Check it out here: http://www.swiftnets.com/products/H20-APEX.asp
Great product! Swiftech ROCKS :worship: :worship: :worship:
Printable View
Check it out here: http://www.swiftnets.com/products/H20-APEX.asp
Great product! Swiftech ROCKS :worship: :worship: :worship:
wow its actually the best kit available right now I believe
Wow that is nice plus it has a Storm block Cathar rocks :worship: :worship: :worship:.
hehe i used to have my bix2 mounted on my radbox like that 4 months ago lol
copycats!
yea... as soon as i finish my rig i'm heading to the patent office LMAO
that kit looks awesome. I love the drive bay reservior, very clean look.
welp my job is done here. looks like maxxxracers soap box is out of buisness.
is swiftech's radiator performance good? or should i just get a radbox and a dual heatercore instead?
looks like someone followed Maxxx's suggestions a bit too well ;)
does anyone actually use the coily things to go around your tubing, i think it makes it less attractive
No it wont. :stick: There will always be idiots recommending other products. So you'll always be needed to :slap: them around.Quote:
Originally Posted by MaxxxRacer
lol bloody. bill just needs to design a shroud for that radiator and they are done imo. well that and make it 14-15 FPI, if it isnt already.
brit, thanks for the kind workds :lol:
They should name it the H20-Cathar kit instead.
It has everything he thinks is best for a kit, even 7/16'' ID tubing.
you know Cathar is getting insane royalties for all of this... and if he's not we have some swiffy spies camping his favorite forums and stealing his ideas lol. BADQuote:
Originally Posted by RaptorRaider
No, I'm not. I first heard about the kit yesterday.Quote:
Originally Posted by moonlightcheese
Swiftech are just listening to what customers want, that's all. Why pay me anything. Anyone could throw together similar stuff and call it a kit. Doesn't worry me one bit - am just happy to see some manufacturers taking 7/16" ID tubing more seriously.
The radiator appears to be 12.5fpi from the pictures.
u serious cathar?!? 12.5fpi.. go bill.
moonlight, stew really isnt in this for the money. if he was he wouldnt be here on the forums helping us out so much. he would be stowed away in some lab designging all of these parts, outsourcing the production to the lowest bidder and then selling them at rather nice profit margins. Atleast thats what I would do if i was in it for the money.
And with watercooling, there's really no money to be in it for, so it's an easy choice. :p:
thats uplifting enjoy :(
dammit, just when i thought for sure that i was getting smart with this stuff, you throw out a new term... FPI? does that have anything to do with that frontal area thing i heard mentioned on the procooling forums? *sigh*
Sorry to burst your bubble Maxxx. I still love you. :)Quote:
Originally Posted by MaxxxRacer
FPI = fins per inch.
so how does swiftech's rad perform compared to others?
:doh: :(Quote:
Originally Posted by MaxxxRacer
sorry for the confusion/ assumption.
it's good to see people who want to bring together a better piece of equipment like that, making no profit but researching for the thirst of knowledge. very refreshing.
its very refressing monlight, and it is the reason that we all respect stew so much.
activate: no idea on the performance numbers. dont think anyone has tested it yet.
will 7/16" ID tubing perform better than 1/2" ID? or is it pretty much the same but 7/16" will prevent more leaking?
Same performance according to Cathar, better fit around 1/2" barbs.
It depends on the amount of tubing and velocity of water going through the tubing.Quote:
Originally Posted by ANDREW76
In an average setup, there is no real performance difference between 7/16'' ID and 1/2'' ID tubing.
At 1.5gpm with 6ft of tubing (average guess i made) here is the head loss
7/16" = .69768ft of head
1/2" = .36433ft of head.
so at 1.5gpm its not bad at all. so if yo u want to go with something a little more flexibible its not gonna hurt you
now at 2gpm with same tubing lenght
7/16" = 1.1879ft of head loss
1/2" = .62035 ft of head loss
now at 2.6 gpm with same length
7/16" = 1.93016ft of head loss
1/2" = 1.007938 ft of head loss.
so unless you have an iwaki, and ur pumping greater than 2gpm you should be fine with 7/16.
Is there a good place to find 7/16"?
USPlastics.com has it, but it's like $13 to ship 10 feet. lol
mcmaster. :D
Remember Maxxx to account for the tubing ID to fitting ID step-down transitional resistances that occur at entry points to blocks, pumps, radiators, etc. Once these are factored in the differences between 7/16" and 1/2"ID tubing closes right up.
cathar, that math right tere was merely from tubing and nothing else.
but i do see your point about the transitional resistance. this is why i think it is absolutely retarded hat companies use fittings that OD's fit perfectly inside of the tubing. this is also why i try to use 5/8 fitings wherever I can to reduce what you reffered to as step-down transitional resistance.
hmm.. time to do some pressure drop testing with fittings i think.
I know this is a rather odd thing to be picky about, but why is it that swifty always shows the example setup with about 1 foot of extra tubing compared to what it really could need. no sense in extra tubing i think.
Okay, the recommendation goes like this:
People should really be using 7/16" ID tubing with a 3/32" wall thickness for a 5/8" OD.
The physical benefits for 7/16" ID over 1/2" ID are these:
- 7/16" is more slender (5/8" OD as opposed to 3/4" OD). Less bulk equals better case air-flow.
- 7/16" is just as kink resistant as 1/2" ID (3/4" OD) tubing. It will turn just as tight radii, but consume less space when doing so because it is not as bulky
- 7/16" tubing takes about 2/3's the effort to bend as 1/2". This is important for waterblock mounting. Tubing that requires less effort to bend won't be applying as much torque against the waterblock mount against the CPU, resulting in better mount consistency, and therefore better temperatures.
- 7/16" tubing, when filled with water, is 2/3's the weight per unit of tubing length as 1/2" tubing. This means that there is less weight dragging off fittings. This is specifically important again for reducing the rotational torque that tries to tilt the waterblock away from the CPU, again helping to improve waterblock mounting.
- 7/16" tubing costs about 2/3's as much per unit length as 1/2" tubing.
Now people will want to know about how to use 7/16" ID tubing and the flow-rate penalties associated with it. The thing here is that it is quite easy to stretch 7/16" ID tubing over regular 1/2" OD hose fittings, and this is how it should be used. Do not attempt to source 7/16" OD hose fittings, it MUST be used by stretching it over your standard 1/2" OD fittings, and here's why.
This is where we get into the flow-rate effects of 7/16" ID tubing as opposed to 1/2" ID tubing. Aside from the flow-resistance effects of waterblocks, radiators and reservoirs, there are two main types of tubing based flow resistance in a water-cooling loop.
1. The resistance offered by the tubing itself
2. The resistance offered by the fitting interfaces where the ID of the tubing has to step-down to the ID of the hose barb.
Now 7/16" ID tubing is going to be more restrictive than 1/2" ID tubing naturally, however point 2) above is important to consider. When the ID's between the tubing and the host barb more closely match, then the flow resistance offered by the step-down in side is reduced. What this means is that 1/2" ID tubing will offer more flow resistance at barb fitting interfaces than 7/16" ID tubing. How much more? Let's explore that using a good pressure drop calculator.
First let's define what a typical water-cooling setup looks like. Let's assume 2.0m of total tubing length, and a cooling loop that has a radiator, a CPU waterblock, a pump and a reservoir. There is a total of 4 tubing-to-fitting step-down transitions in such a setup. 1/2" OD barbs will have an ID of 9.5-10.5mm, so let's assume a middling value of 10.0mm ID for the barb.
