cancel
Showing results for 
Search instead for 
Did you mean: 

HELP High humidity causes failure to boot + quick fix.

Gone_Troppo
Level 7
Hello Everyone,
This is a bit of a weird one for you.

I'm currently trying to get to the bottom of an issue where high relative humidity causes my PC to not start up.
Hopefully someone here might have run into this before and may be able to shed some light on the situation.

Sorry this first post is going to be a long one.
I will try to keep it short yet detailed as possible there is also a potential quick fix at the end of this mega post.

Firstly here are the build specs, details to follow.

MOBO: ASUS Rampage IV Extreme
CPU: Intel i7 3970X @ stock 3.5 ghz
CPU Cooler: Corsair H100i
RAM: 32 GB Corsair Dominator Platinum 2133mhz c9 (4 x 8 Gb)
GFX: ASUS GTX Titan (Nvidia)
PSU: Corsair AX1200i (1200W)
HDD: 512 Gb Samsung 840 Pro SSD (OS)
HDD: 1 Tb Western Digital VelociRaptor (Storage)
DVD: Pioneer Blu-Ray burner
CASE: Corsair Obsidian 800D
OS: Win 7 Pro 64 bit
MONITOR: Samsung 50 inch plasma @ 1920 x 1080

OK my PC has been running just fine for the past few months, the first time this happened was a few weeks ago after it had rained overnight and I tried to switch the PC on in the morning, there was simply no response when I pressed the power button on the front of the case, yes I tried it several times.

At first I thought the worst, That it had suffered the same fate as my last PC ... Death by Lightning Strike.

However I noticed that the mouse light and motherboard lights were still on and working,
So I tried switching it off at the power supply, and unplugging it from the wall socket before plugging it back in and trying again, It still did not boot.

As the PC is still under warranty from the store that it was purchased from I packed it up and took it in to them so they could have a look and see if they could find the fault. I thought that the front power button had probably failed or developed a fault.

The guys in the PC store had their tech's look at it and they re-seated the ram to fix the problem.
Apparently they had 5 PC's come into the store on the same day with the same issue and re-seating the ram fixed them all.
They claimed that it was due to the rapid increase in humidity caused by the first decent rains this wet season and if it should happen again I should re-seat the ram myself rather than bringing the PC into the store and that I would not void the warranty.

When I got the PC home again everything was working as it should, except the clock was now 15 hours slow, easy enough to rectify, reset the time.

I was very doubtful that humidity would cause such a thing to happen.
I have lived here in the tropical north of Australia for long time (10 + years in this house) and have always had an uncovered 6 foot fish tank in this room,
humidity is always high in here, as is the temperature most of the year round, generally in the 32*C - 34*C range.
No Air conditioner only ceiling fans, and my previous PC's have all been located in exactly the same place on my desk.

I have never had any other PC or electrical appliance have this issue, nor have I heard of anyone else having such an issue caused by humidity, had a fair look around on google and could not really find any credibility to this theory / explanation.

It was 9 days later before it happened again, exactly the same symptoms as the first time, rain overnight PC wont start in the morning.

So this time I re-seated the ram myself, PC still would not switch on.
Next I re-seated the GFX card, PC still would not switch on.
Then I pressed the bios button on the motherboard, PC still would not switch on.
So I pressed the bios button on the motherboard again to return to primary bios, PC still would not switch on.
Next I tried the reset cmos button, PC still would not switch on.
I removed the front panel from the case and tried operating the tiny power micro switch manually with a thin bamboo stick, PC still would not switch on.

So I decided I would try something a little different and outside the square so to speak,
I lied to my PC to make it think it had taken a trip into the PC store.
I removed all the external cables USB/KB/mouse/sound/HDMI including the PSU from the supply and left it all unplugged for about an hour maybe an hour and a half, about long enough for the PC to have been taken back into the store anyway.

Then I plugged everything back in again and what do you know, It worked, the PC started when the power switch was pressed the first time .... as you would expect it to.

The clock was out of wack again the date was still correct, but I figured this was simply because I had pressed the cmos reset.

