Just a random thought about chillers
 

I see alot of messages about people wanting chillers for thier PC's.
Most are converting Dehumidifiers or building true Phase change setups.

My random thought is this....
Has anyone ever tried using a old water fountain? You know the kinds made 20 years ago that actually cooled the water. Seems to me it would be a good choice since thats what its job was in its real life to cool a small amount of water but not to the freezing point.
Personally, and If I had a few more dollars and a bit more time, I think it would be worth a shot.
You "Should" get below ambient temps, and no frost build up since it is not below freezing.

Again this is just a random thought I figured I would share with you guys. I have no Idea if it would really work but it seems to me It should.

Aardil

my 5 cents
 

Hey.

Just to clear a few points.
The point on using a chiller is to have sub ambient temps, making a more "active" radiator. (pardon the expression)
You exchange the radiador connected to the loop by the chiller.
You will not have a standard radiator in the cooling loop because it will add heat to the water , due to the fact that the water is below ambient temps. And the radiator works both ways.

Said that, the chiller must be able to remove heat from the water , and maintain it's temperature (in a given interval of idle/load), working as an active element.

Assuming a standard cooling loop composed of CPU (overclocked), chipset and GPU (both with overclock) , you can get as high as 200W dumped in the water (or more).

For that to happen, it must be powerful.
A waterfountain is nowhere near as powerful as you might think. Most are rated for under 50W (some alot less). They work by cooling a tank of water, which is not used continuously, and it's maintained cool by switching on and off the engine when the water temperature reaches the edges of a given temperature interval (set) . So it has time between uses and temperature variations, to work down the temps.

You'll need the same engine as a prometeia/vapo (rated for 200W, 1/4Hp compressor or similar) . This way you'll maintain sub ambient temps, cool all three (or more) blocks and not worrying about temperatures . A good chiller controls the temperature by the same method, switching on and off to maintain the temp in a interval.

Things to worry about is in idle mode , the temps dropping below freezing (and create condensation), and in load mode, the engine not being able to keep up and/or not being able to operate in continuous duty (most aren't).

And no, a fridge heat engine doesnt do the trick, most fridge engines aren't rated for dealing with that type of heat. However , a freezer have alot more powerful engines, and I know a couple of guys over here who sucessfuly converted one into chillers. Althou they are massive (bigger than a cube case). The deposit consumes most of the space for several reasons. Some time delta between switches, and a large reservoir gives them that time , while maintaining temperatures between the given interval (set) , and a large reservoir takes alot of cooled water making the temperature changes "less hard".

Temperatures will be defined by the size of the reservoir, the power of the engine, and the time it is on and off. Of course, if you have a small res, then it has to work more and harder, and vice versa, less and not as hard, and you can even get a "not-so-powerful-engine".

Or, you can go Peltier (but i'm assuming you know the pros and cons of that, if not, there's a thread about swiftech's chiller in pHaestus worklog to give you an idea).

Nice reply. I'd have to agree with everything he says. Most water fountains are only meant to chill a certain amount of water and then stop. That is why they only turn on the compressors some of the time you go get a drink.

Ok,
As I said it was just a thought. I knew there had to be a reason to not use one just didnt see it.

Thanks for the replys, and back to drinkin' er thinkin'

Aardil

wait is condensation caused by the chip getting below freezing or below room temperature... i thought if it was cooler than room temp, water droplets would be formed also.....???

Yes, condensation is below ambient temp, but not a few degrees, I heard you must go 10° below ambient to have codensation.

But I'm using aircooling, so I'm not the expert here :-)

FYI, condensation is a function of temperature and the amount of water vapour in the air.

At any given temperature, air can contain a certain amount of water vapour. As the temperature goes up, the amount of vapour that can be held goes up. As the temperature goes down, the amount of vapour that can be held goes down.

The DIFFERENCE between the amount of water vapour that the air actually contains and the amount that it theoretically could contain at that temperature is what makes the number for 'humidity' that you hear them talk about in weather forcasting.

The temperature at which a volume of air can no longer contain the amount of water vapour presently in it is called the DEW POINT. The higher the humidity, the higher the dew point.

If the amount of water vapour is greater than the air can hold, you get condensation or fog. (AKA Dew)

If an object is cooler than ambient air temp, then it causes the air next to it to cool (basic thermodynamics), if the temperature of that cooled air is lower than the dew point, you get condensation, if it isn't you don't.

Thus a sub-ambient cooling system in the Sahara (where there is low humidity, and a very low dew point) might not get any condensation. The exact same system placed in Southern Louisiana (where you have high humidity and a much higher dew point) would get extreme condensation.

It is important to understand this if planning to go subambient, because it will determine how much insulation you need on your cooling system and hardware.

Gooserider

What he said.


PS:

Not the best example (but we get the point).
At night , temperatures can (and usually do) drop down to freezing point in the Sahara Desert. It's not uncommon to see - 5 ºC or less (much less in some areas).

If you're from, as an example, Algeria, and live in the outskirts of the Sahara (which is not hard, as it is like, most of the country) , you can get in certain regions up to 60 or 70% relative humidity and a temperature intervals of -5 ºC to 50 something ºC .
"Nice weather we're having, eh?". :D

I'll just put in this chart, as it seems appropriate here. It could do with being updated to read higher I know, but it's been pretty much spot on so far with my findings. I'm in the vary variable climate UK and have seen 88% humidity which only allows about 2C under ambient before condensation. Had to bypass the underground tank for that couple of days........humidity is a very irritating thing :mad:

http://www.zerofanzone.co.uk/picture...ationchart.gif

Thank you one and all for your comments and helpful advise.
Since I do live in South Lousiana ( and the Humidity does reach into the upper 90% range Although thats only during the Monsoon season. You know, that short time of the year between January and November) The rest of the year the Humidity is a comfortable 85%. Too Many days around here I see the Humidty Higher than the temp.
So If I was to go to phase change or chiller, I would need to heavily insulate against condensation.

Think I will stay with W/C at least for now. PC's seem content at current temps.

Aardil

Thanks for the great chart Bladerunner. Very useful even for us farenheit fixated Americans ;)

I agree. I used to live down in Red Stick during my pre-computer days, and remember how hot and sticky it got, and how you would get condensation on anything that gave you the slightest excuse. (especially when I worked on the river...)

I would be very hesitant to go sub-ambient at all, but if I did, I would probably be inclined to go with a pelt mounted on the CPU, with a WC system to cool the pelt. That way you would only need to worry about insulating the CPU area, as everything else in the system would be above ambient.

Gooserider

The same goes for a vaphor phase system...(although that should chill colder)

The chart is a little lacking in top end results as it could do with reading to about 45C but I guess a higher ambient than 30C is more often than not accompanied by a humidity over 50% anyway.. so your pretty limited to chilling possibilities. It wouldn't take long to cross reference it to Fahrenheit if you really wanted.. still makes much more sense to talk of anything water and temperature related in Celsius. 0C= frozen water (Ice) 100C = boiled water (Steam)

I do have an idea / plan to "solve" this condesation problem in my system. Yes I could insulate every block and tube, but those that know my system would realise how near impossibly hard this would be to do successfully. It would also cover all my blocks with ugly lagging. :(

What I plan is to make a custom PC case to roughly the dimensions of a large JCB rad I have. The case will be near air tight sealed other than all the base area and a chimney or collection of chimneys in the top made from copper. Each chimney will contain a light inside, (probably 12v halogen car fog light bulbs or similar).

With the large rad shown below, (that I'm currently using as a true rad for the system), laid flat in the base of the case, chilled coolant from an underground tank at between 8C to 12C would pass through this rad after its loop through all the blocks on its way back to the tank. With the heat-up of the chimneys there should be a small but significant drafting effect like a real chimney in a house when you have a log fire. This in theory should draw air through the base mounted rad cooling it and probably dehumidifying it enough so that the insides of the pc are significantly cooler than the ambient and as such keep the blocks and tubes condensation free... will it work? how well could it work?...only one way to find out and I'm working on it :D

http://www.zerofanzone.co.uk/projects/pc/giantrad.jpg


Fahrenheit to Celsius Converter

PS:

http://forums.overclockers.com.au/sh...hlight=chimney . "Chimney Case".

