If you disable Intel TM, will your CPU melt down if cooling fails?

[Seero]

In any case, I have not been able to get within 5 degrees of TJMax. The hottest I have ever managed to get it was 93 degrees C on core 1, which would correlate to 7 degrees from TJMax.

And actually, these numbers are probably wrong. I've read one theory that is most likely correct : different cores in an i7 CPU have different setpoints for TJMax. This prevents them from all throttling at the same time. That would explain why my "hottest" core is always, no matter what, almost exactly 8 degrees "hotter" than my "coldest" core.

Since the temps are really distances from TJ Max, and Occam's razor suggests that the temps should be the same in each core given equal workload, this means that each core's TJMax is different.

Now that I know how Intel TM actually works, I'm going to leave it on with all future overclock builds. 95 C (or 5 from TJMax) is HOOOT. There is nothing gained by letting the CPU get hotter than that. Even the most extreme overclocker will have cooling better than that. (which means TJMax is 85 C for a C2D)

One way or the other, if it is over 80 degree, then it is too hot. I can run torture tests for 10s of hours and never exceed 75 degree. Keep in mind that dust gather under the fins of the heatsink, which eventually blocks air flow, which leads to higher temp, which will then cause the CPU to throttle IF YOU ARE LUCKY. Temperature of the die can shoot up 10 degree, or even 100 degree under 1 ms, killing the chip instantly.

You can put the CPU is an oven and cook it at 200 degree and it will still work. TjMax really isn't mean the chip will melt or malfunction at that temp. It is a safe mechanism that kicks off at that temp to avoid possible damage. Keeping the CPU working close to that temp will not have good ending for the CPU.

 

[Ben90]

One way or the other, if it is over 80 degree, then it is too hot.

Intel actually states that the optimal solution is keeping the processor at Tcase at all times. A 72C Tcase means your cores are near or above 80C. The GPU on my Lenovo T41 has been running at 100*C nearly 24/7 since 2003.

 

[Habeed]

Temperature of the die can shoot up 10 degree, or even 100 degree under 1 ms, killing the chip instantly.

No it can't. Thermodynamics won't allow it. Even a significantly overclocked chip cannot dissipate heat fast enough for that to happen right away, assuming that it is reasonably well coupled to the thermal reservoir of the heatsink.

I will concede that the higher temps my cpu reaches at peak are shortening the CPU's life. Cutting it in half if you believe Intel's estimates. That is, if my CPU could run 24/7 for 3 years at 85 C, it could run 24/7 at 75 C for 6 years.

Regardless, it will more than likely last for 5 years the way I am using it, at which point the computer will no longer be in service. (I build a new box every 2-3 years but keep the old hardware around until it's been approximately 5 years)

More than likely, by the time the CPU would have failed something else will have quit, such as the motherboard. That's what killed my old rig as I was converting it to a fileserver - I accidentally broke the motherboard while inserting RAM modules too firmly. CPU still worked but that become irrelevant since I was able to pick up an atom based system for $80 and use that instead.

 

[Seero]

No it can't. Thermodynamics won't allow it. Even a significantly overclocked chip cannot dissipate heat fast enough for that to happen right away, assuming that it is reasonably well coupled to the thermal reservoir of the heatsink.

I will concede that the higher temps my cpu reaches at peak are shortening the CPU's life. Cutting it in half if you believe Intel's estimates. That is, if my CPU could run 24/7 for 3 years at 85 C, it could run 24/7 at 75 C for 6 years.

Regardless, it will more than likely last for 5 years the way I am using it, at which point the computer will no longer be in service. (I build a new box every 2-3 years but keep the old hardware around until it's been approximately 5 years)

More than likely, by the time the CPU would have failed something else will have quit, such as the motherboard. That's what killed my old rig as I was converting it to a fileserver - I accidentally broke the motherboard while inserting RAM modules too firmly. CPU still worked but that become irrelevant since I was able to pick up an atom based system for $80 and use that instead.

The reason why chip will fry is rarely due to its on cooling, but the voltage it receives. When you set 1.4 voltage to CPU, assuming that it will always be that voltage and cooling is okay, meaning that heat generated by CPU got transfered to heatsink, the CPU will not fry for 100 years, long after the fan of the heatsink died. However, often enough the PSU and mobo degrades, and therefore sending unstable electricity towards CPU. That means, even though bios stated that it is 1.4v, it can shoot up under different scenarios, which is what droop is for. While the temp of the surface of CPU is handled, spikes voltage will generate unexpected heat on wires which may not be radiated fast enough before it hits melting temp.

