Two New Ways To Totally Beat Condensation

[r0ckstarbob]

***edit***

*chuckle*
well this thread has gone through alot of changes and evolutions. i'm adding this now just cause it's gone on for so long.

Conformal coating works very well but is supplementary protection against condensation in that it won't prevent condensation, but it will protect from it.

Drierite is a very good option for sealed or VERY low airflow cases.

these two items aren't the sure-fire end-all be-all answer to beating condensation like i first thought they were but they're some really fantastic products to think about when doing the things that we do, and will be a good read.

if you read the whole thing, you'll come to that conclusion like we did, but thought i'd come back and summarize while i was thinkin about it.

peaz
RSB

***/edit***

christ i know i sound like a commercial about this crap and for that i apologize. it's not intentional, i promise. the stuff just WORKS and nobody knows about it yet because overclocking doctrine says to make one of those big nasty blocks of white foam or to use dielectic grease or some crap like that. there IS a better way. this is it.

read on...

Acrylic Conformal Spray

http://www.marvac.com/products/mgchem/419b.gifhttp://www.techspray.com/images/210312f.gif

http://www.mgchemicals.com/products/419b.html
http://shop.store.yahoo.com/webtronics/419b-340g.html
http://www.techspray.com/2103info.htm

it's an epoxy-like spray on coating that is very thin, completely dielectic, completely transparent, strong, it yields to a soldering iron immediately and it is DRY. it has a flourescent compound in it that will glow green under a black light so you can check for gaps in your application. it's used to protect aircraft electronics from condensation and a hundred other little things. coat the front and the back of your board and spray a little dab in that gap under the CPU and let it dry for a couple of hours. the spray is about 10 bucks and it is 100% effective. it won't prevent condensation from forming but it WILL completely eradicate the possibility of it doing ANY kind of damage whatsoever. used in conjunction with the stuff below and you've got a waterproof system.

Drierite (Desiccant filters)

http://www.speakeasy.net/~language/pictures/drierite1.jpghttp://www.speakeasy.net/~language/pictures/drierite2.jpg

http://www.drierite.com/default.cfm

other stuff it's called is Silica Gel. it's a clay out of norway that sucks up water from air and lowers the dew point to -100F. no moisture in the air means no condensation can happen. ANYWHERE, under any conditions. make a filter and fit it over your fan and point it at your CPU if you're really worried or you can just set some of it inside your case. this stuff is REALLY powerful, it's also really cheap (13 bucks for a 5 pound can, 5 ounces per 100 cubic feet is the recommended amount - your case is NOWHERE near that big), completely dielectic, and although it does reach a saturation point after awhile and you have to change it out once in awhile, you can recycle it by putting it in your oven at 250F for a couple of hours and then you can reuse it at 100% efficiency. the word from the manufacturer of this product is that they will guarantee dryness and a lowered dew point to -100F.

if there is no water in the air then condensation cannot occur. no matter how much heat is applied to how much cold. condensation WILL NOT FORM. not at a -100F dew point level. that means anything below -100F (say -101F) is below the dew point and will make condensation. any temperatures above that will not make condensation. it requires a little bit of jr high science revisited to know about condensation and dew points and stuff like that to know why this stuff works but be guaranteed it definately does.

check out the Acrylic Conformal Spray before you start to covering your board in that gooey crap. either one or both of those options are going to work ALOT better for you and theres no mess involved.

take care and good luck.

RSB

i went into it a little bit better here.. HERE (http://forums.overclockers.ws/vb/showthread.php?s=&threadid=34296) this is the repaired link i dunno where it was pointing before

ya know, i'm just going to write an article about all this stuff and submit it to the forum here. maybe they'll publish it... what i know about this stuff and the little i'm able to expound upon here in individual posts are getting to be infuriating. i dunno. anyone want an article about condensation and how it applies to computers, computer cooling, and extreme computer cooling and the various ways to beat it? just say the word and i can cook one up that everyone will be able to understand...

[Thelemac]

Yeah! That stuff sounds great!

Thanks for the info...I'm definately looking into that when I go into my next level of cooling. :)

[Silversinksam]

Anybody know where you can get Acrylic Conformal Spray at a reasonable price in the USA? Seen it at a few places for $19.95 a can. Any help on this appreciated

[r0ckstarbob]

http://www.marvac.com/products/mgchem/419.htm
http://www.marvac.com/products/mgchem/419b.gif

this appears to be the same stuff for 12 dollars

and i don't know which one works better but theres this brand too...

http://www.techspray.com/2103info.htm

http://www.techspray.com/images/210312f.gif

[r0ckstarbob]

oops... i guess it's 15 or 16... whatever. it completely protects your whole board. and besides, remember that flourescent compund in it that shows up under blacklight so you can check for gaps in the application??? guess what that means? for those of you who have blacklights inside your case, in addition to protecting your motherboards, it'll make your board glow green!!!

wee!!! neato cheato. :D :cool: :cool: :cool:

[Silversinksam]

Hey thanks Bob,

Been meaning to buy conformal coating spray for some time.

Appreciate the linkage.



PS What a horrible day today has been and my thoughts and prayers go out to those that may have lost a friend or loved one in todays horrific events.

[r0ckstarbob]

Originally posted by Silversinksam
Hey thanks Bob,

Been meaning to buy conformal coating spray for some time.

Appreciate the linkage.



PS What a horrible day today has been and my thoughts and prayers go out to those that may have lost a friend or loved one in todays horrific events.

i hear that brotha. my prayers go out... and my hopes that nobody here on this forum was hurt or lost loved ones...

[Silversinksam]

Looked up that part # and found it for $9.95

http://shop.store.yahoo.com/webtronics/419b-340g.htmlhttp://us.st9.yimg.com/store4.yimg.com/I/webtronics_1647_13288728

[r0ckstarbob]

see... i THOUGHT i saw that stuff for 10 bucks.

thanks!

[touser]

wow this stuff sounds great! i will have to go pick some up, i have been wanting to use chilled water in my rig for some time and now i have no reason not to!

[r0ckstarbob]

dear god man i think you may have the most powerful stock system i have seen in a long time. whats an Abit Th7 though?. never heard of that one...

**jealous**

wish AMD would drop a 1.7 gig chip... :(

[Random Nonsense]

by the way, if you want silica gel its the beads in those little packets that come with all computer hardware.

i been thinking of using the stuff for ages, but cant see how to get any of it into areas where condensation is a prob....

[funnyperson1]

Originally posted by Random Nonsense
by the way, if you want silica gel its the beads in those little packets that come with all computer hardware.

i been thinking of using the stuff for ages, but cant see how to get any of it into areas where condensation is a prob....
just tape a pack or so onto your case

[adamtekh]

a 1.4 athlon will perform as good or better than a PIV 1.7

it realy depends on what apps your running

[r0ckstarbob]

Originally posted by Random Nonsense
by the way, if you want silica gel its the beads in those little packets that come with all computer hardware.

i been thinking of using the stuff for ages, but cant see how to get any of it into areas where condensation is a prob....

make a filter and fit if over a fan pointing at your CPU or just set it in your case. silica gel is one kind of desiccant. Drierite is like silica gel on meth. Its a desiccant, but more pure.

adamtekh - yah i know but it requires s00per stud cooling. not that it's a problem but....

you get the idea.

[r0ckstarbob]

you could even grind it up and powder your board with it... the stuff never gets soggy. it transforms the water into another inert chemical. literally, you can't tell the difference between saturated drierite from fresh drierite by feeling it. drop a little powder under your cpu in that little pocket in the socket there.... voila. good to go.

[Fink]

As a chemist I need to say something.

Dryrite is not silica gel, its CaSO4 (calcium sulfate). It is a slow acting dessicant (drying agent) used primarily to slowly remove moisture from dessicators, which are usualy sealed large glass enclosures - think glass tupperware. The problem with CaSO4 as a dessicant it that it is Slow - it will take 24 hours to take a sealed container to 5-10% humidity. Plus it does not hold much water, 1-2 equivalents. Blue dryrite has an Cobalt indicator in it that turns pink when it gets saturated with water.

Silica gel is also a dessicant, IF it is baked at 250F for 4-6 hours. The problem is that it also does not hold much water and would need to be changed often (like every day).

Chemical dessicants work great in static enviornments. For places where air is being constantly circulated, the moisture in the air would quickly saturate and overwhelm the capacity of the dessicant. Plus there is a disposal issue and if the silica gel is of a very fine consistency it is a serious inhalation hazard.

Cheers.

[Random Nonsense]

and i know silica gel doesnt work indefinately, thats another prob, it eventually gets saturated, in aircraft uses it starts of blue and turns to pink when saturated.. i could snaffle some from college i suppose.... although im no longer thinking of using pelts, i could send anyone some if they want the changing colour stuff.

[r0ckstarbob]

wow

very cool. i'm glad someone out there has some kind of experience with this stuff. i'm going off of what i'm able to interpret off of the data sheets in combination with the emails i've gotten back from them about this product. well, maybe as an active prevention method of condensation, it is a bust... oh well., live and learn i guess....

in a sealed enclosure, would somethig like a desiccant scrubber work? something like a box with 4 fans. two fans on one side sucking in air, and two fans on the other side sucking it out again, thrusting the air over some Drierite filters???

...thinkin about stuff

[Fink]

No, blowing 'wet' air over dryrite or anhydrous silical gel will quickly overwhelm the carrying capacity of the absorbents. Plus, neither works very fast and most of the moisture would pass through the filter without being caught. Chemical methods are tough, an electric dehumidifier might be a better idea....

