Made my own waterblock!

[MadDogMe]

I like the idea of this block but I think it sorely needs more surface area. Don't discount the thicker BP, it's strange sometimes the amount of diffefernce it can make being thicker(or thinner), it can spread the heat out quicker to cover more surface area...

I'd try scoring lines in down the lenght not across with a stanlyknife/linolium~knife as deep as you can get them...

PS, Did you know alot of proffessional engravers shops use mills with 'tiny' bits nowadays?. I have a peice of brass that has been milled with words and had them filled with paint to make them standout. there are two depths milled, the top is wider, 2mm and the bottom is 1mm, it's only about 1~1.5mm deep but would be OK for your BP size/thickness...

It's like they milled it with a 2mm bit then went over it again with a 1mm bit in the centre of the 2mm channels...

[UnloadeD]

I like it too. I notice you are comparing it to Iceberg block, are you still using the Iceberg pump/res? If so I'd suggest a different pump, maybe a Maxijet 1200 for $20. Are you still using the 80mm rad from the kit too? If you are using everything but the block, it might be hard to realize any improvments as you tweak the block.

peace.
unloaded

[Fixittt]

maddog,

When they evgraved it, they used 1
endmill, it was either a ball nose which is rounded almost to a point, or a V shaped bit.

[msv]

Originally posted by bigben2k

No, I don't have a sandblaster either. I found that they're only good for two things: cleaning grafiti, and etching glass details.

I have an Opel, so I´ve used sandblasters a lot. Rust removal, You know.

I asked mr. Steve Owens about sandblasted surfaces in water blocks, reffering to rough surfaces and turbulence build up, but he thought that just sandblasting wasn´t enough.
I do am very curious about other opinions and experiences, though.

regards
Mikael S.

[Rayman2k2]

Originally posted by Fixittt
lol cleaning grafitti ... TURK 182!!!!!!!!!!!!!! (the movie for all you youngins)

good old Turk....

[Graystar]

Originally posted by UnloadeD
I like it too. I notice you are comparing it to Iceberg block, are you still using the Iceberg pump/res? If so I'd suggest a different pump, maybe a Maxijet 1200 for $20. Are you still using the 80mm rad from the kit too? If you are using everything but the block, it might be hard to realize any improvments as you tweak the block.

Yes I am, that's one reason why I'm encouraged by the results. I actually have a Velocity T1 pump on the way. This pump will move 540 GPH at 3 ft of head, or 120 GPH at 24 ft of head!! I also bought a heater core. Once I get those hooked up I should get much better results.

[Graystar]

Originally posted by MadDogMe
I like the idea of this block but I think it sorely needs more surface area. Don't discount the thicker BP, it's strange sometimes the amount of diffefernce it can make being thicker(or thinner), it can spread the heat out quicker to cover more surface area...

Well, this has been mentioned before. I think I should note that when I touch this block, it is not hot. That is, when I touch the block as close to the core as possible, I cannot sense a temperature difference. This tells me two things. First, I'm doing a pretty good job of transferring the heat to the water. Second, a larger base will probably have little to no effect on cooling.

I don't understand why people think that a thicker base will spread the heat quicker. This is backwards. A thinner base will spread out heat faster than a thicker base. This is so, because the heat input is constant. This makes the base (at the die) hotter, due to the smaller temperature gradient through the copper. Heat will move through mass faster when there is a greater temperature difference between the center and the edges. If thicker was better than all the air cooled heatsinks would have thick fins.

In any case, moving heat through copper is exactly what I *don't* want. I want to move the heat to the water.

Originally posted by MadDogMe
I'd try scoring lines in down the lenght not across with a stanlyknife/linolium~knife as deep as you can get them...

Why in that direction?

The original idea was to create breaks to the flow, to add turbulence to the water. If I change the direction, I won't get any turbulence.

[bigben2k]

Whoa! I never meant to imply that a thicker base would spread the heat quicker! The purpose of a thicker base is to spread the heat, so that the deltaT is at a point where the water can pick it up better.

Think about an extreme: if the source heat was say 300 degC (assume a 1500W source), the water would pick up a lot of energy, but it certainly wouldn't pick it up fast enough to lower the temp back down anywhere near 20 or 30 deg C.

