Pumpless Watercooling?

[arcsylver]

Before anyone slams the new guy let me explain what got me thinking of this.


My parents live in rural North Carolina and have a well for thier drinking water.

It have a resivior tank that has an internal bladder for pressurising the pipes to allow for enough water pressure at the faucets.

In this setup, the well pump supplies water to the internal bladder as it is used in the home.

Now for a watercooling setup in a closed /sealed loop could something similar be rigged up using a bladder lined resivior to supply the flow of water through a system?

Or would it all just stabilise after a time?

Perhaps as a solution for inadequate flowrates or too long tubing runs?


Just something to tweak a few braincells while I am bored.

[Zhentar]

the problem is that won't work with a closed loop. You'd have to have water coming out somewhere, and therefore you'd have to have a pump pushing water into the pressure tank too.

[Phant0m51]

The problem you have is the pressure you would need. A well has alot of water, therefore alot of pressure. Also, the well doesn't re-send it's water back to itself in a loop, like a WC setup does. Another thing is that water can only go as high as the pressure is (take a garden hose, fill it with water and keep both ends even, and put a 1 liter bottle of water on top of one end. When you lift the other end of the hose up to the water level in the 1 liter bottle, the pressure evens out.

So what this means is that your 'inlet' and 'outlet' are both going to be submerged in water, meaning that no water is going to move without pressure.

I could be wrong, but this is how I understand it. Anybody else have any ideas?

[arcsylver]

Hmmm, then perhaps it could be used to boost a system by increasing the flow or pressure through it by increasing the system pressure throughout?


I wonder if this would interfere with a pump actually pumping the water due to increased backpressure on the pump or if it would actually stabilize and be able to pump the water only at higher overal pressure?


Maybe a better term would be pressurised water cooling not pumpless.

[Phant0m51]

You could do this...Have a 5 gallon bucket sitting on top of your case, and run 1/2" tubing down and use that for your WCing setup, then have another 5 gallon bucket on the ground collecting the water is it comes out of the other 5 gallon on top of the case. Now just use a pump to pump the water back to the 5 gallon container, and you now have pressurised water in your case...

Or you could just hook your 1/2" line directly into your house's water lines, and connect the other end to the sewer line, so you could have cold water, slowly running through your case forever.

That 2nd idea doesn't sound like that bad of an idea, as long as you kept the valve mostly closed so you don't have a $2,000 water bill at the end of the month

[bigben2k]

The bladder that you mentionned, is actually a damper, to prevent the variations of the pump from being felt at the faucet. Without it, the water would pulse out. It also acts to steady the pressure. It doesn't actually pump any water.

Now if you want to use a bladder to pump water, then what you're talking about is essentially a bilge pump, and most of us will tell you that it's not a good idea, because they break fast: they're not meant to run for a long time.

If you're looking to increase the pressure, you'd have to keep in mind that it's a closed loop, so the pressure has to be lost somewhere, making it an open loop. The pump isn't going to like it, I can tell you that right now!

Keep at it, there might still be something in the idea.

[starbuck3733t]

Originally posted by Phant0m51
You could do this...Have a 5 gallon bucket sitting on top of your case, and run 1/2" tubing down and use that for your WCing setup, then have another 5 gallon bucket on the ground collecting the water is it comes out of the other 5 gallon on top of the case. Now just use a pump to pump the water back to the 5 gallon container, and you now have pressurised water in your case...

Or you could just hook your 1/2" line directly into your house's water lines, and connect the other end to the sewer line, so you could have cold water, slowly running through your case forever.

That 2nd idea doesn't sound like that bad of an idea, as long as you kept the valve mostly closed so you don't have a $2,000 water bill at the end of the month

2nd idea is one I've tossed around before, as my landlord pays the waterbill for the complex, and its a flat fee on my rent. It'd rule until he noticed the extra change

My idea involved a water-to-water intercooler. A copper pipe that has constantly flowing (low-flow) ice-cold city water through it would it in a my reservior serving to cool the water. I might try it for a few days this winter just to see what would happen... cept I'd have to condensation proof my setup as the water from my mains is usually 50 to 60 degrees.

[sevisehda]

I tryed the 'second' idea for about a month at college. It worked well but after about a month I got a letter from someone in the housing department about 'some hoses running to and from the bathroom'. So I had to stop. The only problem was there was a good amount of buildup on the tubing from the meinerals in the water.