Think of a 750watt PSU where 90watt goes to the CPU while 100watt goes to other parts of the system, that is 190watt in total. GPU suddenly needs 400 more watt, making the total power demand from 190 to 590 watt. The PSU can supply this, but there will be a fraction of a ms where the power towards other parts take a small dip, like 88 watt to cpu and 97 watt to other parts of the system before everything stabilizes. Now GPU finishes its job and total power draw needed becomes 190watt, down from 590 watt. Again, there is a fraction of a second where the CPU is getting 92 watt and 103 watt the other parts. As PSU degrades, it takes longer meet demand, and there for the min and max voltage that the CPU will receive will be further apart and at longer duration. Good motherboard has multiple layers of power protection to stabilizes the power to CPU, however, unstable voltage from PSU can weaken the mobo as well as any parts connected to it. In reality, each clock cycle impacts the voltage, capacitors weakens, power surge from the wall outlet, and all other fun jazz.

For professional explaination, read this

Again, when disaster comes, those thermal throttle tricks aren't going to save you as the unexpected heat generation is to much radiate off the surface of the CPU, not to mention to the heatsink. If the surface of the CPU is at 90ish, there are no more headroom.

 

[Ben90]

...which leads to higher temp, which will then cause the CPU to throttle IF YOU ARE LUCKY. Temperature of the die can shoot up 10 degree, or even 100 degree under 1 ms, killing the chip instantly.

No, and the CPU will always throttle itself.

Again, when disaster comes, those thermal throttle tricks aren't going to save you as the unexpected heat generation is to much radiate off the surface of the CPU, not to mention to the heatsink. If the surface of the CPU is at 90ish, there are no more headroom.

Just wrong. Perhaps a friendly little bet in any reasonable denomination you choose could help make this forum a more informatively-correct place:

I load up my i7-920 using Intel Burn Test on maximum. After the first pass, I remove my heatsink and let it do its thing. Rinse and repeat 2 more times for redundancy.

If the chip completes 5 error free passes of IBT on maximum with the heatsink on I win, otherwise you win. Of course this isn't a closed thing, anyone else can join as I am always up for some free money.

 

[Habeed]

Seero : Ben's right. A spike in voltage is not going to melt the chip. The reason is thermodynamics - there is insufficient energy delivered with a small voltage spike (large voltage spikes can't happen) to actually damage a physical object like a microchip linked to a heatsink.

What DOES happen is there can be various effects that wear out some of the insulating junctions with charge migration - and at higher temperatures this is much more likely to happen. A totally different effect.

 

[Seero]

No, and the CPU will always throttle itself.


Just wrong. Perhaps a friendly little bet in any reasonable denomination you choose could help make this forum a more informatively-correct place:

I load up my i7-920 using Intel Burn Test on maximum. After the first pass, I remove my heatsink and let it do its thing. Rinse and repeat 2 more times for redundancy.

If the chip completes 5 error free passes of IBT on maximum with the heatsink on I win, otherwise you win. Of course this isn't a closed thing, anyone else can join as I am always up for some free money.

May be I have not been clear. I said, inefficient cooling is not the number 1 reason shall a CPU fries. If everything else is okay, a broken heatsink won't kill the CPU. In real life, anything can go wrong, will go wrong. The reminding question is when?

Crank your vcore to max and see. Based on what you said, it is safe because throttling will always saves the day right? According to my knowledge, over-voltaging the CPU can kill it regardless of throttling or any other fail safe mechanism.

BTW, in the case of heatsink removal, it isn't throttling, which is a not cause by TCC (Thermal Control Circuit) that may save your CPU, but another mechanism called thermtrip.

www.overclockers.com
THERMTRIP

In the event of a catastrophic cooling failure, the processor will automatically shut down when the silicon temperature has reached approximately ~135 °C. At this point, the system bus signal THERMTRIP# goes active and power needs to be removed from the processor. THERMTRIP# stays active until RESET# has been initiated. THERMTRIP# activation is independent of processor activity and does not generate any bus cycles.

Now you look at all this stuff and wonder why Intel couldn’t do something simpler, like the good old in-socket thermistor. There’s a good reason why only on-die diodes provide the best safety valve:

Unfortunately, measuring temperature with a thermocouple on the processor package has some inherent disadvantages when using the resulting data to control a thermal management mechanism.

Thermal conductivity through the processor package creates a temperature gradient between the processor case and silicon. This temperature difference may be large with the silicon temperature always being higher than the case temperature. Since thermocouples measure case temperature, not silicon temperature, significant added margin may be necessary to ensure the processor silicon does not exceed its maximum specification [i.e., fry].

As you can see, the key to prevent frying is having extra "margin" on tolerance, but that is still not 100% foolproof. Operating at temp close to TjMax means higher chance of frying your CPU. That means, working at high temperature will not shorten the CPU's life, but increase the chance of frying the CPU.

 

[Habeed]

Seero : working at high temps will shorten the CPU's life. Silicon does age and wear out, eventually. This happens a lot faster at higher temps - Intel estimates every 10 degree C doubles the rate.