I think that's where your thinking is at. Yes, a higher temp will transfer more heat, but you're also allowing your source to remain hot.

There is an optimal set of conditions at which the water will transfer the most amount of heat, a.k.a. a maximum efficiency point. You're just shooting for maximum heat, without considering the efficiency of the heat transfer. That's what's biting you in the rear end!

[Graystar]

Originally posted by bigben2k
Whoa! I never meant to imply that a thicker base would spread the heat quicker! The purpose of a thicker base is to spread the heat, so that the deltaT is at a point where the water can pick it up better.

Yes, sorry. Didn't mean to imply you, or anyone specific in that comment. I've just seen the idea tossed out in various places.

Originally posted by bigben2k
Think about an extreme: if the source heat was say 300 degC (assume a 1500W source), the water would pick up a lot of energy, but it certainly wouldn't pick it up fast enough to lower the temp back down anywhere near 20 or 30 deg C.

I think that's where your thinking is at. Yes, a higher temp will transfer more heat, but you're also allowing your source to remain hot.

There is an optimal set of conditions at which the water will transfer the most amount of heat, a.k.a. a maximum efficiency point. You're just shooting for maximum heat, without considering the efficiency of the heat transfer. That's what's biting you in the rear end!

Ah!! Methinks you're mixing your apples with your oranges You've introduced a factor that is independent of block design.

Inside the block, water is being heated. However, that water is also moving. Any given gram of water is being heated only for the short time it remains in the block. The quicker the water is removed, the less heat it will absorb. This sounds bad. *BUT*, the quicker you remove that gram of heated water, the quicker you can get in a *fresh* gram of water that is starting off at the base temperature. This is good. This is the simple reason why increased flow cools faster.

So, to your comment I would answer "it depends on the flow rate." Thinking extremes, At flowrate = infinity, heat transfer = infinity, and the copper touching the water would be at the water's initial temperature all the time and all heat would be absorbed by the water. (Yes, in the real world friction would be a problem. But lets stay in the fake world for a while ) The temperature of the CPU die would be limited only by the thermal conductivity of the copper. Start slowing down the water, and now the water has time to heat up, which slows heat transfer through the copper...the CPU die heats up. So you simply slow down the rate to the point that gives you the cooling you're looking for. This will be true, regardless of the size of the heat source or size of the surface area. Of course, you could end up needing a google of gallons per minute for a given situation, but you will get the cooling you want.

Instead of saying that there is an optimal set of conditions at which the water will transfer the most amount of heat, I would put it as there is an optimal set of conditions at which a particular water cooling configuration will transfer the most amount of heat. I would say this because BillA has already demonstrated what physics says should happen...increase flow rate makes a water block work better. So flow rate must be considered, but the radiator's ability to get rid of that heat must also be considered, as that is the true limiting factor to how much heat can ultimately be removed.

I'm looking at a temp reading right now of 41C on a processor that is known to run really hot (early 2200+) that's running at full load. And this is with a crappy rad and pump. I really can't complain. This block is working well (at least, that's what I'd like to think! )

And I didn't even lap the base! Just rubbed it on some 400 to clean it up.

[bigben2k]

I think myv65 is going to jump in here, soon...

Yes, flow rate is one of the factors that I was reffering to.

The turbulence of the water increases with the flow rate, and this turbulence allows a higher heat transfer. It's not simply a matter of higher flow equals higher transfer, i.e. the proportion isn't linear.

Here's an OK article on the subject.


What I'm talking about is the efficiency of the transfer. It's not just a matter of getting a lot of heat transferred, but doing it efficiently, i.e. with minimal flow/pressure, which includes the right deltaT, and hence the right bp thickness.

Are you still up for the 3mm piece?

[Graystar]

I understand what you're saying, and I agree that the design of a block can cause turbulence to kick in at a point, causing a non-linear heat transfer rate.

Are you still up for the 3mm piece?

Sure am! I'll send you my info.

[Fixittt]

as far as I can see, a thicker base plate would allow you to take advantage of gaining more surface area. By allowing any markings in the plate to be deeper. Therefore multyplying the surface area. and by applying your markings in the path of the flow instead of against it, you will minamize water sitting trapped in the recesses.