That's an interesting topic and I hadn't seen that before, but have seen some others in the past. I know it's not a new idea as such and I wasn't claiming I'd thought of it, although I've been toying with the idea since I first water-cooled my PC. I was more thinking how well it may work in reducing the dew point in humid periods to enable the use of chilled coolant without having to continually worry about the humidity %. To get a chimney effect to work well I think you'd need a pretty large temp differential, hence my cool 10c rad in the base and heated copper chimney stacks that could be more than 50C to 60C if need be

Here's an early 2002 topic where I outline the basic idea in my reply... also shows how long I've been posting that chart in threads to help people understand humidity & condensation :p

http://www.ocforums.com/vb/vb/showth...threadid=81696

I could sit here and make paper equations & calculations until I were blue in the face, but it still wont really tell us how well it may work in actual practise with all the real life specific variable factors. The thing is my system MUST remain 100% fanless... along with the fact I want to be using Tec's soon :)

So, the relevant issue stands only on the effectiveness of the convection currents created by the forced deltaT .

You're assuming that the condensation will occur only on the rad and maintain a low humidity of air passing through the case (after the rad). You could put the rad before the loop to maximize the effect (assuming it works). At the expense of a few degrees on the blocks.

I think 50ºC is a bit much thou :) . Are you going to isolate the chimney with neoprene or similar ? And the chimney heating the room, and changing ambient temps to higher , will help condensation on the rad ?. Hmmm.

Something else came up , you'll need something to collect the precipitated water on the rad ... sounds complicated.

However, i'd like to see it assembled and working.

Or, you could just heat up the interior of the case. Less hassle. But it would probably affect the effectiveness of the cooling loop.

I think we may have our wires crossed? either that or I'm misunderstanding you ;)

My system will be cooled by an underground tank the rad will do nothing for the cooling except raise the coolant temp a small amount before it goes back to the tank. At the moment all the blocks in my system add under 1C to the coolant temp in one pass. This might increase to 2C or 3C with Tecs. So if I add this air "conditioner" rad, in the loop, (after its been through the blocks), it will get around 3C warmer coolant than enters the PC, so about 12C is my guess.
The plan is to build an improved underground cooler that will cope with the Tec's and not heat up at all, (probably a large run of copper pipes buried instead of a tank). Assuming I can build a truly effective ground cooler this time, (and It's going to be very large trust me :D ), then it will return the coolant to the PC completely re-cooled to between 8C to 10C )

The rad pictured before would be then working in reverse attempting to cool any air drawn through it by the chimney effect, as its going to be in the base of the PC drawing air from underneath any condensation that forms on it I wont drip on anything but the floor...... It's unlikely to be much, or for long periods where I live but if it turns out to be an issue I'll make a drip tray affair. The heated chimneys won't be any different than having a few light bulbs on, (nowhere near the heat that comes from a CRT monitor, and it would be at the top of the case. Could be outside vented if really need be.

Heating up the PC interior is the exact opposite of what I'm attempting to do as its the hot air combined with the high humidity % that is the issue causing condensation to form on the cool surfaces, (tubes and copper blocks). It's tricky to control the humidity without using fan based de-humidifiers, but if I can significantly lower the air temp inside the PC case the dew point will also fall significantly, (see my condensation chart).

If this is enough then the only place I'd likely get condensation is possibly the outside of the case in extremes, and on the air "conditioning" rad... (again its all untested theory until I build it), I wont need neoprene on the chimney however. Its warm humid air on a cold surface that leads to condensation not cold air on a hot surface.

It all sort of works in my head, but there are many variables....... its really just a case of will it work well enough to be an effective solution to the one or two months of humid & hot weather we get in the UK yearly?? ...... I better shut-up talking about and get on with actually doing it I guess :D

I think you misunderstood me on several points. Or not :) .

I got the general idea at the first post. But with the (limited, i admit) experience i had a few years back with a few AC systems , they do tend to generate alot of water, specialy in the summer. But hey, here , summer we get up to 40º something, and we call winter when temps drop below 14º. I cant complain much about the weather.