So using our calculator, and 26C water, we arrive at the following flow resistance values:
4LPM
2m x 1/2" tubing resistance = 0.09mH2O
4 x 1/2" ID tubing to 10mm ID fitting resistance = 0.12mH2O
2m x 7/16" tubing resistance = 0.18mH2O
4 x 7/16" ID tubing to 10mm ID fitting resistance = 0.07mH2O
Total 1/2" loop tubing/fitting resistance = 0.21mH2O
Total 7/16" loop tubing/fitting resistance = 0.25mH2O
6LPM
2m x 1/2" tubing resistance = 0.20mH2O
4 x 1/2" ID tubing to 10mm ID fitting resistance = 0.28mH2O
2m x 7/16" tubing resistance = 0.39mH2O
4 x 7/16" ID tubing to 10mm ID fitting resistance = 0.16mH2O
Total 1/2" loop tubing/fitting resistance = 0.48mH2O
Total 7/16" loop tubing/fitting resistance = 0.55mH2O
8LPM
2m x 1/2" tubing resistance = 0.33mH2O
4 x 1/2" ID tubing to 10mm ID fitting resistance = 0.49mH2O
2m x 7/16" tubing resistance = 0.68mH2O
4 x 7/16" ID tubing to 10mm ID fitting resistance = 0.28mH2O
Total 1/2" loop tubing/fitting resistance = 0.82mH2O
Total 7/16" loop tubing/fitting resistance = 0.96mH2O
10LPM
2m x 1/2" tubing resistance = 0.52mH2O
4 x 1/2" ID tubing to 10mm ID fitting resistance = 0.77mH2O
2m x 7/16" tubing resistance = 1.04mH2O
4 x 7/16" ID tubing to 10mm ID fitting resistance = 0.44mH2O
Total 1/2" loop tubing/fitting resistance = 1.29mH2O
Total 7/16" loop tubing/fitting resistance = 1.48mH2O
Now what we're looking at here is the resistances being offered by the tubing and fittings, and not the additional resistances being offered by the waterblock, radiator and reservoir. Let's now add on the resistance of a waterblock and radiator. We'll use a medium pressure drop like the White Water (reviewed by Bill Adams here), and the Thermochill 120.2 radiator (also reviewed by Bill Adams here). We'll assume that the pressure drop offered by the reservoir is neglibible.
Total System Resistance
1/2"
4LPM 0.82mH2O
6LPM 1.80mH2O
8LPM 3.10mH2O
10LPM 4.80mH2O
7/16"
4LPM 0.86mH2O
6LPM 1.87mH2O
8LPM 3.25mH2O
10LPM 5.00mH2O
Now flow-resistance is proportional to the flow rate squared. Therefore with the slightly higher flow resistances offered by the 7/16" ID tubing we can predict what the flow rate impact would be, which is the square root of the pressure ratios:
4 LPM => -2.5%
6 LPM => -2.0%
8 LPM => -2.5%
10 LPM => -2.0%
Now in all of that there has been some approximations by eyeballing graphs, but it's working out that there is about a 2.25% reduction in flow rates by going from 1/2" to 7/16" ID tubing. If we then factor in that centrifugal pumps actually push higher pressures as flow rates are reduced, then we can pretty safely assume that we'll see a net 2% reduction in flow rates.
So basically what we're seeing here is that we'll get 98% of the system flow rates with 7/16" ID tubing stretched over 1/2" OD barbs, as we would get with 1/2" ID tubing over 1/2" OD barbs.
In terms of an actual performance difference, we're talking about around a 0.02-0.05C performance loss at the CPU waterblock due to the decreased flow rates, but this is typically offset by the waterblock being able to make better contact against the CPU due to 7/16" ID tubing's benefits as noted at the start of this post.
In all of this, we can conclude that 7/16" ID tubing, for all intents and purposes, is the performance equal of 1/2" ID tubing, but offers a host of other benefits that makes it much more suited for a top-end performance rig than 1/2" ID tubing.
Oh really?Quote:
Originally Posted by MaxxxRacer
I can't find 7/16ID / 5/8" OD Tygon there.
http://www.mcmaster.com/ctlg/loadpag...n+R3603+Tubing
Only lists 3 tubings with 7/16" OD, none with 7/16" ID.
ill find it for you tonight. if there is a size they dont have I'll pass out.