The reason, however, is not for the reason you state. The chip doesn't actually "fry" - a few of the hundreds of millions of gates in the chip wear out. Just one error in memory address calculations and it's BSOD time. At the nanoscale, even "solid state" electronics use moving electrons that cause a form of wear on the components.

Anyways, long story short is that running a chip at lower temps and lower voltage will increase it's life. However, for the timespans that people actually use a computer, this doesn't matter. Every processor I have ever owned still works.

 

[Ben90]

The reason why chip will fry is rarely due to its on cooling

May be I have not been clear. I said, inefficient cooling is the the number 1 reason shall a CPU fries.

It purpose isn't to prevent melt down, but to prevent BSOD.

TjMax really isn't mean the chip will melt or malfunction at that temp. It is a safe mechanism that kicks off at that temp to avoid possible damage.

Your CPU will hang as soon as you hit TjMax, throttling occurs at 5 degree below TjMax.

According to my math, your next post will have at least one wrong fact.

 

[Emulex]

I've got an old boxen it will do a thermal shutdown at 40C ambient - well that is not smart. It doesn't go try to slow itself down or anything just thermal shutdown. so i disabled it. i mean what good is an unreliable box. If ambient does get to 40C i'd rather toast the machine than have it go down

 

[Seero]

According to my math, your next post will have at least one wrong fact.

Let me do you a favour. Maybe, just maybe, you will have better understanding on electronics after watching these:
This is how experts clean their motherboards

Less fancy way to clean it.

Difference in CPU heat generate on a clean mobo vs a clean mobo with good capacitors

Just in case you don't know, there are lots of chipsets on a motherboard. 100 degree is nothing to a CPU, but when the surface temperature of the IHS is 100 degree, then the temperature on the silicon inside the CPU must higher. The temperature is not important in this case, but the change in temperature is. Material expand at different rate. That means, although that are no parts that are made to move, there are parts moving. This alone will decrease the lifetime of a CPU, bigger gradient means bigger movement. Bigger movement means bigger the chance for cracks to occur within the CPU. When cracks occurs, the CPU may or may not work.

But that you already knew and it isn't the question. The real question are, when will it dies and what condition? and the answer is "it is random."

Aside from heat, frequency and voltage also ware down the CPU due to electromigration, which will eventually put the CPU to bed for good. None of these can ever be prevented by TCC as it is really a wear and tear effects. TCC however can keep the rate of electromigration under control by not allowing temp to exceed operating temp. Again, electromigration does occur and will eventually break the CPU, but it is totally random. That is why people don't say the "lifetime" of the CPU but the "expected lifetime" of the CPU.

The above is not the number 1 reason why CPU dies because all CPUs are pretty good in qualities and should last long enough before you decided to ditch it. The number one reason is unstable voltage, which can be due to motherboard, power supply unit, or the power supplied to the PSU.

Expensive mobo have multi layers of power protection to ensure voltage that CPU gets is as stable as it can get. However, they act like bumpers while electrons really don't know how to drive. Now here is the killer:
bad capacitors

When this happens, it will discharge higher voltage than it is specified. So let say you have set 1.375v to CPU and it had been 1.375v for some time and you leave and forget about it, which is fine. You keep temperature within the operating temp, which is good.

So now your CPU uses takes maximum operate voltage without throttling, that is good, and the CPU will still last more than 150,000 hours. What can go wrong?

One bad capacitor and discharge, hmm say 1 volt in 1 us. Unluckly, it is one of those bumper who is now fed up with all those electrons. For 1us, your CPU got 2.375v going into the silicon, and wherever this bulk of electron hits, it generates heat that exceeded the number the IHS can handle(radiate in time), and therefore melting the little, thin silicon as it passes by. When DTS detects this heat, it takes other 20us to stop more electrons coming, but it is a little too late.

Well 1 volt is impossible as the capacitor can't possibly store so much charge, but you get the idea. Bad capacitor can discharge and a different rate with different voltage. While one capacitor can't do alot, a series of them can do it. As time goes capacitors got weakened, and some will go rogue, breaking capacitors that are connected to it. Although we don't know when it will occur, it will eventually occur.

So, has this go off chart yet? I'm so got trolled.

 

[Ben90]

That first cleaning video made me cringe. Do not expect long term durability by smacking the shit out of it as he did in 3:50-4:10. Nothing might get damaged for reletively new boards, but I've had capacitors pop off from much less stress than that. Also the shaking at 1:36 is putting a decent amount of stress on the traces.

Don't post that as a professional way to clean a board. A proffessional would have used an ultrasonic cleaner with distilled water for at least the rinse stage.

 

[Habeed]

Condensing your wall of text :
So, you're saying that the most common way for a CPU to die is for the motherboard to fail and take the CPU down with it.