[Alchemy]

Originally posted by Phant0m51
The problem you have is the pressure you would need. A well has alot of water, therefore alot of pressure.

The amount of water is irrelevant. The pressure depends on the pump and the water level in the well, not the amount of water underground.

I'm pretty sure Ben is right; the bladder acts as a dampener, not a pump. Remove the pump and the bladder would push out the water inside to relieve pressure, and nothing more.

Alchemy

[arcsylver]

Well that is how I thought it worked myself. Guess this was a bust.

Still I can't help but to wonder if when a pump pulls water out of the bladder if it would n't be at a higher pressure than normally capable of the pump.

That and would the water return to the bladder via the loop or would the pressure create too much back pressure for it to be able to pump the water around the loop back to the bladder?

[starbuck3733t]

Originally posted by sevisehda
I tryed the 'second' idea for about a month at college. It worked well but after about a month I got a letter from someone in the housing department about 'some hoses running to and from the bathroom'. So I had to stop. The only problem was there was a good amount of buildup on the tubing from the meinerals in the water.

So you had no heat exchanger (coolth exchanger?) and ran the city water directy in your loop? Ouch. ick.

[sevisehda]

If you pump water from a high pressure bladder into a high pressure bladder the flow would be the same as if you did it from/to a res. The only difference would be there would be more pressure inside the system and a greater chance for a leak.

[bigben2k]

Originally posted by sevisehda
If you pump water from a high pressure bladder into a high pressure bladder the flow would be the same as if you did it from/to a res. The only difference would be there would be more pressure inside the system and a greater chance for a leak.

Exactly.

Your overall pressure would be higher, but it has no effect on pumping, other than the flow restriction.

[CoolROD]

The water lines of the house are a hydraulic system. Without the tank the instant you cracked a water valve the pressure of the system would drop to zero. There is a pressure switch that starts the pump. The bladder tank is an accumulator that puts an air cushion in the line which reduces the number of pump start / stops. The purpose of the bladder is to keep the air in the tank from diffusing into the water and coming out of the faucet.

edit: and as BB2k said: the bladder tank reduces pressure fluctuations at the faucet. It smoothes out the pressure spikes from the pump starting and stopping. The system will not work without the tank -the pump would start and stop repeatedly as the pressure instantly goes high and low!

[myv65]

Originally posted by bigben2k
Exactly.

Your overall pressure would be higher, but it has no effect on pumping, other than the flow restriction.

Close, but not quite Ben. Depending on one's setup it may have no effect or it may have a very substantial effect. True cavitation does not occur often, but when it does raising the overall system pressure is one sure-fire way to eliminate it. Generally speaking, most people that think they have cavitation merely have air in the lines. Generally speaking those that truly have cavitation merely need to clean up their suction line to the pump by eliminating nearby elbows or increasing the line size. Nonetheless, raising overall system pressure also eliminates the chance for the pressure to drop to the flash point.

[bigben2k]

I hear you Dave, but while on this, maybe you could clarify this whole cavitation effect for me. I understand how it works, but I'm having a difficult time putting numbers on it, and figuring out if a typical cooling pump, or slightly larger, is actually capable of creating that effect.

Do you have any wise words on this?

[myv65]

Wise words? Not sure, but I'll try to summarize the gist of it.

Pressure and velocity are interrelated. Water has a vapor pressure that's mainly a function of temperature. If the ambient static pressure drops below this value, some of the water will spontaneously flash to vapor. In doing so, it increases greatly in volume and momentarily raises the ambient static pressure which prevents all the stuff from flashing at once.

In a hydraulic system you've got a static pressure based on gravity, ie the pressure will be higher at the bottom of the case than at the top. You also have the ambient atmospheric pressure, which depends mainly on temperature and altitude. Once you start things moving the situation gets interesting.

The lowest static pressure in the system exists within the pump housing. The water gets pressure from the impeller literally by radial acceleration. Higher flow rates mean higher velocities within the pump. Higher temperatures mean lower vapor pressure. High restrictions ahead of the pump rob pressure that would help keep the fluid liquid.

One universal solution to cavitation is to throttle back on the pump discharge. This obviously reduces system flow and slows down the average velocity in the system. Another solution is a "water tower", basically a column of water tall enough to raise the static pressure in the system such that cavitation does not occur at the desired flow rate/temperature. Yet another is a pressure reducing valve that also artificially raises the overall system pressure. In this last one, a valve sits between a high pressure source and the flow loop and maintains a minimum static pressure.

In my old job, we used the "water tower" approach with hot oil systems to handle the expansion associated with heating ~5000 gallons of oil from room temperature to 550°F. The expansion tank pulled double duty taking up excess fluid and raising overall pressure. We also used the reducing valve approach with pressurized water systems heated to a maximum of ~375°F. These hot water systems would pump ~300 gpm of water that was maintained in a liquid state at these elevated temperatures.

Anyway, getting back to your question, industrial pumps typically state a value known as NPSHr, or Net Positive Suction Head required. This value varies according to flow (roughly rising with flow squared, imagine that). NPSHr represents the amount of positive head that must exist at the pump suction to avoid cavitation. NPSHa is the actual value available and equals ambient atmospheric pressure - V^2/2g +/- the static head applied by virtue of the pump elevation vs water surface elevation. There's also the matter of pump intake resistance, which some makers factor into their respective NPSHr values and others require you to factor into the NPSHa value.

I have no hard evidence to back up the next statement, but consider it "gut instinct". With joe-typical pond pump, cavitation is not likely provided you have a straight run of at least five diameters (preferably a little more) heading into the suction AND you do not reduce tubing diameter below the pump's intake diameter AND actual flow is less than ~2/3 of zero head flow.

[bigben2k]

Originally posted by myv65
...

I have no hard evidence to back up the next statement, but consider it "gut instinct". With joe-typical pond pump, cavitation is not likely provided you have a straight run of at least five diameters (preferably a little more) heading into the suction AND you do not reduce tubing diameter below the pump's intake diameter AND actual flow is less than ~2/3 of zero head flow.

I learn something, every time you post, thank you!

That's a range that may or may not fall within what an average Joe might have.

-The "2/3" of full flow is going to be a yes, in 99% of cases.

-I believe that most pumps are bottom mounted, or at least all the ones I've seen have the 5d run above them. There might be a few exceptions.

-I surely hope that everyone has learned not to reduce the pump's intake, following pHaestus' great article.


On the topic of "d's", I recently picked up an article in "Flow Control" mag that stated that a 20d run would be preferable, above and below a flow meter, without any diameter changes. Any thoughts on that?

Here's the full article.

As you might remember, I picked up a rotameter recently, and I'm looking into the feasability of incorporating it into my case, which is ~ 26 inches tall. The meter has 1/2" NPT connections. Here are the details of the meter.

[myv65]

The only flowmeters I use are magnet-based. To the best of my knowledge, all recommend a 10d in and 5d out minimum straight run.

[Gooserider]

My experience with those bladder tanks on a well pump system is that they act both as pressure stabilizers AND as water storage units - I believe they typically hold 10-20 Gallons of water below the bladder, and about the same amount of air above it. (Think of a 30-50 gallon air compressor tank, w/ a bladder in the center)

The pump runs until the tank is mostly full, and the bladder is under its highest pressure. Then the pump shuts off. The bladder and tank supply the house plumbing, and the pump kicks on and refills the tank when it gets close to empty. There IS pressure variation at the faucet, but the bladder pressure change isn't that great compared to it's minimal pressure. In some setups there is a flow regulator on the feed line to the house to reduce the pressure variations caused by the bladder.

In addition to pressure regulation, this setup means that the pump doesn't need to cycle every time someone gets a drink. (remember that a well pump consumes the most energy and generates the most heat at startup. It is far more efficient to have the pump run for extended periods filling a storage tank and shut off than to have it constantly cycling on and off for short spurts)

My interpretation was that arcsylver wanted to replace the pump w/ a bladder tank, and have the water come out of the tank, circulate through the system, and return to the tank w/o further pump energy input. This wouldn't work because it's a variant on the perpetual motion machine.

What one COULD in theory do, though I don't see any significant advantage to it, is to have one of those tanks, a powerful pump, and a resevoir that had capacity like that of the tank. With some sort of regulator or valve, allow the coolant to flow from the pressure tank through the system and into the non pressured resevoir. Set the pump up on a switch so that when the pressure tank got low it would turn on and pump the coolant from the res back to the pressure tank, then shut off.

It might be a way to get rid of the pump noise most of the time (but when the pump kicked in it would be loud) It might also make sense if doing a big outdoor tank setup ala Bladerunner. I don't think it would be good for a standard system like most of us have.

Gooserider