[r0ckstarbob]

too much bulk and expense. besides, i know how they work. it's pretty much identical to a air conditioner except a dehumidifier reheats the air before it vents it where as an AC unit doesn't.

even in a totally enclosed environment? like say you need 2 cubic feet of air dry dry dry and needed it done as soon as possible. sealing it up and turning the scrubber on wouldn't speed up that process? i don't care about depeleting the drierite, i just want to get the space s00per dry so when i turn on the refrigeration unit and the temp plunges to -30 F i'm not going to get ice all over my waterblocks and walls and stuff... the scrubber isn't sucking in fresh air, it's just cycling the air thats already sealed inside the enclosure round and round through those filters for a couple of hours before i kick on the refrigeration cycle...

[IFMU]

I must admit that Ive never wanted to deal with peltiers before due to the difficulties of having to deal with the condensation... But the more I read on the Drierite the more tempting it seems... No insulation to worry about... day by day more and more i think I might actually have to start a thread on how and what to do so I can setup a peltier set-up... :D

[r0ckstarbob]

well, while the efficiency of the drierite doesn't seem to be in contention, the practical application of it seems to be. if you're going to use it i'd certainly recommend a low airflow kind of case (no aircooled peltier action going on, strictly a pelter/watercooled sort of thing) and some maintenance on your part to make sure the drierite remains efficient. this stuff is certainly going to work better in a static environment apparently. the acrylic conformal spray however i would say is definately a must -have as it's going to utterly remove the possibility of condensation doing any kind of damage to your board even if it does form. besides, it'll also protect your board against other gunk like metal shavings and stuff like that as long as you coat it ahead of time. definately a good thing. god knows our boards could all use a little protection from accidental slipups. 10 bucks to save a fortune in ruined boards, a small price to pay, i say. and besides, it glows green under a blacklight fer chissakes! i mean really, tell me thats not cool... :D :D :D

good luck and let us know how it goes if you decide to go that route.

RSB

[Ottoman]

well the problem is the cool surface hitting warm air,

so either u have to seal it in a vacuum, or u havta prevent the cold surface from touching the air,

similar to thermos insulation, so u have to use a very poor thermal conductor and put it between the to mediums,

like covering the waterblock area with styrofoam or something,
so that the coolness will be trapped in, and the air can't touch it...

the acrylic is good to proctect parrts, but I wouldn't want water pooling on my board...

but I think u def. have something there... there's potential, just need the bugs ironed

[r0ckstarbob]

Originally posted by Ottoman
well the problem is the cool surface hitting warm air,

so either u have to seal it in a vacuum, or u havta prevent the cold surface from touching the air,


actually there is a third method, and thats the one i'm trying to explore. if you remove the moisture from the air then you can beat condensation that way too. by removing the condensation you drop the dew point.

warm humid air + sufficiently cold surface = condensation.
warm dry air + cold surface = no condensation

and yup. i see you understand the problem of using one without the other. using strictly the acrylic conformal spray, you're not preventing condensation, you're just preventing it from doing any kind of damage to your board.

hmmm. well, theres kinks to be worked out. and on i go! :)

[Ottoman]

yeah, I've also heard that moving the air helps,

eg. static air has more potential to condensate than moving air...

so even a fan directly blowing over will reduce it

[r0ckstarbob]

yup yup. thats a fact jack.

[samuknow]

We use a product like this in production here at work. Of course it's industrial but the same type of product. We actually have a method of coating the entire board and elevating the temp and reducing the moisture to improve curing time. This stuff is for high voltage and is meant to stand up to being left outside in the rain. I have been considering using this for some time when I go ahead and build my Tbird system with a pelt.

[r0ckstarbob]

could you elaborate for us exactly what you're doing again? this sounds very VERY interesting actually.....

[samuknow]

We build high voltage PCBs. Electric fence controlers to be exact. Part of the process to pass UL is to weather proof the boards. The process is to hang the board on a motorized chain that goes to be coated then is passed through a low temp oven to BAKE the conformal coat. This stands up to some severe conditions. I am the manufacturing engineer here and I have run some extreme condition testing and they always pass. 150F heat with 90% humidity along with direct spray of water. I would imagine if it could pass this test it would stand up to some condensation.

[r0ckstarbob]

dunno why the forums do this, but after so long, if you post in one of these things after awhile, it doesn't send you an email letting you know that someone has posted in the thread.

at any rate. WOW. thats VERY cool. apparently the stuff really DOES work quite well then. hmmm.

rad. that seriously makes my day.

:)

[TOMATOMAN]

Man if that gell stuff could hold more mosture it would be great to put in a bong coller where the fan blows in :} I coould only inmagine the cooling u could get with no mosture!!

[71skylark]

Wouldn't you have to be very careful you didn't fill in any of the holes in the socket or in the dimms for memory? That would be a big fear for me with the conformal spray.

[kevin_bouchard]

does the conformal spray insulate heat(traps it like paint) if so it would be hell for all those chips.

[r0ckstarbob]

no. that doesn't seem to be a problem.

