Finally finished my setup...

[jtroutma]

Finally got my system up an running once again after my latest watercooling upgrade.

Added a Cool-Computers chipset waterblock w/ 36W TEC. Final results are CPU @ 1.500Ghz (150 FSB x 10); KT7A voltage mod and absolutely no condensation. System will boot @ 1520 (152x10) but Win2000 Blue screens at random.....(Win98 works fine; grrrr)

Idles at about 40C and full at 42-44C depending on how hot it gets (California in summer not very good for OCing

Here are a few pics (Sorry that they are soo large............)



Thanks for all the help guys.
Any ideas how I can improve it (without spending more money?

[jtroutma]

Grrr....stupid space in URL code and it doesn't work

Here is the final pic....


Enjoy

[resago]

most people pump to the radiator first, then to the blocks.

you might give it a try and see if yuor temps change.

also, you could cut a hole in the top of your case and duct outside air through the radiator. maybe good for a couples degrees.

[LiquidCool]

Just curious, but, is that a Rio1100/1400 modded to run inline? I'd be interested in how you did it if so. Is it as simple as threading in the two fittings? thats what I had planned on doing but haven't gotten around to it. That ring seals with an O-Ring so I figure thats all you have to do...

[jtroutma]

Yes that is a Rio 1100 you see there

I actually had to do some moding to the ends in order to get it to work perfectly. Originaly I just "power" threded the ends in and used teflon tape to seal it. That worked for about a week, then I shut down my whole rig to go on a month vacation and when I came back the water had all drained out.

So I completely got rid of the tape and get some 50 grit wet/dry sand paper and sanded out the inside until the threds barely at into the plastic. Threded it in until I liked it, took the fittings out again and used epoxy to permently bond the threds/fittings to the pump. Since then I have yet to loose a drop out of the pump (I grew very fond of epoxy during the making of this project

I'm thinking I should have gotten the Rio 1400 because of the slightly higher GPH but at the time I thought that too much would somehow cause leaks or overpower the system. BUT the 1100 does the job very well and is almost completely silent (used velcro to attach it to the bottom of the case; beautiful for easy detaching, holding the pump in place while system on its side, and absorbs vibrations very well)

Any more questions?

[jtroutma]

Resago

The reason I put the rad after the pump was because that was the best place to put the rad and also the difference in temp should only be MAYBE a degree IMHO but I could be wrong. Also forcing water to the top of my case, back down to the waterblocks and then back up to my res. would probably cut my GPH quite a bit because the water would have to rise twice in that setup instead of only once (I know that sounds stupid but the principle is there)

About ducting fresh air....
I have the front 5 1/4" bays open and I can feel the cool air being sucked into the case from the front (cools my HDDs at the same time as well I also tried it with the cover off and the temps went down 1 degree and noise difference was not noticeable; so the cover stayed off.

Thanks for the idea though..... maybe for my next project.

[LiquidCool]

Yeah, I have 4 of the 1400's sitting here and was planning on modding one or two of them to run inline. I'd just get the tap for whatever size thread fits in there.

Yeah I just checked it and it would be perfect for a 1/2"NPT thread. Maybe they make a model with a threaded inlet? The outlet seem to BARELY just take a 3/8"NPT thread.

Guess I'll go get a 1/2"NPT tap and try it out.

Does it put out much heat? When I had all four running in a 5 gallon bucket they heated up the water pretty hot in a few hours. How hot does the casing of it get?

[jtroutma]

Mike

The Rio 1100 that I have just gets warm to the touch. I really doubt that this one in my system is adding much more than say 5-10W to my systems overal heat. As you can see from my pics the CPU is putting out about 100W and the TEC chipset waterblock is puting out around 50W or so.

I also have that pump mounted in the front with a case fan blowing air on it (not intensionaly butit worked out well)

4 of the RIOs (or any pump for that matter) sitting in a bucket of standing water for hours would heat upo the water no matter what as you probably already know. Keep in mind that your 1400 Rios put out the equivalent of 260-275GPH @ 3 feet head so they do quite a lot for just that little amount of heat.

