The Lemon Block Cu
 

I have finaly got around to milling up a Copper version of my Lemon Block. The Aluminum version beat out my Maze 4 by 1C so I decided to make a Copper one with some modifications to take advantage of Copper.

http://www.customcomp.us/lemoncu/001.jpg
http://www.customcomp.us/lemoncu/002.jpg
http://www.customcomp.us/lemoncu/003.jpg
http://www.customcomp.us/lemoncu/004.jpg
http://www.customcomp.us/lemoncu/005.jpg
http://www.customcomp.us/lemoncu/006.jpg

Unfortunatly the 1/4" acrylic top is a NO GO. It has to much flexability and doesn't snug the O-ring down in the middle enough. It worked ok around 250GPH but once I opened the pump up to 500GPH setting it leaked. So I have to make a new top. Havn't decided to try 3/8" acrylic or 1/4" copper yet... So in other words no testing as been done. I hope to have this completed and running by next weekend because if I don't it may not get done for a long time.

What's it going to take to get you to switch to polycarbonate? ;)

not to insult anyone, but isnt this basically a milled version of a #Rotor? Also, doesnt the area around the square pins lend to having more water flow there and less over the pinned area?

I would have to buy some. I have tons of scrap acrylic at work. Acrylic works fine. Just need it to be thicker which I also have. Just have to cut the piece. Possibly will do tomorrow at work on the laser.

hydrogen: I wouldn't say a milled Rotor Block. Not even close really. His has many more pins that are smaller and not sqaure. I came up with this before I even knew #Rotor existed about 2 years ago. But the concept is the same just as any pin style block like what Swiftech, Koolance, and a few others I can't remember use in their commercial blocks.

And no, I have no idea what you mean "over" then pinned area. The pins touch the top peice. Water is forced "through" them and it works well (note the CENTER INLET). Watching the air bubbles go around every single pin and exit to the main channel and then to the exits is rather cool. The AL block works very well and I expect this one will be better by a few C.

my wording was a bit confusing, let me clarify. Water is injected in the middle, and seeks a path of least resistance, which would be towards the sides of the block, where there are no pins, just channels. So flow would not be maximised across the pinned area I would think.

The middle IS the pinned area? :confused: It starts in the middle and goes left, right, up, and down. Which is around the pins and out the outlets. The pressure forces water in all directions. You can see the water hauling ass all around the pins exiting to the outler channel and then the outlet. How can flow not be maximised around the pin area if that is where the flow is originating from? :confused:

http://www.hydrogen18.com/stuff/lemonblock_flow.jpg

woudlnt the flow be like that?

Below is how it visually looks through it when actual water is flowing through it. Not sure why you think it wouldn't go left to right? Thats the path of least resistance and the shortest route to the exits, not up and down..

Well, you know, McMasters has those 6" by 6" samples of 1/4" clear polycarb, cheap. In fact, it's on my ship list to you, for the making of Radius. ;)

I can't bend it. I can't even make it flex!

I need to get some 1mm endmills and find out if my mill will even handle them. I figured out how to slow the feed rate way down on the mill though. Will have to dig that link up of those 1mm endmill and pick up some. I got all kinds of stuff to try and make with them. :D Including a modified version of this Lemon Bock.

Go polycarb all the way.

I was a staunch "no plastic" policy person on waterblocks. I spoke to my machinists about it and learned that all the cracking blocks were mostly made using acrylic based plastics.

We did some tests on polycarb. Tapped an M3 hole (~1/8") into some 1/4" polycarb, and drilled a 3mm hole through some 4mm thick aluminium plate and stuck the lot in a vice and used a cap head allen key bolt to tighten the aluminium onto the polycarb to test the strength of the polycarb.

The polycarb thread did eventually strip, but only after the bolt had bitten into the aluminium by almost a millimeter deep.

Even after the polycarb thread had stripped there was no way you could pull the two apart by hand as the polycarb just "bounces" back into the thread again.

We did the same test on acryclic and it just cracked as soon as the thread got even remotely tight.

After that I decided that polycarb was an acceptable substitute for copper/aluminium. Super-strong and highly crack resistant under stress.

If I was going to sell these they would be a poly of some type. Being I am not I am not worried about it. I will try the 3/8" acrylic first if that fails I will use copper. I also got some softer O-ring material aswell. The one in the pic is real hard. If I get around to making the revised Jet Block I will get some poly. I might just buy a sheet from work and cut up some blank peices on the laser if it will do it. If not we got a table saw setup with a special blade to cut acrylics and polies.

Life is going to get busy again after this week. the daughter is comming back from grandmas and I will have her on the weekends again which rules out any block making. Will try and tinker with things after work as much as I can.

http://www.drilltechnology.com/

5 Pcs EM2XL-0400 US$4.75/each
10 pcs EM2XL-0400 US$2.50/each

2 Flute Extended Flute EndMills -Square Cut

...and yes, they're center cutting. You have to e-mail them, to get that quote first. No TIN coating required.

