does bong size matter?

[Warlord2]

I am about to make a water tower and was wondering what a good bong size would be?

it will be cooling a 750mhz duron one chipset and one video card
Im hopping to get the duron to atleast 1000mhz

I am going to try and get the bongs water to atleast 4-5f below ambient

if all goes well I may put some pelts on it so I do need some good cooling

I was thinking maybe 4'? and put a 7volt 92mm fan on it

[flounder43]

Sorry, a little off topic, but my duron 750 runs at 1.0 Ghz with the stock heatsink. I have not had a problem with heat at all. Just thought I would share that...

Just checked, 41C full load, 27 case temp, this is at 1.85 volts...

[Warlord2]

I might just try it on my tbird and skip the duron:D

but I realy need a computer for lans:(

[Silver]

Early on I ran a duron 750 at 960 w/1.89 volts and a dominator heatsink. My chip would go no higher stable period. I built a six foot bong, added a vmod and a dd maze2/1 and the chip would go to 1120 stable without a pelt. Presently I am running a 1.4 t-bird with the same setup and at a little over 2.2 volts the cpu will overrtake the waterblock. I am presently looking at possibly building a better water block. I really do not want to go to a pelt until I have the best water cooling that I can think of. This buys time for the xp chips to come down and what the heck, this 1.4 is in the same ball park (more or less) as the xp's. Hope this helps.

BTW this 1.4 on a dominator 2 heatsink barely works at 1.82 volts due to overheating. Bad heatsink for a t-bird that is overvolted.

[Silver]

A couple of more things.

1) Bong (in georgia) holds temps 4 to 10 degrees below ambient (due to humidity and ambient temperature).

2) Return temp on 2.3 volts is approximately 4c higher than entry temp. This (seems to me) allows plenty of cool water to adequately cool a second device (video).

3) Use a large resevoir (I use a five gallon bucket).

[Warlord2]

Silver thanks for the info:)

thats a nice overclock
what temps are you getting? and how big is the fan on your bong? and how much water do you think goes away in a week? and lastly how loud is the waterfall? is it louder is fans?

sorry for all the questions :D but I havent seen very much info on bongs here

I can get my 1.2@1.5 stable at 1.81 volts so Im sure I can go higher with more volts and better cooling

[Spartacus51]

As far as height goes on a bong, who knows what's necessary, what's best, from what I've read they're all good, if constructed well.

Here's proof

http://www.overclockers.com/articles389/index06.asp

Instructions for a 12" bong.

The other 5 pages of that article are also good reading if my memory serves.

[Warlord2]

yup Ive read every last page:)

I think I will go with 4' or 5'

[FrozenInHI]

I built my first one using 4" piping for the tower, it was just over 3 foot tall, and used an 80mm fan in the wye connection to cool the water, the total water that was in the bong and system was around a gallon and a half. the temps were just 3c under ambient ( i live in Hawaii, take that into account) and my cpu temp was about 2 to 6 celsius over room temp. Next i built a 4" pvc bong with a 5 gallon tank and it's 4 feet tall, same fan, added a dedicated sprinkler for the bong and one for the outlet water from the system (2 in the top basically) and the temps of the water are a solid 5c under room temp, have to fill it more per gallon with the extra sprinkling, but it's more performance, my cpu temp is never over 4c above room temp. i'm running on 1.8v most days, sometimes 2.1.

[Silver]

I run at 2.18 volts right now as this is as high as I can go and maintaint stability. If the water is around 16c the computer will boot into 1.77 and almost get into 1.78. The chip will post at 1.82 so I have room. BTW this is at 2.33v and the water can not dissipate this load from the block quick enough (designing new block) as even with vcool running the temps keep climbing to failure. Bong height is not of critical importance and in fact the one item that effects the water temps the most is air temp blowing through the water streams.

Noise? A very soft (barely audible) rain effect. I have a 3m heavy duty pad at the bottom of the pipe (entrance to five gallon bucket which is very effective in killing the water noise. Pipe is cut into five gallon bucket (resavoir (sp)) with 1/2 holes drilled around the submersed bottom of the pipe. Two plastic nineties and a little vinyl tubing gives me a sight glass on the side. An aquarium temp (stick on) on the side of the bucket gives me rough temps. A radio shack probe stuck in line before the bong (shower) head gives me return temps along with ambient temps. Crude but effective. I go through about five gallons a week. Air flow is done by one hacked cd cover with a 80 25 cfm (read quite) fan dc. Had a radio shack 65 cfm ac on before and got lower temps 2c but was loosing a lot of water (sprites). Now it is almost as effective and even quiter with less water loss and the same stable overclock.

BTW the smaller the spray of water returning the quiter it is so you want to keep a small amount of pressure in the line. Pump, block and line design will dictate this.

[Silver]

Temps? Remember the voltage, t-bird and speed I'm running at. Using vcool, I get ambient at idle (got 2c below ambient with 750 duron) and 44c under load. Torture will get me to 48c.

[Silver]

One more thing, water temp is 4-10c below ambient depending on humidity and ambient temps. I have never had condensation during testing even under extreme (13c below ambient) normal type temps. I once put ten pounds of ice in the bucket (melted in 8 minutes) and did get some condensation under abnormal conditions. Computer started giving strange letters and I shut it down immediately. Dried the video and block using paper towel and hair drier (not too close) and let case fans run for 1/2 hour and all was good again.:p

[Greedy Guido]

The lowest water temp you will get with a cooling tower is the wet bulb temp. In real life you may design to get within 4°C of the wet bulb as to get any closer requires a very large tower. For a computer it is possible to reach this point with a cooling tower due to the small heat loads relative to the tower size.

If you want to work out optimum height for the cooling tower without going into complex calculations then you need to measure how high the tower needs to be when the air reaches saturation point. At this point no more latent cooling will take place.