Sand Blasted WB

[Tecumseh]

OK, I've been lurking for a while...and now I'm legit.

The Question: Have any of you water cools guys sand blasted
the **INSIDE** of your waterblocks? It is difficult to estimate
the magnitude of the effect, but easy to speculate that it
may very well increase turbulence, thus reducing temps.

If you try this be sure to clean out the glass beads/grit
before you put the block back together.

[KingB]

Theoretically, I would think it would reduce flow, therefore causing the system to reduce its ability to take heat away from the water block via the water.

(In reality would it really do anything at all noticeable?)

What do you think?

[Krusty]

Sounds very interesting. It would make for enough roughness in the sides to create a little turbulence, but not enough to cause any problems.

Of course, you would need to find a way to protect the top of the water block while doing this. You don't want the part you are soldering another plate on to get all messed up.

[The Overclocker]

i doubt sandblasting a waterblock will add any increase to turbulance or ony other aid. the closest think i can think of sandblasting a watrblock in the consumer world is the becooling jagged edge waterblock.

[jdmcnudgent]

welcome to the forums, where are you from in OHIO?:p

[Bender]

Hey Tecumseh welcome to the forums. I saw the outdoor play Tecumseh here in Ohio and it was excelent.

[Tecumseh]

Thanks for the replies!! I'm touching Wright-Patterson
Air Force Base----Home of the AF Museum....

What I am thinking is that the internal finish would be
a course satin-like surface. Would it reduce water
flow? Maybe, but that **Could** be good. Here is
my thinking: We know if we make the channels too
small we restrict flow and that's bad :( . BUT if the
turbulence is increased this is more of a dynamic
thing instead of static. What I mean is that the
water is fighting itself. Who knows?

The real problem is that this a very non-linear
situation and just how important each feature
is is open to debate until someone tests it.
I don't think sandblasting could hurt.

Bender, I think the outdoor theater for Blue Jacket
and Tecumseh are really Great!!

[Bender]

I han't seen blue jacket but Tecumseh was awsome. I have seen it twice at the outdoor theater. I didn't think I could enjoy a play so much. It fealt like I was actually there and the characters were so real.

[errorz]

If you use a high gritsize, I am sure you could do quite a bit of dmaage. I work for a company that installs spray booths, amogn other things. Did one for disneyland not too long ago. You can spray it with simple sand, or even a very fine green diamond dust which does quite a bit. I'll have to look into it.

[Tecumseh]

errorz, I do belive you can do a lot a damage with a sand blaster.
I once saw a guy who had a small compressor setup who
blasted custom designs in glass, however, who seemed
to have a light enough touch.

I sure hope someone can try this.

[errorz]

I will try :). I will have to speak with the guy before it happens (obviously). Want it nice and rough, but not earth shattering. He shuld know the correct grit/velocity :) thanks for the idea

[KingB]

Hey found this water block from a site in Australia. It has a pretty unique path that promotes turbulence. Now this may make a noticeable difference (at least more than sand blasting).

The Cyclone

[Tecumseh]

KingB, doesn't that look like a genetically modified DD2?
Anyone know of an online test report or side-by-side
test with the DD?

I believe that a rough satin finish on the inside of a
water block might improve the turbulence in any
design.

The real question is: IF it does help, will it be a
measurable improvement?

[KingB]

KingB, doesn't that look like a genetically modified DD2?


Now that you mention it????

Here's a link to there site, if you would like a look.

Silverprop (http://www.silverprop.com/sp/cooling.asp)

[Diggrr]

There was an Aussey Gent on here a while ago who posted saying he Loved his Silverprop block.
And the different finishes look SO cool too, compared to the usual copper that's getting so common place now.

[Aesik]

There have been several engineering studies showing the effects of different surface finishes on the promotion of turbulence. Sandblasting has proven to be quite effective in triggering turbulence under certain conditions. I haven't read a study that had any close correlation to the physical attributes of a waterblock, but there is enough data out there to convince me to try it.

My next series of waterblocks is going to have a base model with smooth walls, then several other models with different turbulence enhancing features. It's going to be a long project to undertake, but I think it will be worth it.

[The Overclocker]

i feel drilling small holes in the waterblock like on the koolance waterblock will so a much larger improvement then what you are considering, turbulance is much more improtant then flow rate or surface in watercoooling because of the high thermal capacity of water. have a look at the swiftec waterblock which create turbulance - they are just an open space but perfrom as well or better then other blocks.

[Tecumseh]

I suspect that there are various scales for turbulence.
Clearly the "macro level" enhancers like jaggedness
and such help.....and I think that they are needed.
However, I think there is a "micro level" that can
enhance heat transfer even more.

Hey, if you are making your own blocks this is easy enough
to try. If I had a mill and sandblaster I would do the test.

Aesik, you know of any online references for this?

[Aesik]

I don't have anything handy. Everything I have is either from my heat transfer texts, or from articles published in scientific engineering journals. Unfortunately, to get a copy of most of the stuff you can pay $30-$50 per article, or head off to the local university library and see if they have a copy. I've got a few printed articles, but unfortunately they are very difficult to read even with advanced heat transfer knowledge and most don't directly relate to our situation.

If you do a web search on something such as "heat transfer enhancement pipe turbulent flow" you can find a few useful web pages but most are just summaries and don't have all the nitty-gritty details. I'm working on building some numerical models and setting up my DOE, but it's going painfully slow since I'm doing it on my own time (of which I really don't have much).

There are definitely varying degress of turbulence and my follow on article to the laminar/turbulence article will try to address the basics of simple calculations. I'm going to try to show how turbulence itself effects heat transfer.

I'll dig into my archives and see if I can find anything that I can post that would be useful to you that I can post.

[Aesik]

I just found a pretty nice thesis about heat transfer enhancement in ducts. This guy was using different styles of ribs to get the effect. Problem is, you have to view the thesis page by page as a gif and it's rather painful. Worth the read though if you have the time.

http://theses.mit.edu/Dienst/UI/2.0/Describe/0018.mit.theses%2f1997-352?abstract=

[Tecumseh]

Excellent!! There is a lot of information here.

What is wrong with those MIT people? :mad:
Wish they just put the pdf out like everyone else.
Kind of makes them look greedy or like ****,
don't you think?

The Thesis is worth a read. Thanks for the link.

[res0r9lm]

Originally posted by the overclocker
i feel drilling small holes in the waterblock like on the koolance waterblock will so a much larger improvement then what you are considering, turbulance is much more improtant then flow rate or surface in watercoooling because of the high thermal capacity of water. have a look at the swiftec waterblock which create turbulance - they are just an open space but perfrom as well or better then other blocks.

I'm going to have to agree with overclocker on this:burn:

[Aesik]

I started downloading all the images and reading them in a ACDSee (so I could 'flip' pages easily) but my mouse button finger was getting tired. I'll have to finish getting it later. It is a very interesting read on the methods used and the patterns that he came up with.

[Tecumseh]

I downloaded part of the thesis Aesik referenced above.
While I didn't get all of it downloaded either I do have
an observation snagged from the paper.

It looks like in the laminar flow region that roughness
does not cause turbulent flow and has little effect.
However, in the turbulent flow condition the roughness
contributes significantly to the overall heat transfer
coefficient.

So, the overclocker is correct about needing structures
to cause turbulence and my assumption about sandblasting
increasing heat flow is also on the money assuming
turbulent flow has already been established.

Who wants to volunteer to sandblast their WB? :cool:

[Silver]

I like the idea, alot. Might give it a go on the next one built.:)

[labrat]

sounds cool, might work, but i doubt

Good luck though:beer: :burn: peace out:burn: