Water Blocks roundup of my own
 

Hi
I finally reserved on day to test all my new blocks.
So the contesters are: two fin spiral Cu 6mm base, AL 6mm base, AL 2.5mm base and last my home made with one fin in the channel 3.5mm base. Here are all together
http://www2.arnes.si/~mlivak/Test1/All%20blocks%201.JPG
http://www2.arnes.si/~mlivak/Test1/All%20blocks.JPG

First three blocks were made on Cincinnati Arrow 1000 CNC mill 1h for AL version and 3h for Cu one. My was made this way: http://forums.procooling.com/vbb/sho...&threadid=1946

So next to my test setup. As my AXP suddenly died (it will be replaced), I had to use old D600 cpu which is even more difficult to cool because of the small core. Cpu was OC to 1079 Mhz at 2,1V (166*6,5) max Q= 84W which is higher than AXP 2000+ at 1,75V (71W) , mobo Epox 8KHA+, ddr 256mb pc2700 Kingmax, water pump Eheim 1250, all tubes 1/2" ID. First I setup mobo for easy and quick retesting of WBs.
http://www2.arnes.si/~mlivak/Test1/Setup3.JPG
Plexi plate on the back of mobo against to much bending of it.
http://http://www2.arnes.si/~mlivak/Test1/Setup2.JPG
And here is the cpu with thermal sensor touching the core.
http://www2.arnes.si/~mlivak/Test1/C...ith%20shim.JPG
I also measure temp. of the water on outlet of the WB.
http://www2.arnes.si/~mlivak/Test1/T...etup%20moj.JPG

For heating up the cpu I used Toast, so here are the results in graph:
http://www2.arnes.si/~mlivak/slo-tech/slika18.jpg
and here the numbers:
http://www2.arnes.si/~mlivak/slo-tech/slika19.jpg
Temp on the outlet after 20min of Toast were 20°C except for the first block that I test (al-2.5mm base) where it was 19.6°C
Here is another pic. with Al block testing.
http://www2.arnes.si/~mlivak/Test1/T...setup%20Al.JPG
From results you can see that Cu blocks are better, but not by mile like some think. CNC machined block will go in small production here in Slovenia by my friend for more info email me.
Btw I'll retest when I get another AXP.
Uf, my longest post so far.
Questions, opinions comments.

looks cool, but for some reason those results are all 404'ing for me.

I forgot that I moved those pics. to another location. I think they work now.

taking the temp at time = 0sec (ambient) and time = 1200, we see that the results are:

Al 2.5mm = 5.3C
Al 5.0mm = 6.6C
Cu 5.0mm = 5.5C
Cu custom = 4.4C

this is assuming that the ambient stays the same throughout the test

Those CNC'd blocks are just beautiful pieces of work morphling. From the looks of things you are getting very good results, bravo! I would deffiniatly be interested in some details and what the possibility of getting one in the US would be if you are thinking about producing some of these blocks. Anyway, good work, very impressive.

temp at 0 sec are idle temp in w2k just before I started Toasting , and the ambient temp was the same through the test.
Me and my friend are going to make those blocks, actualy first batch of AL was allready made and shipped in Slovenia. And shipping to USA wouldn't be a problem. I would send AL block to Joe for testing if somebody need to see his results before buying, he just have to give me the address.

what was the ambient throughout the testing?

VERY nice work :) Enough so to get me out of hiding, LOL!

Your custom block uses one of several design concepts I've been toying with to improve block performance in general (the "split path" concept). Really good to see theory and reality correlating so well :D

I would be curious how the copper spiral would perform if it's base was milled down to 3.5mm, the same as the custom block's base. Might we convince you to do that test? (with the different base thicknesses, it doesn't give the best comparison regarding the channel designs).

The other thing is that I am a bit surprised that the water temp at the outlet is exactly the same for all but one of the blocks, considering that you have a 4 degree C temp difference of the core for the top 2 and bottom 2. Are you sure that you have the thermocouple inserted so that it sits in the center of the water flow and that it isn't touching the copper pipe it's inserted through the wall of? You might also want to put a piece of foam insulation around the pipe and hose where the water temp sensor is to try to isolate it thermally from the ambient environment.