[kevin_bouchard]

sweet i will get some as soon as i can.
i was wondering how i was going to insolate my northbridge and gpu now i dont have to worry about it, i am so happy:D

[IFMU]

Actually that is a good question there... How do you spray it on the board? Do you mask anything off? or just spray it everywhere... which I dont see how it can protect from water damage but still let elecricity conduct through it and not hamper the speed... I understand that the only parts you have to worry about being to transfer electricity.... but what about the insides of the PCI slots? that kinda thing... the socket? for the CPU?

sorry just a lil curious what directions have been taken before...

[JML]

Say I decided to take my mobo out and spray it all over with acrylic conformal coating, would it be okay if it got into any of the slots or the socket? And I think even with this stuff on someone should still put something on to prevent the condensation. Would this spray save your system, if say there was a leak in your watercooling system?

[kevin_bouchard]

I dont think it would be the greatest thing to put into the isa/pci/agp slots or cpu socket because it acts like an insulator of electricity and the contacts may not connect right. Personally what i am going to do is spray my whole system, then put some dielectric(and non-acid)grease into the sockets and pci/agp ports.
Oh ya it may be a good ideal to cover all the jumpers and connectors/ports on the board so that they will still be able to conduct electricity.
My only real concern is if it will trap heat it could easily turn a pc unstable and this stuff probably doesnt come off with out a fuss.

[r0ckstarbob]

yeah, i can definately see where you're comming from and it seems you're on the right track in taking the proper precautions. you'll want to dummy your slots closed and cover your jumpers in something like vaseline or dielectric grease or something and then spray the board so it won't interfere with the connections. as far as it thermal insulation of the board, according to the mfg who i emailed about it awhile back, they said that it wasn't a problem - a claim that seems to have been substantiated by comments from others who have used this product in the past, some of whom have commented earlier on in this thread.

it's got a maximum temperature though and should say right there on the web page as to what it is. you won't want to coat the super hot parts of your board with it, but for the rest it should work great.

[kevin_bouchard]

vaseline, i thought that brakes down some plastics(is this true).
Thanks for the help:D

[FuzzyNY]

I did the acrylic conformal coating part and I wasn't very carefully that I sparied some of my PCI slots, after I boot up the PC, my sound card didn't work, my mouse didn't work and the system could only detect 128MB of memory(I have 512MB), then I pulled the sound card off and I saw lots of the "coating dirt" left on the pins, after serveal times of pull up and push in, my sound card works again, so I concluse it was the coating left in those slots.

another thing is don't use fan to blow the spraied MB just after the coating, wait 20 mins before using fan to dry it more quickly(I wasn't patient enough:)), otherwise you'll see a lot of small air bubbles inside the coating

except this, this is a very good product for conformal coating the MB, thx for the links.

i tried the Drierite, as Fink said it works but very slowly...



Fuzz

[IFMU]

So you actually have tried the Drierite? How well is it at pulling the water outta the air? have you actually tired to set up a peltier setup that would actually cause condenstation and try the drierite? sorry but Im curious how it came out...

[Warlord2]

I just went to a audo parts store today and I was looking at sprays

I found this rubber undercoating that looked pretty good

it was only $5 for a big spray bottle of it

it didnt say if it was dielectric or not but I asume its not sense its made of rubber

[r0ckstarbob]

Originally posted by ifmu
So you actually have tried the Drierite? How well is it at pulling the water outta the air? have you actually tired to set up a peltier setup that would actually cause condenstation and try the drierite? sorry but Im curious how it came out...

just an FYI,

the more i talk with those guys over at the drierite, the more i'm pretty certain that using drierite in an open or even low airflow case probably wouldn't be such a hot idea like i'd first thought. i mean it'd work but you'd have to recycle it almost daily and is probably doubtful that you could get the the dew point low enough for long enough to use it very efficiently.

drierite is definately a good option for a closed/sealed case, (and for my purposes seeing as i'll have a sealed case and using the drierite in a scrubber - perfect).

looks like i got a little overenthusiastic. my bad y'all.

[FuzzyNY]

yes I am currently using 152Watts peltier with water cooling, I experinmented it by puting a tiny bit of water at the side a bunch of Drierites, it did dry the water but in a very slow way, I think it did works but only in a way that help to dry the air, don't use it solely with peltier system.

and the reason I didn't put it in my case is because I did pretty good job to prevent the condsensation and I also have a 120mm blow hole direct blowing the cpu, so I don't think it will help much to put a bunch of Drierite in the case



Fuzz

[FuzzyNY]

oh I see r0ckstarbob, it didn't work so well because I experimented it in open area


Fuzz

[r0ckstarbob]

hm

drierite doesn't work by actually absorbing a large amount of water like a puddle, so putting a little puddole on the talble next to the drierite isn't really a good way to test it. it's not going to absorb straight water very well. the idea behind the drierite is that it drys the water in the air in order to lower the dew point sufficiently so when you DO turn your peltier on, condensation won't form in the first place.

hope that clears things up a bit.

