shah n zahri...thanks for the answers...

but im still waiting for answer to question number 3

@shah81
asking out of curiosity je laaa... to gain more knowledge *knowledge is power* hehe

Answer to question #3 provided

Performance wise if by old school thinking, the 3 barb is a better deisgn as it's a impingement method which is meant to create swirl and improve heat transfer. Disadvantage is that it requires a huge HMax pump and also a very messy setup of Y joints too.


The Storm G4 and above was designed to circumvent the disadvantges of the 3 barb method and in the process has proven to be a very effective solution judging by it's Delta values.

1 month back i didnt see any of those, so settled myself with a brass barbs instead.

Ianho : Keep the waterblock in the mean time, busy with exams.

@almostthere

thanks for the answer.... so i take it a lower resistance block would be better as it requires lesser hmax which in turn results in lower heat dump?

am i right or am i just shooting blanks?

So its,

3 barbs needs Big HMAX to fully maximize heat transfer

2 barbs does not need Big HMAX.

Am i correct ??

Whats the ideal HMAX required for the 3 barb waterblock ??

Cannot assume this way either.

Ok, i'm gonna go a bit deep here, so look at the design pics of the blocks to get what i mean.

The WhiteWater was designed in such a way that it is much more efficient for the water to exit the waterblock via two exits on either sides. It's design implementing an impingement plate jetting into the microchannels made its restriction high. And because the microchannels are in a straight path to either direction (because the inlet impingement is in the centre of the microchannel path) the water has to exit at both ends of the channels in order for this design to be efficient, hence its two outlet design.

The storm series implements a blind hole jet impingement design. water hits the tiny distruptor's at the centre of the cup, then hits the base of the cup, and then flows out around the jet. After the water passed through the jet impingement stage, its job is done, the water is still recirculating between the midplate and the baseplate, and has to flow out of the block. hence even if there's one outlet, it will still do its job well. Because of the tiny dimensions of the jets and the cup, this made the Storm series highly restrictive. But in order for this design to work well, you need high flow rates otherwise you're better off with a simple low restriction pin grid design or direct path.
The Storm G7 implemented a hybrid 2 outlet design by adding another outlet on the other side of the block, but instead of exiting to to another outlet barb, it is channelled back to the main outlet right before where the barb is. The new Swiftech Storm G4 Revision 2 then implemented this as well.
Performance gains (temperature wise) from this improvement is very minimal, because the main part of the storm design that works to extract the heat from the CPU is the jet impingement. However, this hybrid two outlet design allowed a little less restriction to the waterblock and less recirculating water in the area between the baseplate and the midplate.
And so, in general assumption for high restriction waterblocks, the lower the restriction (without affecting its heat conduction properties), the higher the flow rate, and the higher the flow rate, the better performance. However, the performance and flow rate gains are very small on the Storm design because anyway, the most restrictive part of the design is still the jet impingement, not the outlet design.

We cannot apply the Storm G4's (revision 1) single outlet design on waterblocks such as the WhiteWater because, lets say if you were to block one outlet of the whitewater block and only use one outlet, the water has no choice but to only flow out from one direction. Water that is stuck on the other blocked end cannot move and will be stagnant forcing all water jetting out from the impingement plate to go out from only one direction, hence severely reducing the waterblock's efficiency. You could apply a hybrid two outlet concept to the WhiteWater's design also however, but you have to make both outlets to have equal amount of restriction in order to maintain its efficiency. Fluid will choose the path with the least restriction and therefore, more water will flow to the path with the lower restriction, hence also reducing its efficiency slightly because the two exit design is still the core element of the whitewater. You could flow the exit water around the microchannels to one side of the block parellel to the inlet, but that also poses more restriction. So in the end, the most efficient design would still be two outlet barbs.

This post has been edited by MetalZone: May 24 2006, 05:19 PM

Continued,

Ok now on to the high restriction two barb blocks other than the Storm.
Two examples of this is the Alphacool NexXxos XP and the AquaXtreme MP-05 SP LE/Pro. Both of these designs implement an impingement plate to increase the velocity and of the water prior to hitting the pin grid array. Because this design implemented a pin grid array instead of directional microchannels, water exits on all sides on the pin grid. Again, like the Storm design, this is the stage where most of the heat is absorbed. Hence the water's job is done in absorbing the heat, and now has to exit the block. So also in this scenario, one outlet will suffice. Also like the Storm design, the most restrictive part of the design is again, the impingement pin grid array, thus having one or two outlets makes minimal difference. Basically, any waterblock with an impingement design will be significantly restrictive.

Now on to waterblocks that purely utilises surface area. There are plenty of examples of these, from the Swiftech Apogee, Swiftech MCW6000/5000 down to even the ThermalTake BigWater SE and Gigabyte 3D Galaxy.
All these waterblocks just use two barbs, one in and one out. This is because all heat absorbsion of the water takes place by just contact with the surface area of the waterblock. Hence, in very general assumption, the more surface area, the better its performance. Generally, waterblocks of these nature have low restriction. Thus high flow rates are not required for this design to work. Thus, on low restriction waterblocks, higher flow rates does increase its performance, but gains are minimal. A low restriction waterblock will perform better than a high restriction waterblock at very low flow rates. This is because high restriction waterblocks generally have much less surface area to begin with. But when flow rates rise, performance gains on high restriction waterblocks will rise significantly.