And of course, it would be great to see some tests with a higher heat source (100W to 120W range) and possibly even a set with a pelt combo so the heat input is in the 200W to 300W range. In general it provides very useful data and insight into a cooling combo to perform th testing at different Q levels.

Again, GREAT work :)

Thanks,

First to Brad, ambient temp. was around 19-20°C , my radiator is doing great work, with this load it keeps water temp within 1°C above ambient.

EMC2, we're going to make 4mm Cu and AL block for production (I was wrong about base thickness of Cu&AL block, it is 6mm and not 5mm, I was pretty tired and made the mistake and did't see it). As for the water temp I measure it at the and of the 20min test of the first block and it was 19.6°C , and then rose to 20°C with second block and stay there through the whole test. The tip of the probe is definetly in the water, and I don't think insulation would help cos' the ambient is so close to water temp.
I will do another test when I get my replacement AXP overclock the hell out of it and made the test even better, learning from this test.

LOL, I understand the tired comment completely ;)

There is one other important piece of info missing (or maybe I just missed it). What is the water flow rate through the system? And when you make your next test run, could you vary it at all? Wouldn't have to be a lot of increments, maybe just 2 or 3, say around 1gpm, 1.5gpm, and "give me all she's got Scotty" or if just two, then around 1gpm and 2gpm. If nothing else you could put a ballcock valve inline to vary the flow on the cheap. ;)

Also at a lower flow rate (maybe much lower since we don't know your current flow rate, but I would guess in the 2gpm range or more with your setup), we should be able to see a delta in the outlet water temps for the different blocks which would provide good info on the heat transfer capabilities of the blocks. The idea here being not testing how well they cool with high flow rate s, but intentionally lowering the flow rate to get better insight into the heat transfer capability of the blocks themselves. :) Of course, then it would most likely be best to put a temp probe on the inlet side of the block so we could see the DeltaT across the block ;)

(just suggestions in an effort to get the best data, ignore it if it gets to be too much, LOL)

I like that block!!! The removable top is something I have always been partial to and ALL the blocks I make will have one if possible.

This also shows that CU and AL can vary per design as to what the temp outcome is. The first block I made I made it in Copper and AL and totally identical. I had 1/2 a C difference in temps, but i think it was because of the wide open channels. I am betteing that a design like this copper would perform better because of the added surface area.

Great work!!!

morphling, ok, there is a big (5C) differance between Al and Cu in this case then.

have you thought about deeper channels, so there will be less flow resistance

EMC2, It's hard for me to measure flow rate in my system cos' it's closed loop with air trap and no reservoir, where I could simply calculate the flow rate. But your idea is good for variable flow rate to really see same delta T in inlet (I'll definetly have probe on inlet too) and outlet (my pump is Eheim 1200 l/h, so the flow is high), you have me thinking of bilding some kind of flow meter that I would put in the loop and than could simply put valve to vary the flow, that is all doable.
Ok, any idea for digital or analog flow meter, can you buy one, how did Joe measure flow in his test.

jaydee116, I like the removable top too, no problem with leaking with o-ring , for Cu version I'll try and make 10mm plexy top cos' its so much lighter than Cu, or maybe anodized AL one, just have to check the prices for that.

Brad, like I sad will make WB with 4mm base, and that is It for Cu, I have to ask my friend if he make that depth Cu is very,very hard to CNC especialy for desing like this, no problem with AL, as you can see with 2.5mm WB.

I corrected graph and the table, so that all temperatures are normalized to 20°C which was my average through the test.

you've screwed up the graph image again ;)

:D

Yes, at least here in the states flow meters are easy to come by.

Joe would have to pipe up with the exact model number of the flow meter he used, but according to his review it was a really nice Swiss flow meter that puts out a tach signal just like a fan. He said it was a bit over a $100US in his wb review. You might be able to find it for less over there, since it is imported to the US from Switzerland.