This mechanism of failure is the same regardless of whether you overclock or not.

Fair enough, I've heard of this happening, and once the motherboard dies the computer is generally too old for the mobo to be worth replacing. (so it really doesn't matter if the CPU still works at that point)

 

[taltamir]

Seriously scary movies... do not clean your mobo and HDD in such a way!

 

[Rifter]

Seriously scary movies... do not clean your mobo and HDD in such a way!

this x2, that guy has no idea what he is doing.

 

[Seero]

That first cleaning video made me cringe. Do not expect long term durability by smacking the shit out of it as he did in 3:50-4:10. Nothing might get damaged for reletively new boards, but I've had capacitors pop off from much less stress than that. Also the shaking at 1:36 is putting a decent amount of stress on the traces.

Don't post that as a professional way to clean a board. A proffessional would have used an ultrasonic cleaner with distilled water for at least the rinse stage.

Physical test is necessary as people do kick/knock the case from time to time. Shall things falls out, then they are either not soldered properly or the joins have oxidized. It will be dump not to check for bad joins before washing it. When they find bad joins, they simply resolder them back. They would have fixed problematic joins too. If he doesn't smack it, then how does he know parts are firmly in place?

Condensing your wall of text :
So, you're saying that the most common way for a CPU to die is for the motherboard to fail and take the CPU down with it.

This mechanism of failure is the same regardless of whether you overclock or not.

Fair enough, I've heard of this happening, and once the motherboard dies the computer is generally too old for the mobo to be worth replacing. (so it really doesn't matter if the CPU still works at that point)

What I am trying to tell you is TCC (throttle) serves only as a warning telling you that you should fix the problem which causes the high temp. It will not prevent CPU from frying. Usually cleaning, and replacing TIM solves the problem. Sometimes you have to fix the airflow within the case, replaces caps(or simply mobo) or PSU. Ignoring it will eventually lead to a dead PC. It is best to keep CPU idle temp under ambient temp + 10(around 30-40 celsius), load temp no higher than 60% of TjMax(around 50-60 celsius), and torture temp no higher than 80% of TjMax(that is around 70-80 celsius). Temperature Gradient should ever exceed 30c (The temperature difference from idle to torture.) During torture, temp will slowly increase, but that is okay as it means the heatsink is doing its job. Usually, CPU idle should be around ambient + 5, or 30 degree in a room with room temp 25c. Starting prime95 with small FFT will bump temp almost instantly to 50ish and then slowly increase. This increase slows down on every additional degree until it takes 30 minutes to get from say 72 to 73. Good contact between CPU and heatsink + good air flow will a)decrease the gradient, and b) decrease max temp. Since capacitor last longer under cooler temp, having good air flow can prevent high temp within the case, and therefore prolonging the life of each caps. This are things an experienced user will do and keep their eyes on.

Experts will test their PSU from time to time to ensure that it still provide stable voltage before it decides to die while things are connected to it. They will also visually inspect each caps and joins to ensure that everything is in place and functioning. Washing it like the video isn't part of the routine, but there are times when that is necessary. Oil submerge cooling is one of the example.

Seriously scary movies... do not clean your mobo and HDD in such a way!

That is an interesting question where I always wanted to know. I know a single dust inside the HDD will kill it, but if that is the case, then the case must be air-proof. If it is air proof, it should also be waterproof too.

It isn't the wash that will kill HDD, but trying to dry it. I believe the platter inside will bend, just like CDs after leaving it in the car.

As to mobo, don't do it yourself. Those ain't air-dryer, they are heat-gun. They know how to heat the board up evenly without melting anything. It was just to demonstrate the difference between Facts and myths. Water don't kill electronics, it is the conductive metal within the water that causes shorts while the mobo is running. Dust contain metals too, but they don't flow around like when they are in water, and therefor less likely to cause shorts. Dirty boards are more likely to short compare to a clean one.

Compress air is the best way to clean it. Again, washing it isn't a preferred method, even though it can be done.

 

[taltamir]

That is an interesting question where I always wanted to know. I know a single dust inside the HDD will kill it, but if that is the case, then the case must be air-proof. If it is air proof, it should also be waterproof too.

One flaw in that assumption... air is not dust. It needs to be dust proof, not air proof nor waterproof.
And something can most certainly be airproof and not waterproof.

It isn't the wash that will kill HDD, but trying to dry it. I believe the platter inside will bend, just like CDs after leaving it in the car.

CDs are cheap plastic (there are some very heat durable plastics out there)... the HDD platters are metal.

As to mobo, don't do it yourself. Those ain't air-dryer, they are heat-gun.

Irrelevant.

They know how to heat the board up evenly without melting anything.

Do you mean that heatguns "know" not to melt stuff by being weaker then hair dryers? or do you mean that the person doing the video demo knows how to move it quickly enough to not melt it?
Heatguns produce more heat, its the technique that prevented melting, and frankly you can just bake it in the oven at 100 degrees for a safer experience.
Anyways, the heatguns aren't what bothers me, the soapwater bothers me. Oh, and also the hitting it... especially hitting it when it has just been heated with a heatgun (and thus solder might be loose enough for stuff to detach)

It was just to demonstrate the difference between Facts and myths.

No, it demonstrates horribly bad handling of electronics.

Water don't kill electronics

Water, even pure water, causes corrosion if not dried ASAP...

it is the conductive metal within the water that causes shorts while the mobo is running.

Finally something true... well sorta, it might be true but in context it insinuates some falsehood.

Dust contain metals too

Generally not.

Dirty boards are more likely to short compare to a clean one

Dust is completely and utterly harmless to a computer.

The PROPER way to clean dust is with a vacuum followed by compressed air.

The only reason to EVER wash an entire motherboard is if you spilled a liquid on it (say, coke, or cofee)... in which case you should wash it with DDI water ASAP and then dry it ASAP. Even DDI water will cause corrosion and will leach minerals from the board itself to become conductive. And when using regular water (or, shudder, soap water) they leave mineral deposits after drying which can cause shorts.

Also, why in the world is that moron in the videos throwing all the cables to soak in soapwater? ugh!

 

[Ben90]

Seero, I take back all the mean and hurtful things I said to you. Its obvious my technical knowledge is not up to par with yours.

That being said, can you please go into more detail regarding the relationship between capacitors and temperature?

 

[Seero]

One flaw in that assumption... air is not dust. It needs to be dust proof, not air proof nor waterproof.
And something can most certainly be airproof and not waterproof.

I didn't say for sure, that is my speculation. I don't know much about hard drive assembly so I can't say for sure.

CDs are cheap plastic (there are some very heat durable plastics out there)... the HDD platters are metal.

Again, I never tried that on my HDDs, so I don't know.

Irrelevant.

Do you mean that heatguns "know" not to melt stuff by being weaker then hair dryers? or do you mean that the person doing the video demo knows how to move it quickly enough to not melt it?
Heatguns produce more heat, its the technique that prevented melting, and frankly you can just bake it in the oven at 100 degrees for a safer experience.
Anyways, the heatguns aren't what bothers me, the soapwater bothers me. Oh, and also the hitting it... especially hitting it when it has just been heated with a heatgun (and thus solder might be loose enough for stuff to detach)

I am referring to the person who uses the heat gun and I will describe the difference between heatgun and hairdryer later in this post.

No, it demonstrates horribly bad handling of electronics.

Water, even pure water, causes corrosion if not dried ASAP...

To me, you are simply arguing for the sake of arguing. I never said people should soak the mobo and air dry it. I said do not do it yourself as those are experts. Please note that water IS NOT CONDUCTIVE, it is the ions within that makes it appears to be conductive. Distill water is a bit different than Deionized water. Deionized water does not contain ions and there for will on be conductive and won't cause corrosion/oxidation. However, if it made contact to air, it will draw ions from air and become ionzed, and therefore corrosive and conductive. Since air also contain water molecules, meaning air itself is conductive and corrosive, electronics will rust even if you have never ever spill water on it as long as it makes contact with air. This is the mean reason why parts on old boards tend to fall off because the pin has oxidized with ions within air.

Finally something true... well sorta, it might be true but in context it insinuates some falsehood.

Quote the part where it is false.

Generally not.

Dust is completely and utterly harmless to a computer.

The PROPER way to clean dust is with a vacuum followed by compressed air.

Whether dust will cause short depend on how dirty it is. Dust(or dirt in the scope of what we are talking about) itself, however, is conductive as it is a formation of died skin, hair, and some random particles. They do short electronics from time to time, causing itself to burn, producing smokes that may or may not be visible but sometimes it is strong enough for your nose to smell it. This type of short may or may not damage hardwares. Other than shorts, it can become an insulation to prevent heat from radiate away, which will cause burn out. Dust can also form a barrier on top of heatsinks which blocks airflow.

It is interesting that you never quoted the part where I say compress air is the way to go.

The only reason to EVER wash an entire motherboard is if you spilled a liquid on it (say, coke, or cofee)... in which case you should wash it with DDI water ASAP and then dry it ASAP. Even DDI water will cause corrosion and will leach minerals from the board itself to become conductive. And when using regular water (or, shudder, soap water) they leave mineral deposits after drying which can cause shorts.

See above. Note that mineral residues after drying is no worst than dust itself.

Also, why in the world is that moron in the videos throwing all the cables to soak in soapwater? ugh!