This setting (assuming it works properly, which I think you can make it work) , wont be as powerful , but I'm willing to bet it will generate enough, to cause some concern about the rugs :D. Specially in hot days , and , with that new subterranean radiator, you can get temps lower than 8 ºC. That's why my comment on the water issue.

The neoprene is for the outside of the copper chimney . You pointed 50 to 60 ºC . I wouldnt want to touch it :D. And also to keep the heat inside, and avoid exterior condensation. Just making good use of energy. And it's probably not the best idea to connect the chimney outside, the recirculation of air in the room through the system will keep the humidity levels down, which is the whole point.

Again, the system will only work if the convection currents formed are sufficient to move enough air through the system. I mean, the chimney must be able to drag enough air through the rad (and the case). Which is not that easy because the cold air is heavier.

It's all good , but , if you're building an air tight case, you could just try to form a small vaccum. No air, no condensation ;) (I know I know, it's probably a pain to achieve even a partial vaccum. Well ... er cube of lexan maybe, hmmm).

Yeah I see..... a lot of things are relative to our different planet locations I guess, it doesn't tend to get as hot here 35C being pretty much a rare summer high, with mid to high 20C to low 30C more of a norm, (although I've seen almost 40C inside a mobile home at one point), that was horrid btw :mad:. Although the coolant will be cool it's nowhere near a/c cooling temps so I doubt the wetness will be as much of an issue, (around 8C to 10C is the ground temp after around 8ft down), I don't see it going any lower than that.

I think you summed it all up well with this statement however...

"Again, the system will only work if the convection currents formed are sufficient to move enough air through the system. I mean, the chimney must be able to drag enough air through the rad (and the case). Which is not that easy because the cold air is heavier."

That is the real key and no amount of educated guessing or calculations will fully tell us exactly how well its going to work......other than doing it, may be a flop but I can take failure ;)

Another thing is real fires have good draft effect in chimneys, but they are also burning with much hotter temps and the fire is using, (consuming), oxygen from the surrounding air, causing a greater drafting effect.

Most of the time I was fighting the borders of condensation, not a major bath time so it will most likely help.

what made me mostly think you weren't with me was this statement mainly:-

"Or, you could just heat up the interior of the case. Less hassle. But it would probably affect the effectiveness of the cooling loop"

Wasn't understanding that as that would have lead to a swimming pool condensation situation ??

Oh and if I burnt myself on the hot chimney It would be the first and last time...... perhaps I should have "numbskull" embossed backwards on the chimney :D

lastly if it worked a condensation detector on the psu block linked to the hardware shutoff maybe another wise precaution, for the odd time the humidity was very very high?

The problem is that in worse case scenario you'd get condensation just about everywhere. Assuming the blocks , peltiers, are perfectly isolated , front and back. If any condensation appears inicialy, it would be on the tubing. And start to run on the motherboard , cards , the works.

So, at least for me, it wouldnt be particulary useful to have a condensation sensor in the PSU, where it is in a "box". Unless it has a benefit i cant see.
The problem is that , in the project , the floor is the radiator, so a detector cant be placed on the bottom of the case (where usually things drip the most).

So, unless the detector works near (or ON) a tube (the main intake should work) , it will be difficult to detect condensation, IMO. I think it would be easier to detect humidity levels. There's a few sensors around, cheap.

http://www.relative-humidity-sensor.com/ , example.

The ideal would be to read the temperature of the exit air from the radiator , the temperature of the exit air from the chimney , and with the reading of the humidity levels , control the chimney heating. A small uP could do that, similar to a circuitry i think i saw around the forums , i think it was Turbokeu's. Then there'd be the calibration problems , etc etc etc.

Nah, emboss a dragon or a tiger . If anything happens , at least it would look real cool :D.

Firstly apologies for hijacking this thread in general.. although its still topic related.....

I like the idea of a unit that could monitor humidity related to temp and alarm at set points if the water temp was below the perimeters. That however is getting well beyond my capabilities and understanding of electronics.

