your BIOS) can fail due to ESD with no change in outward appearance.
Even if it's an old board and the above doesn't work, you may be able to find another the same secondhand (or at least another of a different make and model which will still take your memory and CPU etc - hoping they haven't been fried too). It could also be your PSU is fried as even though that green LAN light comes on, that does not signify the PSU is providing enough power on startup at the correct stable voltage for the whole PC to boot. When I worked as a computer technician, I once walked into a room just as a PSU failed on one of the computers nearby. There was a loud pop and a flash (and sparks) from the back of the PC - in that case the fan had seized and the PSU overheated. PSUs (being in the direct line of fire receiving mains voltage) are very suceptable to lightning related damage. Just because the PC remained working after the lightning strike does not mean the PSU can still initialise properly anymore.
Do you understand my suggestion about a surge suppressor for future use? I have both a surge suppressor and a UPS powering my PC and the phone line going into my ADSL (broadband) gateway is also protected using the same equipment. An ounce of prevention is worth a pound of cure and a little bit of money spent on equipment like this can save much more money having to be spent down the track in the situation you've described.
Message was edited by: Flying_Kiwi to include more about the PSU and its vulnerability
Your solution has three steps. First, what is damged (and the path of that damage) must be identified. That means trying to fix nothing. Only first define what is now defective. Second and only after step one, fix the defect. And third, eliminate reason for damage.
You said everything is good except a network interface. Did not say why that is known. Including why is always critically important. Apparently a lightning strike (maybe to AC wires way down the street) used your computer to connect to earth. That would explain the spark. A surge incoming on AC electric, alone, does no damage. To have damage, that transient must be both incoming on one wire (ie AC electric). While simultaneously outgoing via another wire (ie network cable). Typical is damage to the network interface when a surge is incoming on AC mains. Now, what else might be in that path?
Step two is fixing the defect. If the network interface (and not other components of that network) is definitely defective, then a NIC interface can be installed. Then a new network card provides a network connection while the existing and defective motherboard NIC is disabled in Device Manager.
Step three: that damage exists only when a homeowner all but invites lightning far down the street to find earth inside the building. Once inside, then nothing will stop a destructive hunt for earth. Nothing includes so many plug-in protectors that do not even claim to provide that protection. Protecting your computer and everything else (including the furnace, dishwasher, and every digital clock) is only about where energy dissipates. You had energy dissipating destructively inside the house because best protection already inside the computer was compromised. Earthing one 'whole house' protector at the service entrance (ie breaker box) means energy would not be inside a house. If no earthed protector and if 10 protectors were adjacent to that computer, then you still only have the best protection already inside a computer - nothing more.
Three tasks necessary to solve and then eliminate future failures.
Next post suggests other damage exists. But again, the reasons for each conclusion are not included. To know more means collecting facts with numbers. Then posting them hear so that others with better knowledge can make a more useful conclusion.
Your only other alternative is to keep replacing parts until something works. Very unlikely is a BIOS failure. Far more likely is damage in the path used to obtain earth ground via your hardware.
Westom does not believe plug-in protectors work. Much of "step 3" is nonsense.
The best information on surges and surge protection I have seen is from the IEEE:
And a more basic one from the US-NIST:
If using a plug-in protector all interconnected equipment needs to be connected to the same plug-in protector. External connections, like phone, also need to go through the protector. The IEEE surge guide explains plug-in protectors work primarily by limiting the voltage from all wires (power and signal) to the ground at the protector. The voltage between wires going to the protected equipment is safe for the protected equipment.
The NIST surge guide suggests most damage is from high voltage between power and phone (or other signal) wires. Sounds like that is what happened to the OP.
Service panel protectors are a real good idea.
But from the NIST guide:
"Q - Will a surge protector installed at the service entrance be sufficient for the whole house?
A - There are two answers to than question: Yes for one-link appliances [electronic equipment], No for two-link appliances [equipment connected to power AND phone or cable or....]. Since most homes today have some kind of two-link appliances, the prudent answer to the question would be NO - but that does not mean that a surge protector installed at the service entrance is useless."
Service panel protectors do not, by themselves, protect from high voltage between power and signal wires. They will very likely protect anything connected to only power wiring from a very near very strong lightning strike.
It's a Intel Motherboard DH55HC, I test the other components and with a new motherboard they works. In basic conditions (processor, ram memory), the motherboard try to turn but fails, threre is no video signal, green light on the mother board comes on, the fan spins for a split second then off, then try to do the same and so on. With a bypassing in the 20+4 pins connector (green and any black cable) turns on but with no signal video and no beeps.
The PC power connection used an voltage stabilizer, but it seems is not enough.
What the motherboard is trying to say about what is happening when I pressing the power button?. I try with jumper configurations of the BIOS, changing the cell battery, without battery and it is the same no change. It's clear the motherboard is ok, but is something that does not allow.
It's a Intel Motherboard DH55HC, I test the other components and with a new motherboard they works.
So you've swapped your lightning rod DH55HC for another DH55HC motherboard or was it a totally different motherboard that worked OK with all the other components in it? Irrespective of your answer to this question, I can't then see why you wrote "it's clear the motherboard is ok" as if swapping it with another get's things working fine, that would tend to indicate the opposite - perhaps this is a language issue and I've misunderstood you?
As for the above posts about US power surge protection, I'm only familar with ways it's commonly done here in the UK so can't comment with experience on those other systems. Similarly, I've not tried jumpering wires from ATX PSU connectors to start a PC so I don't know what colours are connected where. I'm sure there are reference out there and maybe the board manual might mention this be carefull to get it right if your trying such things though as that really can fry bits if you get it wrong.
