• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

how many of you use a passively cooled radiator?

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

crave_silence

Registered
Joined
Mar 14, 2002
Location
seattle, WA
Hey all,

There was recently an article on the front page from a guy that built the 'African Queen', a water cooled system that had no fans whatsoever. I feel the same way he does about computer noise, and have been running a fairly standard water cooled system with no fans on the heater core radiator for quite some time.

I'm interested, how many of the rest of you are running passively cooled rigs? What speeds/processors are you running, and what are your temps like?

thanks!
 
Depends on your definition of passive... Most of he people here would find what I tout and spout is true if they'd ever actually try it.

You really can run any descent load with no fan directly moving air through a radiator. So long as you have enough of a differance between negative or positive airflow, you can get by with that if you can maintain it... It's that you have airflow through one, not nessicerily how much... Diminishing returns set in very quickly, just like waterflow through a waterblock. You keep getting a gain, but it tapers off into insignificgance.

True, the more the merrier, but why run two 120mm fans at stock speed? There wont be 2C differance between that and 1 120mm fan running 7v. We've all seen and accepted that. I've just been waiting for some other people to key in on it.

**********

The actual question is what are you willing to trade off for the desired effects...

I can't maintain 2016mhz off a pally 2000+ under any load by just indirect case airflow, but I can maintain stock speeds, and stock speeds under 40C via undervolting, with load temps in the 40's...

Undervolting is one of the most underused things in our hobby I might add. Of all 5 Athlon XP's I've had, every one will run stock speeds at 1.6v. One would run stock at 1.5, but I can't remember which one it was unfortunatly...

That right there can shave a deal throughout the temp range.

Really my thing is most people can stand low speed fans... PSU, 3 case fans and a good cooling passive NB block is perfect. You can add one 120mm fan and volt it to a slow speed, bingo. Not much more noise and you can still maintain your speed.

Now cool that CPU and GPU effectively with little airflow and we've got it made...

**********

One of my future projects will be low noise with good trade-off WC. Unfortunatly I haven't the funds to do so... Still planning on doing it within the next year, but that's a big time span and things here are uncertian ATM.

Keep in mind a system like this problably wont handle any serious overclocking. Then again... Anyone looking for something for nothing is looking for the wrong thing!
 
Toysrme said:

Undervolting is one of the most underused things in our hobby I might add. Of all 5 Athlon XP's I've had, every one will run stock speeds at 1.6v. One would run stock at 1.5, but I can't remember which one it was unfortunatly...

I've successfully undervolted Durons as well. Load temps dropped 2-3 C at stock speeds with no instability whatsoever.

Back to the topic at hand. I am currently mulling over a plan to passively watercool my main rig within the next year or so. I live in a relatively cool climate, so I would only have to worry about 25C+ ambient temps for about 3 months or so. I don't foresee any temperature issues with a passive rad in those conditions. If you're not overclocking, I would agree with Toysrme that it doesn't take much airflow at all to keep system temps within limits. If you're looking to keep your machine quiet, I'd way it's a good option for anyone to look into.

Ken
 
I tried to go no-fan on my P4 [email protected], but it couldn't hack it. The Northbridge gets too warm, and the video card has overheating issues at stock speeds, so theres just too much heat in the case. Had to go with 2 120mm's & 6v. Still nice & quiet, so I guess it works.

I do plan to try passive with my 2x366@550 celerons in my full tower case, but currently to busy to take on that project.

Don't think I could NOT overclock though ;)
 
You know, undervolting is something I haven't even tried with this chip, and is a great idea. Other chips I've had have been unstable when I've undervolted them, so I guess I'd already chalked it up as something that probably wouldn't work well with the athlon.

My passively cooled solution is UGLY. I got so sick of trying to cram all of the watercooling stuff neatly into my case (the hoses kept kinking because the heater core I picked up at the dump had outlet/inlets pointing in non-optimal directions), that I'm just running an inline setup with the side of the case off, the radiator sitting on the floor next to it, and the pump hanging from a small bungee cord from the top of the case so as to keep its vibrations from getting transmitted to anything that could make sound. UGLY. But for the first time in what feels like years, I've got a working setup that I don't worry about that I can just turn on and USE. It's a feeling that I'd forgotten about -- the fun feeling of actually using the tool you've worked on so much for its intended purpose!

The only fan I have is one of the NMB silent fans blowing INTO the power supply. It's not very powerful, but I'm more worried about the PS being cool than the NB -- it can exhaust out the open side panel of the case.

I've found my temperature monitoring doesn't seem to make much sense at all when watercooling as well. I'm using an Abit KT7A-Raid for this board, which has a thermistor which seems to be more realistic than any of the Asus boards I've used, but when the chip is WC, the chip is totally stable (compiling linux kernels ad nauseum, etc), but temps climb way over readings where it'd go unstable when air cooled. Case in point, I've got it underclocked right now to 733 mHz, and it's finally leveled off at 52 C. If it truly is that hot, I don't even care too much, as the stability is solid and the silence is BLISSFUL. Does anyone have any ideas about how comparable in socket thermistor readings when WC'd will be when there's NO air circulating around them like there would be when air cooled?

I'll try undervolting later. Thanks for the advice.
 
Well, depends on your point of view, but I think I'm running a passively cooled rig: one without radiator but with a huge (30 l) reservoir beneath my desk. Not the best possible cooling sollution, but it's ultra quiet (I can barely hear the pump at all if everything else,including the wife, is completely silent :D).
It does the job well enough to keep my Athlon XP 2400+ (non-overclocked) at about 35°C to 37°C all day.
 
Nice -- you're using another strategy I thought about when I was having troubles with my heater core being filled with gunk from the junked car I pulled it out of. Don't you have issues with your room being a bit on the humid side, or with stuff growing in the water?
 
crave_silence said:
Nice -- you're using another strategy I thought about when I was having troubles with my heater core being filled with gunk from the junked car I pulled it out of. Don't you have issues with your room being a bit on the humid side, or with stuff growing in the water?

Well, I don't have problems with humidity. The container (just a large bucket really) is closed with a lit. there's a small hole in the lit to allow for a temerature probe and the tubing coming out.

The growing stuff, however, is a bit more troublesome. So I have to refresh the water once in a while. And clean out the tubing as well. But all in all it is OK.
BTW adding 1 to 2 liters of anti-freeze helps a great deal of preventing growth in the water.
 
Back