My reasons for putting a condensation detector in the PSU is that tends to be first to show signs of condensation. I think its because the air is warmest inside the psu, even though its an open design it stagnates there more than elsewhere due to the fanless nature, and general crammed in style of atx psu component design.

My understanding of a "humidity sensor" is to take a volume of air and dehumidify it collecting the water... a simple calculation and bingo.. %humidity. I don't think they make electronic humidity testers... though if they did I would surely like to see one (and more importantly how they would work.)

Though back to the original point of the thread... having an "ineffective" chiller in the loop might not be the best solution ON ITS OWN. but with a normal array of radiator(s) it will help keep the temperature hovering around the ambient mark. Though it may not be the most cost effective solution.

~ Boli

I have a couple of digital "humidity sensors" that have been pretty much spot on (using the chart I posted as a reference). As far as I know they don't "collect any water" but I don't know how they work.....

Here's the one I'm using reporting 48% humidity, shown in some random bench testing.

http://www.zerofanzone.co.uk/projects/pc/testing0.jpg

*pesters Bladerunner to take it apart* :D

~ Boli

There are several ways to read relative humidity.

Some sensors uses a capacitance method (condenser). The dielectric is the air, and the capacitance value changes with the humidity of the air. Or some similar design (it's not written in stone).

Another method is a sensor constructed with an inorganic and higroscopic (correct word?) substance. It basically reacts to water vapor. The expansion/contraction of the substance is measured by a moving nucleous inside a bobinated coil (with no friction) which the alteration of fase of the signal through the coil when the nucleous moves , gives the relative humidity. Pretty simple actually.

Number three is even better. A small box with constant air movement , with ambient temperature reading. Inside , a laser is bounced on a mirror surface , which, on the spot the laser hits it, heats up and creates condensation. Then , the distortion of the light caused by the condensation is measured by a optical sensor . There is a relationship on the temperature of the ambient air and the distortion caused by the condensation , gives the relative humidity.

Take your pick :D.

Anyway, I was thinking (i do that sometimes). You (BladeRunner) could get a Matrix Orbital (reviewed by pH) and use it's three temperature probes (pretty good ones) . One for the cold air (rad) , one for the hot air (chimney) and one for ambient temps. It's also possible to control 3 exits with the MO (you just need one really), so you could connect them to a simple relay circuit , and the MO has a nice sofware that you can "program behaviors" , and control the heating of the chimney. Sounds nice and (somewhat) simple. Better than my uP ghetto approach.

Thanks for the timely explanation TerraMex was just about to disassemble my humidity sensor...... had lump hammer & chisel in my hands :p

Anyway I got a couple of MO's but don't remember them coming with temp sensors. Will try to work something in along those lines, I suppose the MO could be used to report humidity using a humidity sensor on one of the temp inputs (take some time to calibrate it fully though).

Bladerunner - I think you and TerraMex have the idea wrong, I suspect through a misunderstanding of exactly what your chart means.... I think that if you built the system the way you describe, it won't do very much, and might very well make your condensation problem WORSE!

Your chart describes the dew point, or what temperature condesation will occur at if the air is FULLY SATURATED with moisture. Or, in other words, given air at a particular temperature and relative humidity, how much it has to be chilled in order to have it reach full saturation and start condensing.

The term "Relative Humidity" describes the RATIO between the amount of water that a volume of air at the given temperature could theoretically hold, versus the amount that it is actually holding. (note that pressure also effects this, but not as much so, and can be ignored for our purposes) At least at the temperatures we are talking about (~10-60*C) cold air can hold less water vapour than hot air.

Thus if you take a given volume of air, and evaporate a specific amount of water into it, when you heat the air, it's relative humidty will DROP, and so will the dewpoint. If you cool the air, then it's relative humidity goes UP, along with it's dewpoint.