What do you think about the possibility that your PSU may have been damaged to the degree that it no longer puts out a stable enough supply within this boards required voltage tolerances, yet is 'close enough' for a different board to allow booting? Some boards are more lenient about this sort of thing than others. If you have a spare PSU to test in the PC, it'd be worth trying that otherwise a bit of money spent at the local computer store (who should have the required components for these sorts of tests) may save you even more buying bits that aren't needed.
Normal is for a defective power supply to fail in one system and work just fine in another. Same applies to a motherboard. Critical to making any conclusions is how each works and the relevant numbers.
For example, a power controller decided where the power supply powers on. Then decides when the CPU can execute. As noted earlier, the problem must be seen before fixing anything. A multimeter (ie £7 in Maplin) and one minute of labor will provide those numbers. Only then can someone say, without any more doubt, what is and is not defective.
Normal is for a defective part to still boot and run a computer. Then it may cause problems months or even a year later. What is your power controller doing? Does it see a defect and power off the PSU? Is the defect marginal as to cause failures sometimes and not others? Numbers from six wires and one minute of labor will answer those questions accurately – without doubt. Then a problem is immediately identified. Or move on to other suspects; without even looking back.
Normal is for a defective power supply to fail in one system and work just fine in another. Same applies to a motherboard.
In my experience this does sometimes happen, although I haven't seen it often and I wouldn't call it normal. I also wouldn't say the damaged parts work 'just fine' (although they may still result in a booting system with issues).
We agree that the PSU is in need of checking out.
Not just the power supply.
Please don't get me wrong I'm not saying it's just the PSU that needs checking, as you mention, other parts of the system may have been damaged by being in the firing line but the PSU is a part that can (relatively easily) be swapped out - something thats not as easy to do with most components on the motherboard for example.
Although a mutimeter is a handy tool for measuring voltages etc, it doesn't give a good idea whats happening as far as stability of several bus voltages at the same time or quick voltage fluctuations etc. I'd try swapping with a suitable capacity known good supply - if the latter works then thats a pretty good sign it's (at least) the PSU. Many people won't have access to a spare PSU which is why I suggested the computer shop.
If designing a supply, then $thousands of test equipment is required. Once that design has been proven by engineering analysis, test equipment and other working products, then anything that might cause stability problems is identified even by a multimeter. If a PSU is not connected to the entire computer system , power controller, etc (if any wires are disconnected), then even the meter cannot accurately identify many problems (ie stability).
Any layman in one minute can collect those numbers from six key wires. But few can actually appreciate a massive amount of information embedded in those numbers. To explain all that would probably double the size of this thread. Any one who thinks numbers from the ATX standards are sufficient to understand what the meter is reporting did not yet learn some important facts. Meter numbers report numerous stability and other problems if those numbers are added to facts including how a PSU works and what a meter does to measure. The layman simply reports those numbers. Then learns from others what is wrong, if anything, in a next post. And without doubts.
Swapping supplies can identify some defects. Not say why. And even make a problem exponentially more complex. In diagnostic analysis, one either shotguns - tries to fix the problem by replacing good parts until something works. Or one first learns what is wrong before fixing it.
Even a car mechanic can find himself unemployed if shotgunning. Many never learn why shotgunning even makes impossible assistance from those who best know this stuff. Numbers from the six wires means a complete answer (even about stability) in a next reply.
Many never learn why shotgunning even makes impossible assistance from those who best know this stuff. Numbers from the six wires means a complete answer (even about stability) in a next reply.
I looked up that term as I'd only ever heard of using parts from one PC to test another as cannibalising the good machine. I guess there are lots of slang terms for it but all the shotgunning definitions I found related to ways of drinking beer faster.
If you mean that it doesn't get to the nitty gritty of what in the PSU failed and why - in the case we have here, I think it'd be safe to blame it on 'lightning/voltage spike damage' and move on, safe in the understanding that a replacement PSU got things working again. In this specific case though we don't know if that would make a difference until we try another PSU. How are things progressing dcarlosi?
Maybe damage is from lightning (a 'current' spike). But that does not say what is damaged. For example, a current spike can bypass superior protection inside a PSU to do damage elsewhere. To other parts of the power 'system'. Swapping a PSU would not necessary identify the problem (although sparks are a good indicator).
Of two diagnostics procedures, shotgunning has even exponentially complicated some solutions. Shotgunning assumes only two conditions - good or bad. The real world involves perspectives. Variations. Or to simplify reality, the world is ternary. Good, defective, or undefined (intermediate) state. Undefined is why a perfectly good supply can even fail in what is otherwise a good computer. And why a defective supply can even work in some computers.
At this point, the OP apparently is not asking for solutions since he has not replied. Has not asked for a solution in the very next post. So this is really about what is meant in an expression, "work smarter; not harder".
An unprofitable organization would even blame unfair competition because the basics (ie shotgunning verses smarter analysis) were never learned. Another example: an organization with quality control inspectors has inferior quality. Increasing quality begins with eliminating quality control inspectors. Another example of why reality does not make sense until concepts are learned. Another example of why shotgunning, that appears to be an effective solution, is often a least reliable and most expensive solution.
The OP would have a complete and unambiguous answer immediately by getting numbers from six wires. One full minute of labor means the next post says what is or is not defective - without doubts or speculation. Until one does this stuff, then the best solution makes no sense. Just like quality. How to increase quality? Starts with eliminating quality control inspectors. Often only makes sense when learned by doing it.