That is why the use of heatgun is relevant. hairdryer pushes hot air out. It heats up areas where it is blowing at, but cooling adjacent areas . Heatgun doesn't do that. If water can find a way into a cable, then air can find a way out. You can't heat up the board evenly with a hairdryer.

As to the oven tricks, it works if it is only used to bake hardwares. Use the one that you (your family) cook with is like putting poison in your own food.

 

[Ben90]

I didn't say for sure, that is my speculation. I don't know much about hard drive assembly so I can't say for sure.


Again, I never tried that on my HDDs, so I don't know.


I am referring to the person who uses the heat gun and I will describe the difference between heatgun and hairdryer later in this post.


To me, you are simply arguing for the sake of arguing. I never said people should soak the mobo and air dry it. I said do not do it yourself as those are experts. Please note that water IS NOT CONDUCTIVE, it is the ions within that makes it appears to be conductive. Distill water is a bit different than Deionized water. Deionized water does not contain ions and there for will on be conductive and won't cause corrosion/oxidation. However, if it made contact to air, it will draw ions from air and become ionzed, and therefore corrosive and conductive. Since air also contain water molecules, meaning air itself is conductive and corrosive, electronics will rust even if you have never ever spill water on it as long as it makes contact with air. This is the mean reason why parts on old boards tend to fall off because the pin has oxidized with ions within air.


Quote the part where it is false.


Whether dust will cause short depend on how dirty it is. Dust(or dirt in the scope of what we are talking about) itself, however, is conductive as it is a formation of died skin, hair, and some random particles. They do short electronics from time to time, causing itself to burn, producing smokes that may or may not be visible but sometimes it is strong enough for your nose to smell it. This type of short may or may not damage hardwares. Other than shorts, it can become an insulation to prevent heat from radiate away, which will cause burn out. Dust can also form a barrier on top of heatsinks which blocks airflow.

It is interesting that you never quoted the part where I say compress air is the way to go.


See above. Note that mineral residues after drying is no worst than dust itself.


That is why the use of heatgun is relevant. hairdryer pushes hot air out. It heats up areas where it is blowing at, but cooling adjacent areas . Heatgun doesn't do that. If water can find a way into a cable, then air can find a way out. You can't heat up the board evenly with a hairdryer.

As to the oven tricks, it works if it is only used to bake hardwares. Use the one that you (your family) cook with is like putting poison in your own food.

Compelling proposition... I'm curious to the relation between mineral residues and heatguns.

 

[taltamir]

I said do not do it yourself as those are experts

No! I am an expert, they are amateurs who don't know what they are doing. Soapwater bath for cables, mobo, and HDD is the most retarded way to "refurbish" an old computer I have ever seen.

Other than shorts, it can become an insulation to prevent heat from radiate away, which will cause burn out. Dust can also form a barrier on top of heatsinks which blocks airflow.

The only time dust is a problem is if it impedes airflow or contact (aka, if it got inside a slot or socket where pins should go)... that is when you use compressed air as I said. Water is a bad solution as it can just serve to spackle it in there.

It is interesting that you never quoted the part where I say compress air is the way to go.

1. it is true, compressed air is the best way to clean hardware.
2. yet you still refer to those horrible mishandling as "Expert way of cleaning". This is wrong. Real experts use compressed air.

To me, you are simply arguing for the sake of arguing

No, I am just concerned someone will ruin his hardware by following your "expert advice"

 

[Seero]

Seero, I take back all the mean and hurtful things I said to you. Its obvious my technical knowledge is not up to par with yours.

That being said, can you please go into more detail regarding the relationship between capacitors and temperature?

May be Taltamir can give you a better answer, but here is mine:

Since there are different parts of the system that requires electricity, the wattage provided by PSU varies. Fire up a game and PSU draws more watt. Shutting id down and PSU draws less watt. Note that there is a time required to go from say 50watt to 550watt, and from 550watt back to 50watt. Let say it takes too long for it to go up, you will probably get BSoD or the system simply shuts down and you will probably be like "WTF?!" That is the good side of failure.

The bad side of failure is it doesn't come back down it time. Since video card no longer needs extra electricity, but it is still coming from the PSU, where does those extra electricity go? In abstract, it turns to extra heat.

Since the time it took is never 0, and therefore the system will have a time tolerance to prevent BSoD and shutdown, and heatsink/TCC is designed to handle those extra heat.

Usually people don't max out there PC. PC usually dies when owner decided to try out some new games, PC dies with smokes. The cause of death can be a combination of cooling and unstable power. Make sure the PC is cooled effectively may prevent this from happening. However, if it is cooling that causes the problem, the PC should throttle. You can see this happen all the time simply by visiting forums of new games. People strongly believe that their PC is perfectly fine and how it handles torture without sweat and therefore it must be the game while others keep saying it is due to poor cooling.

