Lemon Block Cu 2 (development thread)

[jaydee]

I am in the process of building a sound deadening box for my little CNC mill. Once completed and if the box works as I hope it does I can start milling some stuff again. This project is #2 on the list of things to mill, but it also need the most design work. Below is a pic of my first drawing. As I change things I will update. Right now the pins will remain 1/8" x 1/8" and the channels in between the pins will remain 1/8". I would like to go 1/16" but I am not for a couple reasons;


1) Much less mill time. (A little more important than performance at the moment)
2) Mill isn't fast enough for 1/16" and it also has a little to much slop in the table.
3) Higher flow rate. (most important)

Anyway this is version .01. Each whole number (1, 2, 3) will be a working prototype. Each non whole number will be a paper prototype. At least that's how I have it planned.

My goal is to get the same performance (if not better) than Lemon Block Cu 1 but also cut down on milling time. I already have some mods in mind. I also need to figure out the bolt pattern to hold the top on... Also next drawing will be in ACAD. This one is Corel Draw 9.

[jaydee]

Also considering this being a pelt is not going to be used and this will be for Current AMD CPU's without the IHS. This would be much easier to make and I suspect the performance would still be pretty good.

[JFettig]

This is the first time this block has been posted on the internet, The performance isnt amasing or anything but much better than a maze block.



I tried varius jet inpingments, I am now gonna try it without a jet and see what it does.

And I did it on my manual mill.

[jaydee]

Nice work. That is encouraging. Looks like 1/16"ish pins/channels?

I am not going to use any type of jets on mine as performance is not the main goal. High flow rates is. I want to use one pump for 3 blocks. My plan is to build 3 of the final version of this block and use them in a multiple computer loop. I got 2 dedicated Distributed Folding Comps and will have one more by the time I get these done and will be building a big box for the whole thing to fit in. Going to put the 3 comps (not in case's) and the water cooling gear all in it for my Distributed Computing Center. My attemps to try and get more organized...

I would like a single inlet and single outlet design as hooking 3 blocks will be much easier that way.

[winewood]

how thick are the bases below the pins on both JFettig and Jaydee's blocks? with the greater number of pins, will it work on a IHS, have you measured the area you may need to cool?

Looking gooood guys. I think we are on the starting of a large pin-fin wave

[JFettig]

yeah, 1/16" pins and channels,

Nah, we arent on any pin wave, I have made about 3 pin waterblocks in the last 6months or so.

and the one is pyramid style in the center for good flow(when I tried the 1/8" and 3/16" jet)

The base thickness, I choose to keep to my self.

Jon

[jaydee]

Originally posted by winewood
how thick are the bases below the pins on both JFettig and Jaydee's blocks? with the greater number of pins, will it work on a IHS, have you measured the area you may need to cool?

Looking gooood guys. I think we are on the starting of a large pin-fin wave

I think the pin fins are becoming more popular. The designs simply work. I am suprised it has taken so long to really start getting noticed though. I developed and fabricated the Aluminum Lemon Block about 2 years ago now. The Copper one is pretty much the same except with a few changes to work better with Copper. Swiftech has been using similar design for who knows how long.

The Lemon Block Cu will work very well with IHS CPU's and with TEC's. The link is posted above, I belive all the specs are in that thread.

I can guesstimate the specs of Jon's block there but I will keep it to my self.

The block I am designing here will have all the specs posted. Currently I am not even sure what they are yet though. It will be designed around current AMD Barton CPU's though. I have some work to do in that area yet.

[Sin22]

About your 2nd image jaydee116 when I was playing around with my clover block in its experimental stages, I had wanted to do a single inlet and single outlet. But whilst doing some CFD analysis on the flow, I kept on noticing one small problem with it, this;



Fairly large recirculation zones in the areas furthest from the outlet.



Adding another outlet solved the issue quite nicely. But that was specific to my design at that time.

Basically, with your second pic, I feel that the cooling efficiency of the outer pins furthest from the outlet wouldnt be up there, they'd more than likely have fairly stagnent water flowing around them.

The Cascade got away with the single vs. dual outlet issue to my knowledge because of the jets. Each jet forcibly evicted the water already residing in the cups and since that was where all the cooling effectiveness of the block was, it wouldnt have impacted too greatly if part of the block exhibited stagnent water flow.

Hope that helps a little.

[jaydee]

Sin22: I have thought about that and am still thinking about it. Notice the circle in the middle is the "actual" ID of the inlet which will be 1/2". Same type connectors I used in the last one. I belive there will be sufficent force there to cause the water to circulate around those outer pins being the inlet partially covers them all except the very outer 2. Also being this is going on a 3 comp loop I think the performance would still be better as adding multiple Y adapters woold hamper performance.

I am also toying with the idea of moving the inlet over. All these have there drawbacks though..... Still pondering things.

Input appreciated!

[Sin22]

Very true, my concerns were more for that solitary pin left all by its lonesome just out of the inlet area and fairly close to the outer wall. That said, the loss of cooling efficiency of one pin isnt going to be staggering, and anyhow, u've stated u've wanted ease of manufacturing instead of extreme performance.

Good luck, I look forward to see what the end design is.

[Blackeagle]

Like the limited area design JD, and would advance your goal of high flow as well.

You make some nice & interesting blocks.

[JFettig]

Jaydee, All I really want to say right now is that your trying way too hard...

Take a look here:
http://www.wc101.com/JFettig/wb1/

that thing stomps dangerdens blocks.

[Blackeagle]

With the new limited area design can you drop down to 1/16" roughing mill bits Jaydee? Could have 16 pins with nearly the same surface area as the larger diamiter pins. And with the smaller pin area would take no more time in machining, or at least not much.

And if you were to use those bell shaped inlet/outlets as in your last Lemon copper block you could use one that would give a velocity boost just befor going into the block.

And, depending on base thickness perhaps a few dimples with a drill between the pins to further increase surface area and improve turbulance.

I relize your goal is high flow + ease of making these. But the roughing mill would not add a great deal to machine time would it? Same with dimpling the base.

And with the limited pin area the flow should still be very high.

Edit:

Possible alturnative, 3/32" roughing mill bit? I'll be quick to admit I don't even know if this size is offered.

Wish I had a mill and space to set it up. Has to open up many interesting chances to tweak things, and play with design.

[MMZ_TimeLord]

Jaydee,

A thought just struck me on your second design (ver 0.02). Sharpen the center pin like a pencil (lathe?) so the water has a straight shot to the baseplate. Then all your impengement is the water having to find it's way around the rest of the pins. :shrug:

[jaydee]

Yeah I could very well be making it more complicated than need be.... But I think I will still stick with the pins. Maybe...

The smallest roughing endmill I can find is 1/8" and I would have to do it in one pass. In other words I would mill out the block as I want it then stick the roughing endmill in and make one pass with it to rough the sides up. Thats going to be about a 1/4" deep pass and I am not sure it will work with my small mill. Non the less I will be trying it. If i stick with the middle inlet I could add some dimples around the inlet area. Being three blocks will be in the same loop I don't want to add anything to increase volocity as that would also create a added restriction. I don't want to use to big a pump.

I have considered making the middle pin a cone shape. If I do I will just take the dremel to it. if I stick with pins....

[jaydee]

Hummmm, might try something like this to. Similar to the Maze 4 except bigger....

Blue area is where the water would go...

[jaydee]

I decided to go with this one below. The red is approx. the die size of a barton core. This also will be the easiest/quickest to mill. I will try to get it drawn up with tool paths in ACAD and run one out this weekend. If it looks good i will make a couple more down the road when I get the rest of the project going.

[JFettig]

MMZ, did you see my pic? theres quite a nice point on the middle pin, and on the rest the edges are beveled.

Jon

[MMZ_TimeLord]

Indeed in my haste I missed it behind the text... that's what I get for reading...

Nice work

[jaydee]

Going to have to come up with a new name for this thing. Has no charicteristics of the Lemon anymore.... Maybe the "Egg Block".

[Sin22]

Been used before jaydee

CMF Forums

Well it is deceased in some sense so I guess the name is up for grabs

[winewood]

Jaydees KingPin block
Jaydees PinPin block. or Pim-Pin block?

[jaydee]

I got looking at the picture with the red box representing the core and started thinking if the pins where not 45'ed then the pins would cover more area of the heat output from the die.....

Also by making the pins normally square milling will be considerably easier. Now all I have to do is strait passes with a 1/8" endmill. Will make using the roughing endmill so much easier aswell as I can use the hand wheels for that and go much slower. In fact this entire block could be made with hand wheels! Not that it will be though. If the performance is good then it will be a good example of an easy block to make on a manual mill.

KingPin eh??? Hummm...

[Blackeagle]

How's about this, "The Edge Block"

For the edges of the pins done with care to increase surface area of the pins edges.

Also liked the idea of having the central pin tapered to a point, with dimples around the pins.

Will be very interesting to see how much of a performance increase this will provide in a design that should still provide really high flow rates due to the lower number of pins. In the limited area just above the core this design should offer a large amount of surface area for heat removal.

While you may have posted the pump you intend for use in the 3 system folding set up I can't seem to find it. What pump will power your Tri-Folder Jaydee? Or will that choice wait for a final design of the CPU blocks?

[winewood]

What if that inlet was a little to the right. Therefore ALL the water would be directed right to left over the pins. The pins would all have direct downward water on them but you would eliminate the impact of water returning back from a trip rt to left. It may help flow increase through the block.

Before
Outlet Inlet
XX <-- <- P -> \
XXX <---- <- I ->
XX <-- <- N -> /

Outlet Inlet
XX <-- <-- P - -
XXX <----- <--- I - -
XX <-- <---N - -

sigh.. the forum takes out spaces. Oh well.. there is always the written explination.