This is where I started to think that there may be some credibility to the whole humidity being the root cause of the issue.
So I went and had a look at our local weather bureau readings. (www.bom.gov.au)

Relative humidity was highest at 96% at 6 am on this morning.
Relative humidity @ 84% at 8 am when I tried to first boot the PC.
Relative humidity @ 72 % at 9 am which was about when I last tried to boot it after clearing off the desk re-seating the RAM etc. ( still not working unplugged everything )
Relative humidity @ 64 % at 10 am, down to 54 % at 10:30 am, which is about when I plugged the thing back in and it just worked without any issues.

Today ( 11 days later ) it happens again, exactly the same as the 2 previous occasions, lights are on but nobody's home.
Only this time I figured I would first rule out the front power switch being the problem before I tried anything else.

So I removed the power switch header from the motherboard and shorted the 2 pins with a flat head screw driver, this did not work.
I thought maybe the screw driver was not conductive enough so I stripped a piece of copper cable and used bare copper wire to short the 2 pins, still the PC did not boot.
I reconnected the power switch header.
I then also tried the Start button on the motherboard, this too did not work, the PC would not boot even though it clearly still had power running through it, ( USB mouse still lit up, motherboard lights on).

I have one last ace up my sleeve to try and rule out that it was the high relative humidity,
before I go re-seating everything again or taking it back into the store for another warranty job.

I hit it with the missus's hair drier increasing the ambient air temperature thereby reducing the relative humidity.

I only had to run the hair drier for about 1 minute on the lowest possible heat/fan setting blowing warmish air all around the internals of case and moving the warm air flow back and forth all over the motherboard while intermittently pressing the front power switch every 5 seconds or so before the PC decided that it would boot normally.

I did not re-seat any components.

Yet once again the clock was out of wack ( correct date, but 3 hours slow ) this time there was no reason for it to be so.
This makes me think now ( just a theory ) that possibly the humidity is creating a moisture film between the cmos battery and its contacts which is why the time is out of wack, I just can not think of another explanation at this point.

Basically it would appear that the quick fix is to warm the ambient air temp with a hair drier and reset the clock.

I don't like it because, hair driers like vacuum cleaners are notorious for creating nasty static and should never be used on a PC.
I know a couple of people who have claimed to have killed their PC's by being lazy and vacuuming them out instead of getting / buying a can of air to blow them out, only to find that they never ever booted again afterwards.

I am looking for a better solution if anyone knows of one.
Thanks in advance to anyone who can help !

Until then I will have to just keep carefully hitting it with a hair drier every time it rains overnight when it wont boot in the morning ...
who knows, maybe this will even help someone else get their naughty misbehaving PC up and running again.

USE OF THIS QUICK FIX IS ENTIRELY AT YOUR OWN RISK !!!
81,028 Views
18 REPLIES 18

Zka17
Level 16
Now, that is an interesting experience!

One thing is sure: electronics does not like high humidity... It is clear that water condensates on some of the components and the short protection is kicking ON...

Man, that level of humidity is already harmful for humans too! If I would be you, I would look into some ways to lower the relative humidity in the room where you have that system... soon or later it will cause a short which will kill your system...

If the AC is not an option, there are devices to extract water from the air... or just setup your fans to run before you would turn on the system - maybe from an another PSU...

Gone_Troppo
Level 7
Well it has happened again this morning ... NO rain overnight, just nice and humid.

Removing the side panel and giving it a quick shot of warm air from the hair drier did the trick again, only about 30 seconds.

AC is simply not an option for me, I don't have one and can not afford one, even if I could, I could not afford to run it 24/7 as would be needed to avoid puddles inside the PC if the AC was to be switched off.

Trying to dehumidify the air here would be a pointless exercise like trying to hold back the tides in the ocean, or like trying to dehumidify the Everglades in Florida. (USA reference there for you)
Humidity is just a fact of life here.

The humidity does not affect any other appliance in my house, Including my missus's laptop, the PS3 and Wii which are all located in the same room, it has never affected any of my other old PC's.