[FuzzyNY]

I understand now thanks, but if it needs a seal case, is that mean that we cannot use any fans/blowholes? and if this is the case then it will increase the case temperature a lot, just my thought

opps you've already mentioned that, nevermind
Fuzz

[kevin_bouchard]

hey warlord is that the stuff they put under cars?
If it is its normally black and it is a fairly thick coat, which could possibly cook some of the chips on your mobo.
But if its not good luck :D .

[Warlord2]

hey warlord is that the stuff they put under cars?

yup it is:(


I was just looking for something cheap


thanks for the warning:)

[r0ckstarbob]

Originally posted by FuzzyNY
I understand now thanks, but if it needs a seal case, is that mean that we cannot use any fans/blowholes? and if this is the case then it will increase the case temperature a lot, just my thought

opps you've already mentioned that, nevermind
Fuzz

yah, i mean you CAN use it in a ventilated or unsealed case but you're going to have to change it out and recycle it the more airflow you have i'm afraid.

[kevin_bouchard]

try it on an old isa or a crap card, the way i would test it is by putting the card in and taking the temp of a chip that heats up fairly good, and then spray the stuff on the chip and then put the card back in the system and check if the temp has dropped alot, if so this isnt good because you will know that the heat isnt escaping. You could still use this stuff but it will be harder to use(tape up all the top of chips) and if its black i dont think you will want a computer that is only black(or do you?). Also note the thickness(the thiner the better).
But for the price of conformal sprays, unless you are really on a strict budget I dont see any reason to use something else.

[r0ckstarbob]

Originally posted by kevin_bouchard
try it on an old isa or a crap card, the way i would test it is by putting the card in and taking the temp of a chip that heats up fairly good, and then spray the stuff on the chip and then put the card back in the system and check if the temp has dropped alot, if so this isnt good because you will know that the heat isnt escaping. You could still use this stuff but it will be harder to use(tape up all the top of chips) and if its black i dont think you will want a computer that is only black(or do you?). Also note the thickness(the thiner the better).
But for the price of conformal sprays, unless you are really on a strict budget I dont see any reason to use something else.

well i don't know about the automotive undercarriage sprays (which i believe to be black) but the links to the acrylic conformal sprays that i provided are both transparent and very thin. they've also both got flourescent compounds in them that will glow green under a blacklight so you can check for gaps in your application. i do know that urethane conformal sprays are opaque (cannot see through them). the fact that the acrylic conformal sprays are transparent is what drew me to them in the first place.

[kevin_bouchard]

Oh maybe i should of directed my last post to warlord2 to stop the confusion, sorry:(

One of the main reasons why i would use the confromal spray you posted is because i have a blacklight and it would look sweet:D (oh and to avoid condensation problems).
Has anyone actually tried to use this stuff yet?

[bsblues]

wow I am glade i am not crazy...

I saw this stuff at http:\\www.mouser.com

I figured this would be a godsend to overclockers

[Warlord2]

Im thinking about buying the acrylic conformal sprays now but dont you still need insolation of some kind? to keep the condensation from making a river down your board?

does the board get hot enough to evaporate whatever condesation hits the board?

I was thinking of spraying some insolation foam by the cpu just for a precaution but if I dont need it I dont want to buy it=]

[r0ckstarbob]

no, you're right. unless you have a sealed case (in which case you can use the drierite) you'll need some kind of insulation to keep a river from running down your board. the conformal spray is supplementary protection to prevent condensation from damaging your board in the event you DO get some.

[FrozenInHI]

Hi everyone!
Well, yes i'm new around here, but i'm not new to computers and hardware. I've just recently found this site and started reading around at all the tweaking everyone's done. Now on to the topic. I have a bit of input you might be interested in. I am a micro/miniature electronics repair tech and use conformal coatings regularly. I use Dow Corning's in my computer which is watercooled and soon to be peltier cooled as well. I've set up a peltier cooling solution in a friend's system fairly recently (2months ago) and I'll give you a few tips myself:

1-Conformal Coating is the easiest/best way to insulate the board from water while maintaining thermal dissipation of all the board's components, but:
a) you must make the layer of coating very thin, nothing
thicker than a sheet of paper will work well at all. Remember there are resistors on these boards and you must know that they are designed to drop voltage as well as dissipate heat, very important for stability in voltages.
b) you should avoid coating the tops of any ICs, these are not designed to be insulated from electricity at all, they in fact need to give off heat or they will deteriorate. Coating the "feet" or pins of them would be optimal.
c) it would be better to use dielectric grease in the pci/agp/memory slots, for obvious reasons.
d) do not pay more than 15 bucks for a decent (10oz) can of coating, it is very cheap to make and they can and will try to rip you off in places where you find it, the links at the beginning are pretty good pricing scales.