No! I am an expert, they are amateurs who don't know what they are doing. Soapwater bath for cables, mobo, and HDD is the most retarded way to "refurbish" an old computer I have ever seen.

Well, soap is the best way to remove oily and dirty stuffs. It is unnecessary, but can be done. Hey, what is the point of having a clean board with dirty cables?

2. yet you still refer to those horrible mishandling as "Expert way of cleaning". This is wrong. Real experts use compressed air.


No, I am just concerned someone will ruin his hardware by following your "expert advice"

Compress air is the best way to clean it. Again, washing it isn't a preferred method, even though it can be done.

I don't recall telling people to do what they see. I remember I said not the do this over and over. Now if your intention is to warn others, then I am with you.

 

[Ben90]

May be Taltamir can give you a better answer, but here is mine:

Since there are different parts of the system that requires electricity, the wattage provided by PSU varies. Fire up a game and PSU draws more watt. Shutting id down and PSU draws less watt. Note that there is a time required to go from say 50watt to 550watt, and from 550watt back to 50watt. Let say it takes too long for it to go up, you will probably get BSoD or the system simply shuts down and you will probably be like "WTF?!" That is the good side of failure.

The bad side of failure is it doesn't come back down it time. Since video card no longer needs extra electricity, but it is still coming from the PSU, where does those extra electricity go? In abstract, it turns to extra heat.

Since the time it took is never 0, and therefore the system will have a time tolerance to prevent BSoD and shutdown, and heatsink/TCC is designed to handle those extra heat.

Usually people don't max out there PC. PC usually dies when owner decided to try out some new games, PC dies with smokes. The cause of death can be a combination of cooling and unstable power. Make sure the PC is cooled effectively may prevent this from happening. However, if it is cooling that causes the problem, the PC should throttle. You can see this happen all the time simply by visiting forums of new games. People strongly believe that their PC is perfectly fine and how it handles torture without sweat and therefore it must be the game while others keep saying it is due to poor cooling.


Well, soap is the best way to remove oily and dirty stuffs. It is unnecessary, but can be done. Hey, what is the point of having a clean board with dirty cables?




I don't recall telling people to do what they see. I remember I said not the do this over and over. Now if your intention is to warn others, then I am with you.

Compelling proposition... I'm curious to the relation between gaming and TCC.

 

[aigomorla]

One flaw in that assumption... air is not dust. It needs to be dust proof, not air proof nor waterproof.
And something can most certainly be airproof and not waterproof.

HOW? when proofing something u are limited the size of molecules from entering.

How is air mol. larger then a water mol?
If you have a fan in your case, its by far not airproof.
Even the tubing cant hold water perfectly from osmosis because its pourus.
The only reason why u dont have air going out is because in a closed LC system it works under the pretense of vacuum when the pump goes on.
Which is why over time your tubing collapses unless u tap air.

Do you mean that heatguns "know" not to melt stuff by being weaker then hair dryers? or do you mean that the person doing the video demo knows how to move it quickly enough to not melt it?

Actually ive seen heat guns with built in timers, and also temps which they spit out.
Heat guns are just high hair dryers and yes your right, they are all far more capabile of dishing out more heat then any hair dryer.

the soapwater bothers me. Oh, and also the hitting it... especially hitting it when it has just been heated with a heatgun (and thus solder might be loose enough for stuff to detach)

+1 who the hell is telling people to use soap water on hardware?
Your best off giving it an alcohol dip.

Water, even pure water, causes corrosion if not dried ASAP...

The metal would need to oxidize. Usually pure water drys clean and evaps. Very little oxidation will occure.

This is why tap water usually leaves hard water stains, while distilled water will evap clear without water stains. .

Dust is completely and utterly harmless to a computer.

Whatcha talking about Willus? The amount of dust can impede air flow, and clog sinks.
This will reduce efficiency of the heat sink and disapation of heat from your other components.

Dust kills, especially if its charged, and you happen to static it.

The PROPER way to clean dust is with a vacuum followed by compressed air.

The only reason to EVER wash an entire motherboard is if you spilled a liquid on it (say, coke, or cofee)... in which case you should wash it with DDI water ASAP and then dry it ASAP. Even DDI water will cause corrosion and will leach minerals from the board itself to become conductive. And when using regular water (or, shudder, soap water) they leave mineral deposits after drying which can cause shorts.

Totally depends... What i do is this:
I will compress blow dry, and then take a fine makeup brush... yes the kind your wife/GF uses to put powder on her cheek, and brush it off.

Dipping boards, or pouring liquid over the board is what i call a Hail Mary.
Hail marys involve Alcohol Dip.
Yes you heard me correct, you pour 99% pure rubbing alcohol all over your board in hopes that the alcohol replaces the water.
Then allow the alcohol to EVAP so it leaves nothing.