When the PC booted the time was once again out of wack.
Started the PC at 8:00am, the time on the windows clock said 6:30am
This corresponds directly with the peak in humidity this morning, it was 96% at 6:30 am, but only 72% at 8 am when I tried to boot.

It would appear that with the limited data I have so far that the magic number to stop the PC from booting is 96% relative humidity, I will know more accurately as the wet season progresses, but it looks like this will have to be my practical everyday solution to boot the PC for the 8 months between August and April where the weather is always like this.

Even though it is a rather hazardous one with a serious risk of inflicting electrostatic discharge and I should not have to do this considering the supposed "high quality" claims of all the components in the PC.

The quoted price for this PC at the time of purchase AU$6350.00 including a 12 month warranty.
Just as well I only paid $250 for the excess on my insurance to replace the old PC when it got hit by lightning the second time which actually killed it, surprisingly it managed to survive the first lightning strike 6 months earlier with only a dead network port, Gigabyte GA-N680 DQ6 sli board had 4 ports so was a non issue, I was even able to salvage the HDD's out of it with no data loss after it was fried.

Flaming ultra sensitive high end "quality" product that is looking like it will not play nice in this tropical climate most of the year ... not impressed, but I am glad that once it gets going it is a nice machine to play with.

In the 5 months I have owned this PC I have also already had to replace the Corsair H100i under warranty due to the led's failing.
And while I waited for the replacement H100i to arrive the tech's at the store decided that it would be OK to fit a standard intel 2011 socket air cooler to the i7-3970X ... massive fail on their part, after 1 hour at idle cpu temps were at 65 degrees then 5 minutes of load blew them up to 85 degrees and I shut it down and returned it to them and directed them to intel's website where it clearly states that the 2011 socket air cooler is inadequate for this CPU, now a new H100i is back in it and they were very lucky I chose to monitor the temps and not just take their word for it that it would be OK, another few minutes and the CPU would have been cooked.

Anyone else living in a similar climate having similar weird issues with their PC not booting early in the morning ?
Anyone know where this short detection sensor is located and if it can be bypassed or disabled ?
though now that I think about that ... it's probably a worse idea than the hair drier.

Zka17
Level 16
The thing is that here in South Carolina we do have a very humid climate in the summer time... - but I can survive that only with AC... which in turn does keep the humidity lower too... don't dare to take my computers outside though...

The problem you're experiencing is due to metallic parts cooling down very fast and reaching the lowest temps while your system is off - and some condensation starts to happen... once you turn on your system, will warm up and it won't be the lowest temp for condensation...

So, if you can not have AC or an air dehumidifier (they exists and actually works in a closed room), then you can try the other option I suggested... get an another PSU and connect all your case fans to it... always power that one first and let the fans run for couple minutes - if you have a minimal condensation, the airflow may dry it up (based on the very short need of the hair dryer), then you can turn on the main PSU with the system on it...

Gone_Troppo
Level 7
Thanks for your input Zka17, I do appreciate it.

BTW just so you know the daytime temps will go up to around 35*C approximate daily min-max temperature range atm 24*C - 35*C (75F - 95F)
My 3 bedroom house is a concrete box and holds the heat all too well and is generally 30*C - 35*C (86F-95F) inside.
The PC / desk is located near the windows in an attempt to have some of the heat generated while it is on directed outside and unfortunately it is also located fairly close 1m (3 feet) from the western wall which has the sun baking it all afternoon and is pretty much the warmest place in the whole house, not exactly the best place to have it but the lounge room is the only place I have that I am allowed by my wife to have it, she at least wants to be able to see the back of my head sometimes, she says if I put it in the spare room she will never see me lol, and also the 50 inch plasma is used as our TV and also connected to the PS3, I would relocate it all to the other end of the house if I could.

Happy wife happy life as the saying goes.

The PC lost an hour and a half again last night (no rain) though it started normally when switched on at 9 am, (clock @ 7:30 am)
I'm guessing that either the humidity has been enough at some point overnight to break the connection to the battery again for a while yet evaporated and re-established contact before I had switched it on.

or maybe I just have an issue with the motherboard battery itself ?
I'm going to get the store to test it too before I proceed too much further on my own.