Correction: the 1400 puts out 290GPH @ 3' head and 420GPH @ 0' head. Also yours uses 27W of power (mines 22W) however I doubt that 1/2 of that is going straight to heat. I just checked the spec. chart that I got with my RIO on all this info.

In short, IMHO that they run relatively cool for what you get out of them.

Hope that this answers your questions.
Anything else?
Good luck and let me know how it works out for you.

BTW: yes the outake on the RIO 1100/1400 will hold the 3/8"NPT and the intake will have to be taped for (i think) 1/2"NPT. I would have taped mine but didnt have the tools available

[LiquidCool]

I had a small transmission cooler hooked up circulating the water, I just didnt think they would get THAT hot that quick!

Course my computer room is 3C++ warmer than the rest of the house, even with the door open!

[jtroutma]

Just let me know how everything works out for you..

Thanks

[redleader]

All energy a pump uses goes into heat and most of that enters the water. That is unless your pump glows

[resago]

then whats pumping the water? the watercooling fairy?

SOME of the energy is used to turn a shaft or create a magnetic field.

[LiquidCool]

Only a small portion of the energy is turned into heat. This will mostly depend on the quality of the motor.

You can't say that your 120VAC 25Watt pump puts 25 watts of heat into the water. It's most likely maybe 10% if that, which is only 2.5 watts

[redleader]

then whats pumping the water? the watercooling fairy?
SOME of the energy is used to turn a shaft or create a magnetic field.

You're telling me their's no resistance then? Because if the coolant encounters any resistance then there is friction. And friction leads to heat.

Tell me what happens to flow rate when you unplugg the pump? Unless it continues to flow at full speed all of the mechanical and magnetic energy will be converted to heat (save a tiny amount lost as radiation, etc).

Seriously think about it. You have a constant flow of current going into the pump and no kinetic energy leaving the pump (in a closed loop anyway). Unless the energy builds up endlessly it has to be converted into heat and move into the coolant.

Therefore 99% of the pumps energy becomes heat and most of that enters the coolant.

[redleader]

You can't say that your 120VAC 25Watt pump puts 25 watts of heat into the water. It's most likely maybe 10% if that, which is only 2.5 watts

I can say that those 22.5w of kinetic energy transfered to the water cannot leave the system in any form other then heat. So while you are correct that the pump itself maybe 90% effcient, all of the kinetic energy it produces will be converted to heat. Otherwise energy in the system would build up endlessly (current going in is constant), and of course this doesn't happen.

[resago]

If the water is moving, its not being converted to heat.

and the very little friction between the water and the lines is greatly overcompensated by the smallest of radiators.

point being: if you have constant flow given a constant energy source, the heat to kinetic energy ratio would also be constant.

and the radiator takes care of any friction generated heat in the lines.

I do agree however that without a radiator, the water would heat to a certain critical temp and would equalize at whatever temp radiation could release through the lines into the air.

I also agree some heat from the pump is radiated into the water from its contact in the pump.

ex. on my system with my computer off. with my pump running my water temp equalizes at about 33C with no fans blowing on my raaidator. since the water is still flowing, you cannot say that all the energy a pump consumes is turned to heat.

[redleader]

point being: if you have constant flow given a constant energy source, the heat to kinetic energy ratio would also be constant.

Yes but remember your coolant is not constantly accelerating. Once it reaches a certain speed it no longer can go any faster due to restrictions. However your pump is still inputing the same amount of kinetic energy, therefore the same amount of energy per unir of time must be converted to heat and as you said leave through the radiator or be converted out through the pump itself into air.

If the water is moving, its not being converted to heat.

Really? Like I said, unplug your pump and see if the coolant keeps flowing at full speed. It doesn't? Then all that kinetic energy was converted to thermal energy.

This is simple high school physics, all energy inputed into a closed system eventually becomes heat. A closed loop watercooled computer is no different.

[ 09-13-2001: Message edited by: redleader ]

[LiquidCool]

Dude, WTF are you talking about? The friction of the water and the tubing/blocks/whatever creates, effectively, no heat. If we were creating high pressures, 40PSI+ then you would get some heat from the pressurization of the water, but we aren't.