And once again, you need oxidation for rust to occure.
Without a metal which can oxidize, you will not rust.

Copper does not rust for example, it tarnishes.

Also, why in the world is that moron in the videos throwing all the cables to soak in soapwater? ugh!

Because he has not been taught by aigo nor has he read any of my guides

Big sarcasm...

Let me do you a favour. Maybe, just maybe, you will have better understanding on electronics after watching these:
This is how experts clean their motherboards


So, has this go off chart yet? I'm so got trolled.

Yes you are, because even i have to agree he's a total idiot.

But what would i know eh? ive only watercooled for more then 10yrs, and have done just about everything in salvaging leaked systems to complete nightmare wrecks.

But hey, if u say he's an expert for using soap water... then maybe we should educate you the proper real ways.

Only things u can dip in soap water:

1. Your heat sink
2. Your fans.. yes fans are OK...
3. Your fingers
4. IC Diamond in hot water to loosen it a bit.
5. Your case without any hardware.
6. Your mascott after he has made a mess of your TIM.. er i mean your pet dog..

Everything else ur asking for it.

 

[taltamir]

HOW? when proofing something u are limited the size of molecules from entering.

How is air mol. smaller then water mol?
If you have a fan in your case, its by far not airproof.
Even the tubing cant hold water perfectly from osmosis because its pourus.
The only reason why u dont have air going out is because in a closed LC system it works under the pretense of vacuum when the pump goes on.

1. We are discussing HDDs.
2. You are incorrectly assuming that air proofing and water proofing works via filtration... preventing dust (which is hundreds, thousands, or more atoms big) is done via filtration. Water and air proofing requires a seal (say, a rubber one) which is limited on the pressures it handles, a seal can be damaged by water, or water could seep in via wicking actions, etc etc...
Theoretically, you could make a watertight seal that reacts with atmospheric gases to decompose, but I don't know of any... it is fairly simple to do the reverse and make something airproof but incapable of surviving water.

regardless, we are talking HDDs which are dust proof but let water and air pass right through.

+1 who the hell is telling people to use soap water on hardware?

take a look at that:

Let me do you a favour. Maybe, just maybe, you will have better understanding on electronics after watching these:
This is how experts clean their motherboards

Less fancy way to clean it.

Difference in CPU heat generate on a clean mobo vs a clean mobo with good capacitors

To his credit, Seero said that while those "professionals" are cleaning with soapwater, average joe shouldn't, and should juse use compressed air... I don't think those youtube "experts" deserve the name experts... PS, that video? where they throw a mobo, cables, and a HDD into soap water? "that is how I refurbish old computers"... yap, they don't even have a specific reason, its an old computer so it must be "refurbished" in the soap water.

Your best off giving it an alcohol dip.

ooh, I haven't considered that... plus rubbing alcohol is expensive...
but I do think you have a point, pure alcohol will work better than DI water

Whatcha talking about Willus? The amount of dust can impede air flow, and clog sinks.
This will reduce efficiency of the heat sink and disapation of heat from your other components.

Dust kills, especially if its charged, and you happen to static it.

In the context of which I said it, I was referring to dust that is merely TOUCHING the hardware... Seero and said experts are saying that dust causes shorts by merely touching the mobo, I am saying dust in itself is harmless unless it clogs a fan, vent, or gets inside a socket/slot, but dust doesn't magically damage hardware just by being on it.

The charged dust causing static shocks is something I haven't heard of or seen before and I am taking with a grain of salt unless shown proof.

And once again, you need oxidation for rust to occure.
Without a metal which can oxidize, you will not rust.

Copper does not rust for example, it tarnishes.

Same difference, yes it is tarnished not rust... oxidized iron is rust, oxidized silver, copper, and any other metal that is capable of oxidation is called tarnish. I know, I misused the word and should have called it tarnish, this is just splitting hair.
What I was referring to specifically is that water, if not dried completely right away, will, over the time it takes it to dry naturally in air (potentially a few days), cause accelerated tarnishing... Tarnishing can cause it to fail. I have actually repaired electronics by merely physically scraping off tarnish off of the etched circuits of PCB using razor.
Electrolyte leeching from the PCB itself will take quite a long time though, and I don't THINK it will happen fast enough to matter (aka, before the water naturally dries out), when I mentioned it was talking about the general properties of water... but whether it is part of the cause or not, soaking your hardware in water and letting it dry off naturally is not the best of ideas.

If you spill a soft drink, the carbonation causes it to be acidic, and it is full of conductive minerals to deposit... it will ruin the board via both tarnishing and short circuiting... this is when you wash with DI water or ideally DDI water. You want to dry it off so as to prevent water damage. If you dry it off quickly you should have no damage.
Using alcohol to remove the DI water is a good idea.