If that makes sence

[Les]

Dunno,but maybe the different configurations could behave like this:-


The flow-rate being controlled by the pressure-drop (dP).Estimating the dP is fraught with difficulties, but a summary of my guesses:-

[bigben2k]

Les to the rescue! (Thank you!)

As you can see Graystar, a 6mm baseplate performs significantly better than a 1.1mm bp. The 6mm bp is 40% better than the 1.1mm. (All this is in theory, but Les is pretty sharp about these things)

You can optimize your design for a 3mm bp, with what Fixittt mentionned.

[Graystar]

I'm sorry, but I just can't figure out what those charts are showing me. I don't know what "30x10x1.1mm" means. The calculation for Convection Coefficient considers area only. What's that third dimension? Is that the channel height? Is that implying that my channel should be taller?

Also, as Convection Coefficient is a property of the boundary, it is unaffected by base thickness, and can be considered a constant when trying to calculate the effects of base thickness. Therefore, I am confused by its appearance in the charts.

Finally, there is absolutely no explanation whatsoever as to how the C/W is calculated. "C/W" is Thermal Resistance. The thermal resistance through a solid increases with the length of the solid. That is a fundamental. In the matter at hand, base thickness is the length. For any given area, it is impossible for a solid with a longer length to have a lower thermal resistance than a solid with a shorter length. So I simply don't understand what that first chart is demonstrating.

I also have no idea what the second chart is showing, or why it's there.

Les, can you explain the charts and how you arrived at these numbers?

[Tweety]

Hello all... I'm gonna poke my nose rigth into the fun... Although I'm new to this site, I'm not entirely new to making blocks...

My instinctive reaction to that block was.... "Hmmm wonder how that block would perform with a baseplate similar to #rotors"...

Another thing... those temps with the rad and pump you currently use, on a dual, that tells me that that little block is an overachiver, performing much better than many blocks twice it's size... I think personaly I would however extend it to rest on the pads. Just my own paranoia thoo...

I think you will get better and more universally comparable results with the heater core and the better pump. But after reading the discussion I'm only more curious... What results would grooves give???

I got the impression you had a dremel? that could be used to make those groves, shallow lines with the cutting wheel... A way of getting both turbulence and avoiding trapped water is to make "diamond shaped" grover ie a cross pattern but turned 45 degrees...

Anyways... I'll be back... count on that... here is a lot of fun stuff to read around here...

[Graystar]

Originally posted by Tweety
Hello all... I'm gonna poke my nose rigth into the fun... Although I'm new to this site, I'm not entirely new to making blocks...

Welcome! And thanks for the vote of confidence!

Originally posted by Tweety
I think you will get better and more universally comparable results with the heater core and the better pump. But after reading the discussion I'm only more curious... What results would grooves give???

I'm almost tempted to go buy an Eheim 1048 and hook up my new heater core, just to give out some comparable numbers. I have a Velocity T1 on the way, but that's not good either because few, if any, have such a pump.

We are going to find out about the grooves. I'm going to make a couple of blocks with groove and with different base thicknesses. Then we can see what factors will make a difference. The grooves will be very shallow lines; not deep enough to trap water. I actually did think about the cross pattern. I decided against it because the grooves are so shallow. I think more turbulence is better at this stage. I probably will create a block with a diamond pattern at some point, just to see how it works.

[Les]

Originally posted by bigben2k
Les to the rescue! (Thank you!)

As you can see Graystar, a 6mm baseplate performs significantly better than a 1.1mm bp.

Oh.
The intention was to suggest that bp thickness has a relatively SMALL effect on the predicted cooling for the small(30x10mm) Convection Area (with the envisaged Convection Coefficients).

[Les]

Originally posted by Graystar

1)I'm sorry, but I just can't figure out what those charts are showing me. I don't know what "30x10x1.1mm" means.
2) The calculation for Convection Coefficient considers area only. 3)What's that third dimension? Is that the channel height? Is that implying that my channel should be taller?

4)Also, as Convection Coefficient is a property of the boundary, it is unaffected by base thickness, and can be considered a constant when trying to calculate the effects of base thickness. Therefore, I am confused by its appearance in the charts.