2-you must not try to use only conformal coating and not insulate your chilled or peltiered cooling solutions, this is suicide (or murder, however you want to look at it) to your pc. No amount of coating will save you from shorting things out in there, it's damn near impossible to coat all areas of the board, relief, or surface, mounted components have areas that are tough to coat without industrial equipment designed for this. Manufacturers tend to mount some of the components, spray it, then mount some more, spray it, then mount the ones that do not need coating, so don't try to get away with a "blanket" of the stuff, i've seen this done and it ended up being a dead RF passive limiting circuit board in an expensive weapons system (I work in the military).

3-This may have been covered in the other link, didn't have time to read it just yet, but i'll get to it. One place you might want to coat VERY thickly, and it's not easy with a spray can, unless you spray the liquid into a container or something for "painting or pouring" it on, is inside the cpu slug, and the back of the mobo, where the cpu socket pins mount through the board. that whole area is virtually free of high temp components and you can be free with the coating, I put almost 1/16" of it around the cpu socket, inside the socket, and on the back where the socket mounts, and it still got a tiny bit of frost, i reapplied about another 1/16" and it worked flawlessly. I live in Hawaii where it's humid though, so take that into consideration.

4-The drying stuff, Drierite, is awesome stuff-for shoe salesmen and retail packaging companies. Dessicants were not designed for this application, with the obvious exception of the sealed casings, these would benefit greatly from it, provided they could make some sort of "air tower" see the vent design at the site for further details on drying towers. It's just not efficient enough for our uses, we move more air than they were designed to clean. there is no easy way to dry air for this application, aside from a large dehumidifier. I'd not worry about humidity in the air, i'd just protect myself from it, cheaper, more effective, and certainly a better solution IMHO.


Well, thats about it for now, feel free to ask me anything that might fall into my line of work. I've got tons of experience in this field, i've done it for 5 years now and I've seen a ton of solutions in electronics, also a ton of mistakes. And if you think this is all bullcrap, then that's fine too, like i said, i'm new here and you can take that for what it's worth to you. I look forward to learning more with you all and hopefully bringing some interesting info your way, as you all have brought lots my way. Sorry to make it a long post, just felt compelled since it really hit home for me, i have conformal coating under my nails daily it seems, and removing it with acetone doesn't work when you use the chem resistant stuff too! lol!

[robertm]

one thing about this is it will rune your connectors you must tape over you PCI slots and memory. Or put it in and leave it in the slots. Once this is in the connector you cant clean it out. Well you can but you would just junk it its very very hard. ( I know I have not done a motherboard but I have other devices.)

other bad thing about these kinds of products is they tend not to lend themselves to helping cool. Its one more layer of a paint really that you have to cool though.

Just wanted to warn people so they don't get a bunch of messed up boards.

Just saw FrozenInHI post just take mine as reinforcement :) Its not a cure all by any means :)

[r0ckstarbob]

Originally posted by FrozenInHI
Hi everyone!
Well, yes i'm new around here, but i'm not new to computers and hardware. I've just recently found this site and started reading around at all the tweaking everyone's done. Now on to the topic. I have a bit of input you might be interested in. I am a micro/miniature electronics repair tech and use conformal coatings regularly. I use Dow Corning's in my computer which is watercooled and soon to be peltier cooled as well. I've set up a peltier cooling solution in a friend's system fairly recently (2months ago) and I'll give you a few tips myself:

1-Conformal Coating is the easiest/best way to insulate the board from water while maintaining thermal dissipation of all the board's components, but:
a) you must make the layer of coating very thin, nothing
thicker than a sheet of paper will work well at all. Remember there are resistors on these boards and you must know that they are designed to drop voltage as well as dissipate heat, very important for stability in voltages.
b) you should avoid coating the tops of any ICs, these are not designed to be insulated from electricity at all, they in fact need to give off heat or they will deteriorate. Coating the "feet" or pins of them would be optimal.
c) it would be better to use dielectric grease in the pci/agp/memory slots, for obvious reasons.
d) do not pay more than 15 bucks for a decent (10oz) can of coating, it is very cheap to make and they can and will try to rip you off in places where you find it, the links at the beginning are pretty good pricing scales.

2-you must not try to use only conformal coating and not insulate your chilled or peltiered cooling solutions, this is suicide (or murder, however you want to look at it) to your pc. No amount of coating will save you from shorting things out in there, it's damn near impossible to coat all areas of the board, relief, or surface, mounted components have areas that are tough to coat without industrial equipment designed for this. Manufacturers tend to mount some of the components, spray it, then mount some more, spray it, then mount the ones that do not need coating, so don't try to get away with a "blanket" of the stuff, i've seen this done and it ended up being a dead RF passive limiting circuit board in an expensive weapons system (I work in the military).

3-This may have been covered in the other link, didn't have time to read it just yet, but i'll get to it. One place you might want to coat VERY thickly, and it's not easy with a spray can, unless you spray the liquid into a container or something for "painting or pouring" it on, is inside the cpu slug, and the back of the mobo, where the cpu socket pins mount through the board. that whole area is virtually free of high temp components and you can be free with the coating, I put almost 1/16" of it around the cpu socket, inside the socket, and on the back where the socket mounts, and it still got a tiny bit of frost, i reapplied about another 1/16" and it worked flawlessly. I live in Hawaii where it's humid though, so take that into consideration.

4-The drying stuff, Drierite, is awesome stuff-for shoe salesmen and retail packaging companies. Dessicants were not designed for this application, with the obvious exception of the sealed casings, these would benefit greatly from it, provided they could make some sort of "air tower" see the vent design at the site for further details on drying towers. It's just not efficient enough for our uses, we move more air than they were designed to clean. there is no easy way to dry air for this application, aside from a large dehumidifier. I'd not worry about humidity in the air, i'd just protect myself from it, cheaper, more effective, and certainly a better solution IMHO.


Well, thats about it for now, feel free to ask me anything that might fall into my line of work. I've got tons of experience in this field, i've done it for 5 years now and I've seen a ton of solutions in electronics, also a ton of mistakes. And if you think this is all bullcrap, then that's fine too, like i said, i'm new here and you can take that for what it's worth to you. I look forward to learning more with you all and hopefully bringing some interesting info your way, as you all have brought lots my way. Sorry to make it a long post, just felt compelled since it really hit home for me, i have conformal coating under my nails daily it seems, and removing it with acetone doesn't work when you use the chem resistant stuff too! lol!

frozen-

thank you mr man for the time and effort you've put into your in-depth and very knowlegeable reply. i'm not, by far the best expert in the field on this stuff (or desiccants for that matter) though at the time - when i posted it a couple of months ago - nobody i'd ever heard of had either considered it or brought it up on any of the three forums i'm on so thought initially that i'd just found a new 'cure' to the condensation problem that ails us.

turns out that yes, it's one technique, but not for everyone as i first thought. regardless, it still seems like good information to have and to talk about here - having some options and knowlege of several possible techniques when approaching a problem (ie the issue of condensation) is far better then having just ONE cure for cancer so to speak.

RSB

PS good info is good info. age is no longer a bar to wisdom, as is our post count. newbie or not; experience, wisdom, and common sense and will make itself apparent - and yours is readily apparent as far as i can tell. thanks again for the knowlege and your first hand experience and let me be one of the first to welcome you to the forums FrozenInHi.

[Tiger]

This may seem a very simplistic approach but I'll put my thoughts in for what they are worth.
You are using sub-zero cooling i.e. the cooling solution is sub zero. In order to prevent condensation the air needs to above 13C (dew point). I read an article recently by someone who was using a Pelt and he tried running it with the air blowing out around the CPU in the conventional fashion and he had no more condensation problems.
The problem lies in the fact that the air is being cooled below 13C and therefore you are getting condensation. Why not insulate the water block with lagging and then blow warm air over the area using a reversed pelt.

[r0ckstarbob]

you're close, but not quite there.

condensation forms when warm air hits a cold surface, not the other way around. its not the AIR that would need to be above 16C, it would be all the surfaces i'm cooling down. kind of defeats the purpose, ya know? and that dew point varies depending on how much humidity is in the air. i hope to be able to have system wide cooling at -20C to -25C or colder (i've got some leads on some REALLY amazing refrigeration processes i'm exploring currently).

i guess for me, i've got my solution to the condensation issue and don't think theres a better one to be had (without replacing the internal gas with something else or trying to foam block everything that will recieve subzero temps as a result). maybe this thread and this technique will be of some help to someone doing something similar, or maybe it will spark a different idea in someone to try similar methods in a different technique. i wish this was the sure fire cure-all that i thought it was, but i know now it's not (at least not for everyone). maybe some good will come of it somewhere. one can only hope i guess.

live and learn. and so we go.

RSB

[FrozenInHI]

Thanks RSB! I've found tons of info on these boards that's been very useful, so I really want to try and give back where possible. Anyhow, the dessicant looks like a great solution for you, as you will have a sealed environment, If you have any results, do tell, because i'd be intrigued to know. Id consider sealing my case, i've got a clear one around here somewhere that would work well for it, but it's really hard to do here in Hawaii, as I'd have to cool the air and have some sort of scrubber as well.

Insulation-I've got an application for a TEC in my system at work, it uses about an 80watt peltier with a waterblock on it, not really like the ones in computers, but similar, it's more of a modular design instead of hoses and blocks, this thing's got a water jacket around the entire structure. The insulation used is pretty simple really, they put 2 part foam there. Truthfully, it looks like that spray foam you find at auto parts stores. It doesn't look like it would be too pretty to apply, but it does look like it would work well, i might consider trying that when my pelt gets here, have to think about that, maybe experiment with it's effects on mobos and how easy it is to remove if i don't like it, that stuff's sticky.