The most visible common example of this is a summertime thunderstorm - Air close to the ground is heated and becomes saturated with water, then rises as a 'thermal'. As it goes up, it cools, becomes super-saturated, and condenses, making a cloud. When the weight of the water in the cloud becomes more than the rising air can support, it falls as rain. The falling rain cools the thermal, shutting off it's flow, and the now unsupported water falls abruptly. (This is a gross over simplification of the process)

As another example, if you go into a dry heat type sauna (NOT a steam bath) with a cold can of Buttwiper (I wouldn't insult Guinness by cooling it that much!) you would not get much (if any) condensation on it, because the relative humidity is so low that the dewpoint is lower than the temperature of the beer can. (Once the experiment is completed you can cook some hot dogs with the lousy beer, and get a nice dark beer out of the properly non-functioning Lucas Electric beer cooler :D )

Now as it applies to your design proposal, what you seem to be trying to do as I understand it, is cool the air entering the case, without changing the amount of water vapour that it's carrying other than possibly having some condensation on the incoming air radiator (which really won't help much).

To pull some numbers off your chart, picking a fairly bad case scenario....

Assumptions - per your previous posts, and my guestimates....
1. Your rad is at 13*C and your blocks are at 12*C
2. Air going through the rad will be cooled 5*C from ambient.

Assume 30*C and 75% humidity (which the person from Southern Louisiana would consider wonderfully comfy...) The dewpoint is 25.1*C. Condensation would occur on your blocks at 12*C.

Run the air through the rad. it is now at 25*, and the dewpoint has now dropped to 20.3*, and condensation is still occurring on the blocks. Even with a really big drop to 20*, your dew point is at 15.4* and condensation is still happening. :cry:

But wait! you say, wouldn't there also be condensation on the rad? Yes, there would be, but the amount of water removed would only be enough to leave the air exiting the rad near saturation for the lower temperature. The dewpoint of air at 25*C and 95% humidity is 24.1*C, or ONE degree less than the result with no rad at all.... :shrug: :cry:

If I were trying to design something to deal with this, I would probably try to make a box with little or no airflow and try to dehumidify that air some how. Possibly you might also be able to do the same rad and chimney setup, only locating the CPU box ABOVE the chimney.

This way you make a sort of 'fanless dehumidifier' where the air would enter the rad, be cooled so as to get lots of condensation, and then reheated to ambient (or a bit higher) which effectively lowers it's relative humidity. If you can drop the humidity to 30% at 30*C by doing this, your dewpoint in the PC case becomes 10.5*C, and you'll have no condensation. :cool: (I would note though, that those numbers would leave less margin than I would feel really comfortable with...)

No , i think you're wrong on several issues.
But i (and Blade does too) do appreciate the input. Let the hostility begin :D .

If it's fully saturated then it has a relative humidity of 100%. That's the definition of fully saturated.
That's not what the chart means. It means that at a given relative percentagem of humidity, to diferent temperatures, when the condensation will occur. It doesnt have to be fully saturated.
I can have very low relative humidity and at low temperatures it will create condensation. More on that later.

That's one of the main points. But it's harder for a cold air to release the one it has, if it's a low value of relative humidity, to another cold point (relative cold). The delta T needs a minimum value.

If you have a given volume of air, with a given percentage of water vapor, if you evaporate an extra quantity of water vapor, then, for the same temperature, it will have a higher dew point. See the chart.

When you increase the temperature , for the same humidity level the dew point will increase. This goes against what we want.
We want to remove some humidity and maintain a cool temperature. If a good deal will precipitate on the rad, then , the air (still relatively cool) that goes through the case will have less humidity and lower dew point. See the chart. And that's what we want.

But I'm assuming the rad is very efficient on doing that.

Gross, but i get the point.

But this is better :

When moist air cools, a cloud can form. This much is true.
Sure, a cloud may form as the temperature drops, but not because some mystical holding capacity of the air has decreased.

The air (mainly nitrogen and oxygen) no more has a holding capacity for water vapor, than, say, water vapor has for nitrogen. The atmosphere is a mixture of gases. While saturation (which involves bonds between different molecules) is a real phenomenon in liquids it does not describe the interaction of atmospheric constituents.

Water molecules are constantly coursing back and forth between phases (another word for the three states: vapor, liquid, and solid). If more molecules are leaving a liquid surface than arriving, there is a net evaporation; if more arrive than leave, a net condensation. It is these relative flows of molecules which determine whether a cloud forms or evaporates, not some imaginary holding capacity that nitrogen or oxygen have for water vapor.

