PDA

View Full Version : ? for the H2O coolers out there...


dreadlord79
01-03-02, 07:55 PM
...I was looking through some of the articles and reveiws out front and got thinking about the following. Would it be of any help or benifit to put a temp. sensor in a water block? To show you what I'm thinking, I've attached the drawing below. If you guys think it will work, I'm going to learn how to melt and cast copper so I can build a working model. Thanks for the input! :)

ken257
01-03-02, 08:06 PM
I think I would put take the temp reading after the radiator but before the block. It's nice to know what the temp of the water is being cooled down to before it enters the block.

Warlord2
01-03-02, 10:43 PM
I agree


putting a temp senser before and after the block would be best


that way you know if your making good contact with the cpu

dreadlord79
01-03-02, 10:53 PM
So putting a sensor in the block is a bad idea all-around? What about putting one by the inflow and keeping the one I have in place where it is? I'm trying to simplify the "wires" thing, because as we all know, we have MORE than enough wires in our cases all ready. Thanks for the input! :)

Kunal
01-04-02, 11:05 AM
If you really want to go mad, you could have one temp sensor for fluid coming in and one for going out (both inline)

Then two normal thermal sensors, one on the CPU and one on the external of the waterblock/coldplate.

ButcherUK
01-04-02, 11:10 AM
What is that sensor trying to measure?
For cpu temp mesure you want a sensor in the baseplate of the block directly over the core. For water temp you can put inline sensors in the water tubing.

dreadlord79
01-05-02, 07:13 PM
I, personally am one of those crazy people who don't want to drill a hole in the heatsink that I just paid $30 and up for. My Fiance doesn't let me have a lot of $$ for computer stuff.:( I was just thinking that this might help in having some place to put the temp. sensors that is quick, easy and is still in the flow so you could tell if the rad. is actually doing its job of cooling the H2O) correctly. The sensor position in the bitmap shows that it would be put through the "wall" on the block and then sealed so no water would come into the sensor area. I might have put it in the incorrect place now that the fact has been mentioned. My line of thought on this was that copper can absorb a lot of heat as can water, so since most blocks I have seen are large, they are absorbing a lot of heat that the small water channels in them can't completely take away. I could be wrong here since I don't have a water cooled rig yet, but doesn't the extra copper harbor heat?

ButcherUK
01-05-02, 09:43 PM
copper holds a lot less heat than water per unit mass, I'm not sure of the relative masses of the copper and water but in general, no the ciopper wouldn't. The water is moving through the coper so there is a constant stream of cool water to prevent heat buildup.

As for drilling, you're willing to cast copper but not to drill a hole in it?! uh, ok...

eh?
01-05-02, 10:09 PM
Originally posted by dreadlord79
I, My Fiance doesn't let me have a lot of $$ for computer stuff
my gf doesn't either.... stupid women.....

dreadlord79
01-05-02, 11:50 PM
When I said no drilling, I meant not a purchased HS; sorry for not clarifying that. One thing I guess I am missing, when you drill the hole through the HS; don't you do this right above the die? If so, do you go all the way through? If so, wouldn't that cut down on the area for heat transfer from the die itself? I know these are probably dumb questions, but I guess I need a little help understanding some of these ideas.

ButcherUK
01-06-02, 11:20 AM
You drill in from the side about 1-2mm above the base of the HSF till you get to above the die and stop.you don;t want to make a hole in the base above the die, that would badly affect cpu temps. And make sure it's a small hole. :)

dreadlord79
01-06-02, 06:34 PM
I wondered why the bases of HS were always so thick, that would explain it! Thank You for all of your help! :)