Intel P3 800 cB0 Peltier needed

Alives · 01-31-2001, 05:11 PM

Ok I have a cB0 800 that I want to break the GHz barrier with (1.2 would be nice, but I'll settle for 1000mhz). I have the 1.5 mag drive pump and a dangerden waterblock on it now. I was wondering what the ideal peltier would be for this.

Thanks

Rich W · 01-31-2001, 11:44 PM

All that depends on the thermal resistance of your cooling setup, how well you've insulated, and your intended power supply.
Here's a good 'reference' formula (I say reference because it assumes ideal conditions, and 100% input power):

CPUTEMP = Ambient - ((1 - HeatOfCPU/MaxQ)*deltaT)
Figure out what kind of temps you want, and then solve for MaxQ. Remember, HeatOfCPU changes as the voltage changes. Look for an article from me soon.

Rich W.

Alives · 02-01-2001, 01:43 AM

on intel's site the heat of cpu for my 800eb is 20.8 ( http://support.intel.com/support/pro.../sspec/p3p.htm ). Is this the normal or load amount?Also would it be better to have 9 degrees or 20 degress? I guess if its gonna be below 32 then condensation will be a problem anyway so why not go more? Or am I wrong? I was either looking at the ICE-71 (19 deg) or the Drift-0.8 (9 deg) from this site: http://www.tedist.com/specs/index.html

thanks!

Joe · 02-01-2001, 08:33 AM

I have an 850 I took to 1Ghz with a 130watt Pelt from Radioshack (dont laugh.. they sell em!)

------------------
C-ya
Joe - Owner/Editor
www.ProCooling.com
Where the Completely Addicted Come to Cool Off

Somebody set up us the bomb.

Rich W · 02-01-2001, 10:44 AM

Well if you can handle the insulation, go for as cold as you can. The 20.8 watt figure you speak of is the heat the CPU gives off at normal voltage. If you raise the voltage, the heat the CPU gives off will be (about) proportional to the increase in voltage... ie, if you increase voltage by 10%, heat output will go up by the same amount.

Rich W.

Alives · 02-01-2001, 02:31 PM

Heres my idea behind insulation. Basically youre trying to keep as little air inside your big foam surrounding. There is still some air and it will condensate, but it doesnt really matter because there are no components on the top of the processor. In my previous setup I pulled the foam up just enough for any water to drip out...I know air rushed in to replace it, but its still slowly getting rid of the water. you think that would be ok?

Joe · 02-01-2001, 02:44 PM

AIR doesnt condense.. Water in the air does.

If you do a good job sealing it all you will not have any worries if there is still some air pockets in the sealed area. I would suggest getting some Silicone coating to seal the entire works.

Alives · 02-01-2001, 03:26 PM

nah...i'll just let it condense a little...and bleed out the water when it collects in the botom of the foam jacket. I did it this way with my first setup. Once you bleed it out say 2 times, the only water left in it is like tiny drops on the metal. Know where I can buy a potted Drift Peltier? I was thinking about 1.15 maybe? also, tedist doesnt sell potted ones.

Alives · 02-01-2001, 03:30 PM

nevermind im gonna go with this one: http://www.radioshack.com/product.as...Fid=900%2D8562

Freakyfrank · 02-08-2001, 07:24 AM

Quote:

Originally posted by Alives:
Know where I can buy a potted Drift Peltier? I was thinking about 1.15 maybe? also, tedist doesnt sell potted ones.

??? i've ordered 2 172W POTTED pelts from tedist..

------------------
OC'ing is my middlename

Kevin · 02-08-2001, 07:33 PM

Basically w/ condensation, if it is air-tight, then there can't be condensation really... No water should condense on the components even if they are not in direct contact w/ the foam. Of course, putting them in direct contact is always a better idea. If your insulations is air-tight, the water cannot condense in cold temps, because of the pressure of an air-tight seal.