The rate at which vapor molecules arrive at a surface of liquid (cloud drop) or solid (ice crystal) depends upon the vapor pressure.

The rate at which vapor molecules leave the surface depends upon the characteristics of the surface. The number escaping varies with:

the phases involved --- molecules can escape from liquid more readily than from the solid (ice);
the shape of the boundary --- molecules escape more readily from highly curved (small) drops or ice crystals (convex);
the purity of the boundary --- foreign substances dissolved in the liquid or ice diminish the number of water molecules which can escape;
the temperature of the boundary --- at higher temperatures the molecules have more energy and can more readily escape.

The temperature of a cloud droplet or ice crystal will be (nearly) the same as that of the air, so people imagine that somehow the air was to blame. But, if the (other gases of the) air were removed, leaving everything else the same, condensation and evaporation would proceed as before (the air was irrelevant to the behavior).

What appears to be cloud-free air (virtually) always contains sub microscopic drops, but as evaporation exceeds condensation, the drops do not survive long after an initial chance clumping of molecules. As air is cooled, the evaporation rate decreases more rapidly than does the condensation rate with the result that there comes a temperature (the dew point temperature) where the evaporation is less than the condensation and a droplet can grow into a cloud drop.

Evaporation increases with temperature, not because the holding capacity of the air changes, but because the more energetic molecules can evaporate more readily (with, of course, the caveat that evaporation is also influenced by things other than temperature, as described above).

And when the temperature drops below the dew-point temperature, there is a net condensation and a cloud forms. This can even happen at ground level, fog, as an example.

Cool it and create condensation. Not one or two. Both. Both will change the dew point considerably (Bladerunner hopes). Assuming it can change by a large percentage, both the relative humidity and the temperature. But it's a big IF.


Wrong assumptions, assume BladeRunners numbers.
Assume a 20º C ambient , average.
8ºC or less at radiator (or even average 10ºC). No forced air.
The issue is how much the temperature drops as it passes the rad.
And how much a drop of relative humidity.

On the contrary, for me anyway, as stated. You misread. I stated condensation will occur, probably. Again, depends on the rad.

But honestly, if the rad reduced enough the temperature of the rising air, it will be suficient to avoid any type of condensation without tampering with the relative humidity. Given the chart that is.

And you forget that Bladerunner stated that the cooling loop will be cooling pelts, so the water wont be that cold, and the interior wont need very low temps.

Damn, i gotta stop making big posts.
Forgot to make a comment.

---
Dehumidifiers solve the high relative humidity (RH) problem by simply pulling the moisture out of the air.

A small fan pulls the rooms humid air into the dehumidifier and the air is passed through a cooling coil and is chilled to its dew point on a cold surface where condensation occurs, similar to a refrigerator. The condensation then is then released into a bucket or is released through a hose. The moisture reduced air is then heated back to room temperature and blown out into the room.
----

But. Because of the non fan setup pictured by BladeRunner, the air current formed (theory) wont maintain a very low temperature after passing through the radiator. In fact it will be cooler than ambient, but the cooling system is already cooler than ambient as it is , in particulary the blocks and tubes (due to the nature of BR setup). So no harm done there , assuming not that large detal T. It will (theory) have less RH , which is the whole point. Cooler than ambient, less RH , very low risk of condensation . In theory of course.

It's all very nice in theory but only by testing it (IMO) can be see what are the real flaws. Alot of tweaking can be done , shape , size, temperature of the chimney, etc etc etc.

Sorry about the jumps. I've got the organize my posts better. :) .

Hmmm just had a thought:

I would hate to sleep in a room with an active Dehumidifier, or even work in one with it. Anyone who wanted to install them need to get their head seen to, after ten minutes in there it'll be like a bad hangover.

Any solution should not include a dehumidifier as the human body will complain unless you can find a way to put the moisture back into the air or vent the dry air COMPLETLY out of the room. It won't be nice.

Sometimes people forget when they turn off the heating in the computer room - or vent all the moisture out of the room to "improve their computer" that people have to work in there. (NB: I know turning off your heating is the worst thing you can do in a computer room)

~ Boli