[hunter00]

here are my two cents . . .

drierite is great stuff, anyone who does anything in a chem lab knows that.

buttttt. . . i also know that the only place you really use the stuff is in a dessicator, usually an air-tight one.

i don't know if you know what a dessicator is, well, its like this big thing that you put drierite in and make air-tight. its big, glass, and you slide the top of sideways (if you lift it, the whole thing tends to lift with it, then release when you're a foot in the air, which makes a lot of little dessicators :) )

it also has this thing on the top you attach to a vacuum and then kinda rotate to seal, (usually use one of those water vacuum things, forget what they're called).

i know right now in research im using tetraethylorthosilicate which will hydrolyze if left out, even with a sealed cap, so we seal the thing with parafilm then put it in a vacuum-sealed dessicator filled with drierite to keep it dry.


now why did i say all this, well ...the moral-type thing as far as i see it is, drierite isn't particularly useful unless its in a dessicator(or similarly sealed container i guess), thus i'm guessing it would be close to useless in a case with any kind of airflow.

i of course am not positive about any of this, its all from lab experience (im only in high school, but pretty into chemistry research).

anyone with more experience want to comment? or maybe someone has a workaround type of suggestion?

[r0ckstarbob]

again, more good info.

yeah, i know what a desiccator is now and is more or less what i'm building with the core project. according to the brochure, it doesn't sound nearly nearly as tempermental as apparently it is. ah well. thats the difference between the sales brochure and practical experience i guess huh? i'm building an utterly sealed, low pressure case (due to the internal temperature drop after it's been turned on) with an integrated desiccant scrubber to speed up the process of dehumidification so when the Core hits -30F, i'll be good to go.

[hunter00]

so you're gonna make a completely sealed water cooled/dessicator/peltier case?

well . . .i have to say it sounds like a pretty cool idea, i definitely can't wait to see your results, if it works out there might be a sudden shortage of drierite in the lab . . . :D

well . . . i do have some more issues too you may want to think about,

if your case is completely sealed as i understand it is gonna be, i think your ambient temp is gonna be wicked high.

everything that isn't directly water cooled is gonna be really warm, so it would probably have to be a really extensive water cooling system set up . ..

just stuff to think about . . .

[NuclearFusi0n]

Stop leeching bandwidth! :rolleyes: :mad:

-Fusi0n:eek:

[r0ckstarbob]

Originally posted by NuclearFusi0n
Stop leeching bandwidth! :rolleyes: :mad:

-Fusi0n:eek:

um - who are you talking to?

[r0ckstarbob]

Originally posted by hunter00
so you're gonna make a completely sealed water cooled/dessicator/peltier case?

well . . .i have to say it sounds like a pretty cool idea, i definitely can't wait to see your results, if it works out there might be a sudden shortage of drierite in the lab . . . :D


that would rock :)


well . . . i do have some more issues too you may want to think about,

if your case is completely sealed as i understand it is gonna be, i think your ambient temp is gonna be wicked high.

everything that isn't directly water cooled is gonna be really warm, so it would probably have to be a really extensive water cooling system set up . ..

just stuff to think about . . .

yup. already taken that into consideration. check my sig for a REAL bare bones concept explination behind what i'm doing (theres lots of concept drawings to illustrate what i'm talking about, including the desiccator). i'm working on the new site as we speak. if you like what you see in the sig you can check out the new site HERE (http://languagehammer.net/core/index.html). just be warned, the new site is pretty rough still.

[hunter00]

very cool, very cool.

i'm definitely interested in seeing how this fleshes out.

[Maximus Nickus]

IM getting watercooling and i want to waterproof ALL my components in case of leakag when ITS ON, will the spray mentioned allow me to do this? Is there any chance that the spray will damage compnents? Can i spray it all over? In pcis in the cpu socket gcard sound etc? Will it protect my system if it leaks and the powers on?

[Hachet]

Well rOckstarbob, you've done it again. Excellent work I must say. I Hope you don't mind that I linked to this thread over at Icrontic forums... I felt it was a "must read" for anyone trying to prevent condensation.

Thanks for all you've done.

-Eli :)

BTW, Mods, have you thought about making this a sticky?

[FrozenInHI]

Originally posted by nick_cw
IM getting watercooling and i want to waterproof ALL my components in case of leakag when ITS ON, will the spray mentioned allow me to do this? Is there any chance that the spray will damage compnents? Can i spray it all over? In pcis in the cpu socket gcard sound etc? Will it protect my system if it leaks and the powers on?

You most certainly can NOT do that. If you spray it in your pci slots you will definitely have problems, not to mention the cpu socket too. You will need to smear vaseline on all the components to protect them, it works better that way, but....
you will run into high temps on the board's transistors and other high temp components. If you install your watercooling properly and test it for a few hours before you start the computer up, you will be able to tell if there are any links, then fix those and do not turn the system on until it's 100 percent sealed. Once you've got that, you should have no trouble at all with leaks. Make sure you're not using some 1800 gallon per min firepump either, otherwise you're likely to have a seal bust.