01-31-2001, 05:11 PM	#1
Alives Cooling Savant Join Date: Jan 2001 Location: Columbia, SC Posts: 228	Intel P3 800 cB0 Peltier needed Ok I have a cB0 800 that I want to break the GHz barrier with (1.2 would be nice, but I'll settle for 1000mhz). I have the 1.5 mag drive pump and a dangerden waterblock on it now. I was wondering what the ideal peltier would be for this. Thanks

01-31-2001, 11:44 PM	#2
Rich W Cooling Neophyte Join Date: Feb 2000 Posts: 77	All that depends on the thermal resistance of your cooling setup, how well you've insulated, and your intended power supply. Here's a good 'reference' formula (I say reference because it assumes ideal conditions, and 100% input power): CPUTEMP = Ambient - ((1 - HeatOfCPU/MaxQ)*deltaT) Figure out what kind of temps you want, and then solve for MaxQ. Remember, HeatOfCPU changes as the voltage changes. Look for an article from me soon. Rich W.

02-01-2001, 01:43 AM	#3
Alives Cooling Savant Join Date: Jan 2001 Location: Columbia, SC Posts: 228	on intel's site the heat of cpu for my 800eb is 20.8 ( http://support.intel.com/support/pro.../sspec/p3p.htm ). Is this the normal or load amount?Also would it be better to have 9 degrees or 20 degress? I guess if its gonna be below 32 then condensation will be a problem anyway so why not go more? Or am I wrong? I was either looking at the ICE-71 (19 deg) or the Drift-0.8 (9 deg) from this site: http://www.tedist.com/specs/index.html thanks!

02-01-2001, 08:33 AM	#4
Joe The Pro/Life Support System Join Date: Dec 1999 Location: Denver, CO Posts: 4,041	I have an 850 I took to 1Ghz with a 130watt Pelt from Radioshack (dont laugh.. they sell em!) ------------------ C-ya Joe - Owner/Editor www.ProCooling.com Where the Completely Addicted Come to Cool Off Somebody set up us the bomb.

02-01-2001, 10:44 AM	#5
Rich W Cooling Neophyte Join Date: Feb 2000 Posts: 77	Well if you can handle the insulation, go for as cold as you can. The 20.8 watt figure you speak of is the heat the CPU gives off at normal voltage. If you raise the voltage, the heat the CPU gives off will be (about) proportional to the increase in voltage... ie, if you increase voltage by 10%, heat output will go up by the same amount. Rich W.

02-01-2001, 02:31 PM	#6
Alives Cooling Savant Join Date: Jan 2001 Location: Columbia, SC Posts: 228	Heres my idea behind insulation. Basically youre trying to keep as little air inside your big foam surrounding. There is still some air and it will condensate, but it doesnt really matter because there are no components on the top of the processor. In my previous setup I pulled the foam up just enough for any water to drip out...I know air rushed in to replace it, but its still slowly getting rid of the water. you think that would be ok?

02-01-2001, 02:44 PM	#7
Joe The Pro/Life Support System Join Date: Dec 1999 Location: Denver, CO Posts: 4,041	AIR doesnt condense.. Water in the air does. If you do a good job sealing it all you will not have any worries if there is still some air pockets in the sealed area. I would suggest getting some Silicone coating to seal the entire works.

02-01-2001, 03:26 PM	#8
Alives Cooling Savant Join Date: Jan 2001 Location: Columbia, SC Posts: 228	nah...i'll just let it condense a little...and bleed out the water when it collects in the botom of the foam jacket. I did it this way with my first setup. Once you bleed it out say 2 times, the only water left in it is like tiny drops on the metal. Know where I can buy a potted Drift Peltier? I was thinking about 1.15 maybe? also, tedist doesnt sell potted ones.

02-01-2001, 03:30 PM	#9
Alives Cooling Savant Join Date: Jan 2001 Location: Columbia, SC Posts: 228	nevermind im gonna go with this one: http://www.radioshack.com/product.as...Fid=900%2D8562

02-08-2001, 07:33 PM	#11
Kevin Slacking more than your weird uncle Join Date: Nov 2000 Location: San Diego, CA (UCSD) / Los Angeles, CA (home) Posts: 1,605	Basically w/ condensation, if it is air-tight, then there can't be condensation really... No water should condense on the components even if they are not in direct contact w/ the foam. Of course, putting them in direct contact is always a better idea. If your insulations is air-tight, the water cannot condense in cold temps, because of the pressure of an air-tight seal.

Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)