On the lower cost side you can get inline flow meters for about $45 US that you read directly off the side of the gauge. They are more flow restrictive than the one Joe used, have to be read by eye, and not as accurate, but less than half the price. For an idea of this type of flow meter go to http://www.McMaster.com/catalog/108/html/0444.html and scroll down to the 3rd group of meters, the Acrylic Block flowmeters. A 1/2 NPT meter that'll read 0.5 to 5gpm is $47.

Morp, I am glad U are finally getting some good results with that block. It is very nice. Beautiful machine work, good attention to deburring.

I have a suggestion that might help shave off a degree or 2. Have U thought any about sand blasting the inside> (this goes for U 2 Jaydee) An endmill makes nice smooth walls, and sandblasting will pit the walls, very small, but pits none the less, I am sure this would create a small amount of turbulance, and increase surface area as well, I have a small sand blasting gun I got from wal-mart ($12) but U need an air comressor, if U dont have one, I am sure the shop that U had these made at does.

Just a thought.

Sand blaster!!! That sounds like a good idea, but not possibe just yet for me. Would make a mess of the apartment.:D

if you don't want to do that (even outside your apartment) you could buy some 50 or 80 grit sandpaper, and use that

Maybe i'm just talking out of my ass
 

Your block looks good and it's an impressive design, but i think your design restricts flow, if you expanded the channels where they split off to give space for water that the extra channel walls take up it would increase flow rate, by removing resistance when the water flowing into the block is forced through the overall smaller space.

just an idea.

Nice work! :). I like your custom blocks.
-Kev

Warning: Sleep deprived mind, be on the look out for gibberish
 

FYI, if the channels are 3/32" wide then they only need to be 400 mils tall to have the same cross sectional area as the inside of a 1/2" fitting. Even if you go down to 1/16" wide channel, you only need them to be 600 mils tall to maintain that cross sectional area.

As with most things in life, there's a trade off involved between opposing factors. In this case it's flow restriction, surface area, and block height.

I would be interested in getting a waterblock from you, they look good and seem to perform pretty good. Could you post or e-mail the prices you were thinking of, I'm in the UK, London, I think I'd prefer one of the custom copper ones.

It is hard to get sand paper inside the channels, I already tried with little sucess. I little sanding wheel on my dremel might work though.

Thanks, Fixittt . Sandblasting is doable but I don't think it would help a lot in my case, you see my friend is using one final machining step that I don't know english word for, but the surface isn't smoth at all. I'll try and make on very closeup photo so you can see. As for my custom made block I use dremel and roughen inside surface of the block.

Ultra PeePi ,like EMC2 wrote it's the cross section that matters most and that's larger than the cross sectional area of 1/2" fitting (9mm ID).
What now everybody likes my custom made block more :D

Prices hmm, I have to checked with friend but informational prices for cnc blocks, actualy we named it M&B blocks (morphling & Boeing)M&B Al 45$, M&B Cu 65$ it's not just material, that almost doesn't count, but 2h longer machining (3h for Cu, 1h for Al block), they both come with 1/2" OD brass fittings, teflon washer, and all the nuts and bolts for instalation to mb.
For my custom block I would say I could make it for 60$, but now everybody would want that, and my time is very limited I'm home only on weekends until end of April, so I could make few blocks, but for CNC it shouldn't be a problem once we optimize process.
And Btw production block would have 4mm thick base. And of course I'll retest them with higher power AXP and posted results here and if somebody would give me Joe's address, or where I could send M&B block for test, I would really apreciate. Any other info you can email me: turbo1@email.si

Yes I am also very very interesting in obtaining one of these beauties. I am in CO USA. If you could PM me, or E-mail with a price and some ideas for shipping that would be awesome.

Ok, people, here is close up pic. the best I can do with my cam.
http://www2.arnes.si/~mlivak/close%20up1.JPG
http://www2.arnes.si/~mlivak/close%20up2.JPG
You can see the how the surface of the chanell looks.

decodediesel, send you pm.

Boy that takes an hour to machine? WOW, if U notice most machine shops charge $60 an hour, that is $1 a minute. So right there your block costs $60 without materials or setup time. OUCH!

The cost of tools and machining is just covered, and we can make a little money.