This post has been edited by MetalZone: May 24 2006, 05:17 PM

to MetalZone for that badass explanation

but... for a n00b like me...i was blur as hell reading the thing without pix... so i decided to scour the net and find pix of waterblocks and its internals mentioned by MetalZone (also added not mentioned ones)... combined them and decided to post here.... so that other n00bs like me can read and look at the same time to better understand what MetalZone was rambling about and actually make some sense out of it....

so on with the pix (thank god for spoilers)

The WhiteWater


The Storm


The Storm G7


Alphacool NexXxos XP


AquaXtreme MP-05 Pro


Swiftech Apogee


Swiftech MCW6000/5000


ThermalTake BigWater SE


Gigabyte 3D Galaxy


Vantec Stringray


will edit this post to add in more if i ever have the time and actually able to find any

This post has been edited by mADmAN: May 24 2006, 10:23 PM

to tell y'all the truth..... I'M TOTALLY LOST!!!.... so stop hurting me and yourselves.

Question by mADmAN = comparing 2 barbs with 3 barbs waterblocks... which is better/preferred and why?

I'd rather you'd compare the water block resistivity-> water contacts rather than numbers of barbs... cause there's no definite answer if comparing barbs.. it's the w/block(design for heat transfer via surface area -> c/w) that determine the best blocks...some blocks design uses 2 OUTFLOW barbs cause (in simple PLAIN INGGERIS) to create a lower pressure point than the INFLOW... elementary fluid dynamics... water flow from high pressure to low pressure... so in theory the more the OUTFLOW barbs the merrier it is... right? hehehe there's more to it than that. read on

In some design, geniuses uses the INFLOW pressure not oni for jet impingement effect but simultaneously creating turbulance to further increase the surface contacts with thermal median.. turbulance will oso dampen the kinetics of flow which in turn lowering pressure of flow (a good example is the G4 block... the cups, channeling and such).. thus a single OUTFLOW barbs is adequate, without sacrificing the thermal transfer efficiency.

but.. imagine if the G4 uses a dual/triple/quadruple barbs OUTFLOW.... okay ladies, some food for thouhgt.. lemme know what you ladies think and why.. I think that's a healtier reply post rather than spam like post... (no offence).

Right. Just that designs like the WhiteWater are much more efficient with two outflow barbs because water flows out from the microchannels in ONLY two directions, out of each side.

Another good way of explaining.
To answer your last question, i mentioned it in what i explained, more outlet barbs DO increase flow rates (and higher flow rates=lower thermal resistance C/W). but then again, flow rate gains are minimal because jet impingement by itself is still very restrictive, in fact much more restrictive than a single outlet. Adding outlets will make it easier for the water to exit the block, hence less restriction for the water coming in from the jet impingement, thus you will still gain a small amount of flow rates.

my earlier post spam like post ar? I was just trying to explain what I knew hoping that it would at least make some sense. heheh


Anyway, to add to madman's pics, heres pics of the Storm G4 revision 2 to explain what a hybrid 2 outlet design looks like.

This is the midplate. As you can see, it has two outlets.


This is the top/upper plate. You can see that the right hand side, the water exits to an outflow barb. But on the left hand side, the water is channelled back to the main outflow barb. This will slightly increase flow rates, although gains are small. You could make the left hand side outlet to exit to another barb, but since the performance gains are so small, I'd rather only have two barbs, so does most people who'd like clean looking setups.


This post has been edited by MetalZone: May 25 2006, 03:09 AM

Any comment about this wc kit since got passive radiator. My friend want bring it back to Malaysia and he ask me if i want order from him. But look at the radiator.....

Y so susah buy from overseas? Here oso got a really good passive cooler in the form of the Zalman Reserator. Any problems can RMA sumore.

Anybody here have faulty Laing D4 pumps ? I need the impeller to exchange with mine. PM me if u have one.

Can order 1 IIRC. Or mebe it was Eheim I read about. Try google n c?

ucai@.. try finding Laing retailer, mebbe can order/repair parts from them.

Good News to all Water Cooler Enthusiast.

Now we have another supplier whose bringin in Danger Den Products to this Forum.

So lets Recap:

1. Danger Dinster's Danger Den's Watercooling supply.

2. Metalzone's Maniac Swiftech Supplies

hehehe just like WWE Monday Night RAW... so ladies, get your watercooling kit from em...

ps. (hint, hint, check out Danger Din's budget kit )

edit/O.T.: so now it's my turn to spam. , fair what.

This post has been edited by amok: May 29 2006, 09:56 AM

Yeah, will do it. I also recalled that we can order the parts but its for Laing pumps.

Wah, nice Dinster. Soon Lowyat.Net would be flooded by Watercooled Opty's.

OOT - Nieh terang2x inform lagi gempak.

Hawt! His prices are really good the advantage of taking directly from DD.
Budget kit V2 already coming on the way real soon!

GO go go, you guys got ample WC supplies now.

A big thank u to dinster. Now we dont hafta worry bout our WC supplies. Coz LYP is like a desert in terms of water cooling.

If any of you guys are looking for tubings, I would recomend the one that OP is selling, RM5 per feet. This is the same tubings that Aleck used to sell last time. Very flexible and doesn't kink easily. I find them to be better quality compared to the Tygon R3603 that I bought for about RM12 per feet last time

Is it a clear tube....I mean pure colourless?