The idea of a second PSU and connecting it to either the existing fans or preferably using additional case fans would certainly be a much safer option than using the hair drier.
Although without the heat to change the relative humidity level it would just be air flow to evaporate the moisture as you have said which may yet prove to be enough and may also be achieved much cheaper by removing the side panel and using a 30cm (12 inch) pedestal fan to do the job it will only be a minor inconvenience if it works I will let you know.

Now if I was to go for this PSU option I would prefer to have it set up as a separate system that I would activate it whenever the PC refuses to boot normally and also would not need to be running a second PSU all the time while the PC is on so ... please correct me if I have this wrong.

I'm guessing I would need to rig up a permanent bridge of some sort to get the PSU to fire up and power the fans by connecting it to whichever relevant pins on the chosen PSU's 24 pin atx connector give/receive the start signal, in a similar way to as you would conduct the PSU paperclip test.
http://www.corsair.com/en/blog/testing-your-corsair-power-supply/

Then just use the power switch on the back of the PSU to switch it on/off
Correct ? ? ? I think this is what you are suggesting ?

Although this does not address the cause of the issue it will certainly be a much better option than the hair drier.
I will still have to reset the clock every time it happens which is only a minor inconvenience.

I think I may also have found another way which will possibly address the cause of the issue if it is in fact the humidity causing the problem which it seems to be.
Tell me what you think of this for an idea.

If I setup a low wattage incandescent or halogen globe (eg: a 10-40W desk lamp) inside the PC case to be turned on when the PC is turned off the warmth generated may be enough to stop any possible condensation from forming in the first place and I wont have to reset the clock every time as with the second PSU / fans solution.

The main problem I see with this idea is that it may possibly get too hot / radiant heat affecting the PC components.

The other considerations are:
1) The PC is still under warranty from the supplier until end of May 2014 and I don't want to do anything that may void it.
2) I am currently unemployed so the budget to deal with this is next to zero ... at least until I get a new job, which may be a while given the current jobs market.

The suppliers solution so far is for me to bring the PC into the store where they will reseat the ram/gfx card, which of course temporarily solves the problem as the store is air conditioned but I will see what else they can suggest as well as running the above ideas past them before I proceed.

Any additional suggestions or comments would be appreciated.
Thanks.

nikosa43
Level 12
Hi there Gone. I read about your problem and the first thing I thought, for a quick, easy and cheap solution is this. http://www.silicagel.com.au/ It's easy to make a test and not spend a fortune with experiments. Just find the right package and place it inside your case. Silica gel will absorb the moist from the atmosphere inside the box and you are good to go. I hope it will work for you 🙂

Gone_Troppo
Level 7
I had actually thought of that thank you for the suggestion anyway nikosa43.

I have used similar product available from the supermarket DampRid elsewhere in the house and also at my previous place of employment for moisture removal.

At work it was used to remove the humidity from the paper before it went into the printers.
Reams of paper had to be stored in large plastic tubs with the container containing the silica inside.
Helps to stop the paper curling, reduces photocopier jams and gives better results for the toner bonding to the paper FYI.
Even with the air conditioning they could not get the humidity below 65% in that building at this time of year.

You would probably be surprised to know that a pallet of paper can leave the depot down south travel ~3000km's in the back of a truck and arrive here 200 KG heavier due to the humidity, that's the equivalent of a 44 gallon drum full of water hiding in ~150'000 sheets of A4 paper with no signs of wetness.
Eventually they installed portable industrial dehumidifiers around the office and paper stores, that had hoses and holes drilled through the buildings walls to pump the water outside 24/7, but it did not make very much difference to the humidity levels in the air but it did help to dry out the paper faster.

As I said before trying to reduce the humidity in the air here is about as futile as trying to hold back the tides in the ocean.

Borg Collective: "Resistance is futile you will be assimilated."
My reply: "Assimilation is futile you will be approximated."
Sorry used the word futile and that popped into my head, back to reality.

Basically all you end up with is a plastic container full of water and it needs to be replaced or recharged quite often in this climate.

With the luck I have had this year, I would probably drop it and have it go all through the PC.

And as far as recharging it in the oven goes ... I will burn it ... guaranteed, if the wife would even let me put it in there in the first place.

Every suggestion welcomed, sorry if I seem to be shooting them down, I appreciate the help.

sectionate
Level 12
Well this is a red flag here,if they built these also -> "Apparently they had 5 PC's come into the store on the same day with the same issue" Almost all the hardware you purchased is under manf. warranty for 3+ years
Who did your build?
Crosshair IV 1090T 4093.7 - Crosshair V FX - 8350 Batch # 1243PGT 4.422MHz - 4816.08 MHz 1.5v - GSKILL F3-14900CL9D-8GBSR(x2)1087.6 11-12-10 1.5volts - ViewSonic VX2439wm - Acer GD235HZ - SyncMaster P2350 - PowerColor Red Devil RX 480 8GBD5-3DH/OC - Current Mode: 5760 x 1080 - (retired) ASUS GTX660DC2O2GD5 - Nvidia 3D Vision Wired - HAF 932 - Corsair HX850 - Corsair H100i - latest quakelive.com - Day thirteen 60fps - as always I'm powered by Futurelooks.com

Gone_Troppo
Level 7
Leading Edge Computers
Darwin NT Australia

They were the ones to build it, they also gave a 12 month RTB full warranty on everything.
I guess the remaining 2 years on the warrantable items I would have to deal directly with the component manufacturer myself.

Not my first choice ... but they provided the cheapest quote for the insurance claim. (AU$6350)
The other 2 quotes were AU$6800 + AU$7200 both with 12 month warranty.

The insurance company would not let me build it myself even though I could have bought all the components at retail prices including shipping for around AU$4500
I was only out of pocket $250 for the excess, insurance were picking up the tab and I don't care if they want to spend $1800 more than they needed to.
I gave them the option, they refused it.

It's only adult lego, rather expensive adult lego, but easy enough for those with the aptitude and self confidence to try it.

I also had pretty much the same response ... when the missus brought it home from the store after work.
"F'N BS IT'S HUMIDITY"
"What are these clowns trying to pull here"

They are happy enough to honour their warranty, however taking it in to their air conditioned store to have them reseat the ram and say everything is working ... must have got a bit of humidity in there, well I'm not going to do that regularly, I will end up in jail.

This is also pretty much the same response most people give when told or asked about it.
It is always an initial response of disbelief.
Which is why I am here looking for an answer, test to conduct or any other suggestion that may help me to get to the bottom of this issue.

However I can not yet rule humidity out as a possibility.
It goes against everything I know and I have lived my life experience in this climate.
The hair drier trick seems to confirm this theory of moisture, as does the clock losing time, every time it happens ...
Hazardous as doing this is for the PC it has worked so far, the moisture in the air makes it less likely to build/hold an electrostatic charge but it still can do so and it still can be very damaging to to PC as I'm sure you understand.

I can NOT stress enough that hair driers are as bad as vacuum cleaners and should NEVER EVER !!! be used on your PC.

What next ... get them to check for dry solder joints on the board ? ? ?
A common problem with surface mount tech that is mass produced, usually though that would cause total failure not intermittent issues.
Might explain why the hair drier works (heat re-bonding the joint) but I would think that it would take longer than 30 seconds with a hair drier.
though it doesn't explain why leaving it until later in the morning/day also works.

Poor quality / malfunctioning short circuit detection unit ... I doubt it, anyone know if I can test it ?
Over sensitivity to high humidity, maybe a design flaw for ASUS to rectify.

I have tried pretty much everything I can think of, and will keep trying to find the fault as long as it takes.
At this point though it would appear that humidity / condensation / moisture film forming on the cmos battery is the problem.

To quote from Sherlock Holmes "Eliminate all other factors, and the one which remains must be the truth."

Thanks for all your input and reading through the walls of text, keep 'em coming.

Zka17
Level 16
What if you would insulate your mobo like for sub-zero cooling?

Use some liquid electrical tape...