5)Finally, there is absolutely no explanation whatsoever as to how the C/W is calculated. "C/W" is Thermal Resistance. The thermal resistance through a solid increases with the length of the solid. That is a fundamental. In the matter at hand, base thickness is the length. For any given area, it is impossible for a solid with a longer length to have a lower thermal resistance than a solid with a shorter length. So I simply don't understand what that first chart is demonstrating.

6)I also have no idea what the second chart is showing, or why it's there.

Les, can you explain the charts and how you arrived at these numbers?

1) "30x10x1.1mm" is a reference to the water chamber dimensions(lenght x width x height) of your wb.
2) The Convection Coefficient is calculated by Kryotherm*. Kryotherm uses the Sieder-Tate relationship. Have calculated at various flow rates for two Channel sizes(10x1.1mm and 40x0.8mm).It is plotted against flow-rate(LPM)
3) Third dimension ( "1.1mm and 0.8mm"?) is channel height. Not syuggestions on channel height made. Used 0.8mm for larger bp area(ed) wb since had calculations to hand.
4) Only plotted Convection Coefficient against flow-rate.
5) C/W(for 10x10mm heat-die) is calculated by equating the Thermal Conductance at the bp/water interface with the Film Coefficient in a Waterloo Spreading Resistance calculator** . The Thermal Conductance through the bp/water interface(Edit****) is taken as the reciprical of dimensionally corrected "Kryotherm K/W" for a 0.1mm bp. The "Total Resistance" from Waterloo is added to an estimate(0.1C/W)*** of the TIM Thermal Resistance to give a C/W value.
6) The second chart shows the predicted Pressure Drop(dP Total)) v Flow-rate (LPM) for the two considered wbs and th P/Q curves for three pumps. The intercepts of the pump and wb curves gives the predicted flow-rates - from which graph1 gives the predicted C/W for the different different pumps.


* http://www.kryotherm.ru/soft.htm
** http://www.mhtl.uwaterloo.ca/old/onl...ce/strip2.html
*** From Billa [url}http://www.overclockers.com/articles654/index02.asp[/url] making an allowance of 0.05 C/W for die RTD offset.

**** EDIT: "through the bp/water interface" did read "of the wb".

[Graystar]

Originally posted by Les
Oh. The intention was to suggest that bp thickness has a relatively SMALL effect on the predicted cooling for the small(30x10mm) Convection Area (with the envisaged Convection Coefficients).

Yeah, that's what I gathered. However, I'm still confused by the chart a bit. But I'll mull it over some. It should click eventually.

[bigben2k]

Originally posted by Les
Oh.
The intention was to suggest that bp thickness has a relatively SMALL effect on the predicted cooling for the small(30x10mm) Convection Area (with the envisaged Convection Coefficients).

Hum... you're correct, as usual! The difference is a mere 10% at best, but the 1.1 mm baseplate actually outperforms the 6mm, at flow rates greater than 5 lpm.

What's clear is that the velocity pump will take this block where it needs to be.

[mikildemion]

I have a thought may work may not work not sure just thinking here.

keeping the base the same thickness but adding 3 layers between the top and the bottom. 2 of these layers will be hollow (center cut out) and in between these 2 layers will be the 3rd with holes cut in to it using a drill bit (2 - 3 mm in size) all over it.

Now my reasoning for this thought is 2 things, 1 it will increase the amount of water in the block (lower flow but higher volume) and second the holy layer will cause turbulance.

Would this cause more issues (higher temps worse performance) or would it increase performance?


I love the design, and the thought on the thickness of the base is good also. Since I have used this thickness in my own designes along with using 1/2" plexiglass as the block its self and the copper just for a base.

I am not a total n00b to making water blocks but I do not know all that much. Just enough to spend money and break stuff I guess.

MD

[Graystar]

Originally posted by mikildemion
I have a thought may work may not work not sure just thinking here.

keeping the base the same thickness but adding 3 layers between the top and the bottom. 2 of these layers will be hollow (center cut out) and in between these 2 layers will be the 3rd with holes cut in to it using a drill bit (2 - 3 mm in size) all over it.

Now my reasoning for this thought is 2 things, 1 it will increase the amount of water in the block (lower flow but higher volume) and second the holy layer will cause turbulance.