JD,
what about off setting the pins so the water doesnt have a direct path out . Each pin will create turbualnce which will be effected by the other pins around it creating even more turbualnce.

http://homepage.ntlworld.com/lee_smith/lemon.jpg

do you have any way of giving the walls of the pins a pattern for greater surface area on your engraver ?

what is the base plate thickness

I did this on my personal CNC mill not the engraver at work. The engraver isn't capable of this. I might be able to find a roughing endmill (see pic attached) that would add mini channels to the walls of the pins. Something I have thought of doing for a long time but just never bought the endmill to do it. :D

That pattern is possible but I am not up to doing it. It would add about 10hrs (literally) to the milling process on my mill. But it is possible. The way I have it setup now would be hard to beat with that pattern thought. The flow rate through this thing is unreal as I have it. With that pattern flow rate would decrease, the pump would work harder and I think that would offset any gain. Even the 170GPH doesn't even flinch pushing water through it. That was one of my main goals when I designed it as back then (2 years ago!) high flow blocks were all the rage. And I am going to start going back to higher flow rate designs simply because I would rather use smaller and lower GPH and cheaper pumps in the future even if it causes a few C warmer temps.

As of now the base thickness is .175" or 4.445mm. and the overall is 0.375"or 9.525mm. The top will be .25" or 6.35mm Copper plate which is almost done. If the results are not as I expect I might shave the base down to .125" or 3.175mm but I do not think a design like this will benifit from a thinner base than that as I want the heat to spread on this one a little as opposed to the other designs where I try to keep the heat staged in the area over the die.

Forgot to attach the pic of the roughing end mill.

Very nice. Some "inprovement" over the aluversion such as the oring. Very beautiful indeed! I think you'll get some very good numbers with this one!

Any numbers yet? I'm curious how well this works out.

Finished the copper top and installed is just a little while ago.

http://www.customcomp.us/lemoncu/007.jpg
http://www.customcomp.us/lemoncu/008.jpg
http://www.customcomp.us/lemoncu/009.jpg
http://www.customcomp.us/lemoncu/010.jpg

Will report temps after it has been running for a few hours.

looks nice.

WOOT!

Love the belled inlet/outlets JayDee those are sweet! Have you considered changing the inlet one however so that it would go from line size down to the next smaller size to increase water velocity a bit as it enters the block? Might help just a bit.

Perhaps some small dimples in the base between pins? Easy to do with a drill press.

Those improved low resistence inlet/outlets combined with the copper construction should sure push a high flow rate and lower temps.

Looking forward to seeing your resulting temps.

Have you tested your flow rate with the new inlets yet? What change in flow vs the aluminum block do you see?

Really nice work man!

After running it load all night here are the temps.

The Maze 4 results were:

Air: 26C
Water: 30C
TC: 43C
Onboard: 40C


Lemon Block Cu:

Air: 27C
Water: 31.1
TC: 39C
Onboard: 37.5C

Considering the air and water temp are 1C higher than when I tested the Maze 4 I would say it is doing very well.

BlackEagle: Flow rate is excellent but I have no way of testing it right now. Can't even do a bucket test at the moment. I got the Hydrothruster 500GPH at full speed (as I did also with the Maze 4) and the pump is not stressed a bit. Next time I tear everything apart I can drag it all in the bathroom and do a bucket test. The Aluminum Lemon Block is currently hooked up to a 200GPH submersed chepo pump which usually has a hard time but it is near freee flowing through it. Just from the eye I can't see much difference. I would imagine the Aluminum one would be a little less restrictive because of deeper channels but the barbs are narrower to so I don't know. I went with the bell type connectors as they give a true 1/2ID as to where the brass barbs are no where near it even though they say they are 1/2ID.

As for the modifications I will keep that in mind. It will be a while before I can do any more messing around with this stuff again. Life is catching back up in a hurry. :D

I'd say a 4c+ improvement is pretty good for starters.

With a bit of tweaking it should do even better. And yet has high flow rates.

Use of that roughing end mill with a few other tweaks and this could become a very high performing block. While still holding on to almost all the flow rate advantages it now gives.

I don't suppose you'd have a White Water or Cascade about to test with? Would be interesting to see how much differance in temps and flow rate there would be.

Looking forward to you having more time to test & post more on this block.:)

I decided to make 2 more of these and use them in my multi computer single loop project. I got 3 desktop computers. 2 are dedicated Distributed Folding crunchers and 1 is my gaming and everything else rig + a D.F. cruncher. What I want to do is put them all in a case of their own (now 2 of three are just mobos sitting on a table) and water cool all of them on one loop with one pump and 1 or 2 rads that are mounted in the window blowing the hot air outside. My Hydrothruster 500GPH has plenty of power as these blocks are high flow rate blocks.

I think this will pretty much end my water block designing and building. This block surpassed all my goals and is easily adaptable to future CPU's. As Blackeagle mentioned there is definatly room for improvment yet but I am just not up to doing it. I been having some life changing events happen lately and water cooling just doesn't seem as important to me anymore. Got stuff in life to do other than design and build blocks and spend the rest of my time in the net. :cool:

This project is going to be a neat one though. At least for me. Part of my "get shit together/betterment of life" plan I am putting together. I simply am to addicted to the net and water cooling. Taking up to much of my life. You may have already noticed a lot less posting and I quit posting at the other sites.

Once I get started on this project I will document it on my website and I will start a thread here and update it as I make progress. :)

wow! now how about making me one and letting me test it?