The electrical energy of the pump is NOT completely converted into heat. If it was then the motor would not spin. The only reason heat is generated by the pump is from the resistance of the materials of the motor and also from mechanical moving parts in the motor(bearings/brushes).

I dont know what the hell your talking about unpluging the pump. Of course if you unplug the pump the water stops moving. You just removed it's source of movement. The reason it stops so soon is it has almost no inertia and because the pumps impellers have effectively damned the water off...

[resago]

the thing about the water not accelerating is because the pump is not accelerating.
even in a frictionless system, the pump would still only produce a constant flow because of how motors work.

then then only heat added to the system would be from conact with the pump, which is really where all the heat comes from anyway.
coils produce heat as well as magnetic fields.

take for example a blower motor for an air conditioner, you really think that motor is making 1200watts of heat?

[ 09-13-2001: Message edited by: resago ]

[ 09-13-2001: Message edited by: resago ]

[redleader]

Please think about this.

Energy is going into the loop in the form of kinetic energy from the pump. Do you agree?

The same amount of energy that is inputed must eventually leave the system. Do you agree?

The only way this energy can effectively leave the system is in the form of heat. Do you agree?

Really I don't see how there can be any doubt here. Of course energy is going to be conserved

[resago]

The kinetic energy in the system in fairly constant, the pump spins at a constant rpm and the fluid flows at a constant rate.
the pump isn't try to spin any faster than it already is. thus, once a fluid is in motion and you are not accelerating it, no extra energy is being added to the system.

seriously 99% of the heat from the pump is from contact, but that heat is only a fraction of energy the pump consumed.

[LiquidCool]

"Energy is going into the loop in the form of kinetic energy from the pump. Do you agree?"

Yes but it is an infintisimal(SP?) amount. Mostly at startup of the pump. Once it's running 99.99999% of the heat comes from the motor of the pump.

"The same amount of energy that is inputed must eventually leave the system. Do you agree?"

Yes that extremely small amount must leave the system.

"The only way this energy can effectively leave the system is in the form of heat. Do you agree?"

Yes, the extremely small amount of energy is turned into heat.

HOWEVER

What we are talking about is the electrical energy of the pump. The only way the electricity inputed to the pump can be turned into heat is from the resistance of the materials of the motor. This is the wattage rating you see on the label. Which has nothing at all to do with the kinetic energy of the pump spinning...

[redleader]

Hmm good point. I just assumed the wattage rateing on the pump was average, not full load. Since things are generally rated conservatively, you are probably right.

Another thing, I don't really understand coils well, but how much can the workload of the impeller even be varied? I was under the impression that the impeller is riding the alternations in AC current. If so that would sugust that the same magnetic field is being induced regardless of the impeller. There must be some feedback from the magnet on the impeller, but would that be enough to significantly alter the current flow into the coil?

[ 09-16-2001: Message edited by: redleader ]

[LiquidCool]

If you know how a normal electrical motor works then you know how the magdrive type pumps work...

With a electric motor. You have 2 pieces the coils and the stator. The coils are the outer piece and the stator is the piece that spins.

All they are doing with the magdrive pumps is creating a hollow stator with magentic material on the inside of it. Then inside this hollow is the waterproof chamber for magnet. When the stator spins the magnetic material attracts the magnet on the impeller which causes it to spin with the stator.

I dont think the magnet causes any interference with the flow of current. I think AC doesnt have a problem with magnets, not to sure though. I dont think DC motors like magnets to much, but again not 100% sure...

Maybe I'll break one of my old pumps apart and check out the insides and see what I can see...

[phatman42]

Originally posted by Michael Huck:
The electrical energy of the pump is NOT completely converted into heat. If it was then the motor would not spin. The only reason heat is generated by the pump is from the resistance of the materials of the motor and also from mechanical moving parts in the motor(bearings/brushes).

Think what he should have said is all the electrical energy is eventually turned to heat. Electrical -> Kinetic -> Heat in one form or another