Would this cause more issues (higher temps worse performance) or would it increase performance?


I love the design, and the thought on the thickness of the base is good also. Since I have used this thickness in my own designes along with using 1/2" plexiglass as the block its self and the copper just for a base.

I am not a total n00b to making water blocks but I do not know all that much. Just enough to spend money and break stuff I guess.

MD

Thanks for the kind words. I appreciate it.

At first thought, your suggestions would go against the design goals I had in mind for water flow. I had two goals. The first was to have a very thin channel. Really, the only reason the parts are .043" thick is because I didn't think I'd be able to work with anything thinner! Otherwise I would have an even thinner base and thinner channel. The second goal was to get the water in and out fast. That's why it's a straight run from one side to the other, and why the outlets are so close together.

So here's the concept. Everyone knows that turbulence improves cooling. It does so by taking water that is in the middle of the stream (and not being heated) and bringing that water to the block surface where it can get heated. My idea was to make that "middle" so small that the inherent turbulence of simply flowing through the block would be enough to fully heat the water.

However, I don't want the water to get too hot. NO NO NO! I want that water out of there so that I can bring in fresh cool water. Therefore, I don't have any barriers to flow, except for the narrow opening. The narrow opening, coupled with the thin channel, helps to boost the flow rate of the water through the channel. This gets the water out in a hurry. Water that gets out fast is heated less, but it is better to have 2 grams of water zip through the block than to have 1 gram of water linger in the block for twice as long. The 2 grams, though at a lower temperature, actually drew more heat from the block.

So, to address your suggestions. I believe that the lower flow/higher volume idea would decrease performance, as higher flow is always better (regardless of block design.) Also, I would want the turbulence action to occur on the base of the channel, not on the ceiling. I've touched the block and it isn't even hot. So all the action is occuring on the base side of the channel.

On the next block I will be scoring the base in an attempt to increase turbulence. So we'll see if that has an effect.

[mikildemion]

I understood the original reason for the design and agree with it. What popped in my head was a different design based on your original idea. I posted it to possibly see if my idea would even work. Sorry if was confusing on what I was proposing.

The idea about the turbulance layer was not to be at the top but in the middle of the block. Lets see if my ASCII art is still worth a damn:

[=========] <- top
[=========] <- hollowed out layer
[=========] <- turbulance layer (2 - 3mm holes)
[=========] <- hollowed out layer
[=========] <- base

(ok my ASCII art sucks)

Maybe i should ask it this way: would adding the middle turbulance layer to your design (or any WB design for that matter) increase or decress performance of the WB?

The very small foot print of your WB is something I have been looking for. That is why I am following your successes and failures. But was just wondering if by modifying your original design and increasing the core size slightly would give even better results then I may have found one of my next projects. if not then I'll keep looking.

I appreciate the time
MD

[Graystar]

I *do* think that making the channel wider will help. I didn't do that initially because of the unknowns of working with such a thin base. However, I'm over those fears now. The .043" thick material is *plenty* strong! I'll probably try the wider channel at some point. I have enough material to make 20 of these, and they only take about 2 hrs to make (if that) so I can try all sorts of variations!

However, I don't think that your turbulence layer would work. I actually did understand your earlier description. But you don't actually have a middle turbulance layer of water...you have a top layer and a bottom layer of water and a middle turbulance "initiator", for lack of a better term. Each one of those water channels has a ceiling and a floor. I guess it's like a two story building!

The edges of your holes would create the turbulence as the water flows past them. The turbulence would occur on the floor of the top layer, and on the ceiling of the bottom layer. That's not what we want. We want the turbulence on the base, as that's the only place where we can pickup any heat.

However, I may be misunderstanding again. Are you suggesting that the water be pumped through the holes in the middle layer? I ask because I was thinking that the top and bottom layers would be fed in the same way that it works now. If you're thinking of pumping the water through the holes, then, I don't like the idea, mostly because you've now created a place to trap air (the original design was also meant to pump air through as well.) Also, in that configuration there doesn't appear to be any real "route" for the water. I mean...yes, the water will go in and come out, but you don't get a sense that every water molecule is on a definite track getting in and out. You might have some hot spots.

Of course, trying it is the only real way to tell. Go for it!