I must say I am impressed by this kit, nice block, nice pump, nice everything, even the price is right. Hopefully some sites will review the kit shortly (against some euro kits) and let us europeans, see the errors of our ways :P
does anyone know how this block stacks up against the TDX series? From the little reading ive done, it looks to me that the storm would be the best possible choice for a CPU water block atm.
this is the tubing i use from mcmaster, its cheap and bends well enough for me.
part # 5233K44
The Swiftech Storm G4 is based off Cathar's Storm G4. So as it is, this is the best retail block available. The only better block out there, Cathar's Storm G5, isn't readily available and is made in limited batch runs and costs about 2.5 times more.Quote:
Originally Posted by Frambosie
Here ya go, Tygon 7/16.Quote:
Originally Posted by Psyche911
I haven't been following the redesigned Radiator discussions. How well should this new Swiftech radiator perform? Anyone have any guestimates as to where it falls in the performance lineup?Quote:
Originally Posted by MaxxxRacer
So what if you are actually using 5/8" o.d. barbs? Do you guys use 1/2" i.d. tubing for those? Also, anyone have any problems with tubing splitting when stretched that far? I am building up a system with all 5/8" o.d. barbs wherever possible, so for interconnects between two 5/8" barbs, should I pick up some 9/16" i.d. Tygon or just stretch the 1/2" over them? I figure if you are connecting to two 5/8" barbs, going up to 9/16" tubing will only help flow. Then when there is a 1/2" barb on one end use the 1/2" tubing. Sound good?
cart, if your going 5/8 barbs, why not go 5/8 tubing. its what i have on some of my stuff. its pretty darn big, but i dont use a case, so i dont really care.
but on some of the stuff i have 1/2" tubing going to 5/8" barbs becuase on one end of the tubing its 1/2" barbs and the other is 5/8".
Brit, no testing or discussion done outside of Swiftech's labs. but apparently there has been ALOT of testing of that radiator in the lab.
Yeah 5/8 tubing would work too, I just though 9/16 because the barbs on my mcp-650 are pretty tight with 1/2" so 9/16 would probably work fine on it. Plus the 9/16 is way cheaper than the 5/8 Tygon.
Isn't it about same FPI as BIP.... wouldn't be surprised if Swiftech outsourced HWLabs to do the job :P
more FPI = bigger surface area more restrcitve airflow.
less FPI = smaller surface area better airflow.
It's just balancing act for fan noise vs cooling. :)
That is an interesting point Stew that I had not considered. Being an RF/antenna eng. by trade I can see the direct analogy to transmission line impedance mismatch, discontinuities, and VSWR.Quote:
Originally Posted by Cathar
Jinu, I doubt this design was one of HWLabs, UNLESS it is one of the new XFlow radiators from them to be released soon.
and the BIP has the same FPI as the BIX. 18 I believe. Its In my CoolRad 12-T review on the front page.
I would consider 7/16 tubing, but Ive got 5/8" OD barbs on everything except the Fusion GPU block, plus it would be almost impossible to get the tubing on the 18mm barbs of the RD-30, that is the biggest issue for me, it is incredibly hard to get the 1/2" on there...
One thing i don't understand is the c/w rating
they say their's is H20-APEX 1 radiator x 2 120mm fans at 12 Volt 100 0.125 ~37
But i found this review on the xp90c 4870 rpm, 78 dBA² 25.5 C 19.2 C 6.3 0.11 21 C¹
Does the xp90c cool better than this wc setup?
You have 5/8" OD barbs on your G5? If so...pm me please. :slobber:Quote:
Originally Posted by nikhsub1
Yes, expensive and from taiwan... but they rule an OC forum member got me 4 of them... blurry pic but you get the idea.Quote:
Originally Posted by EnJoY
http://www.anonforums.com/builds/bla...barbsblock.jpg
C/W ratings are not comparable between different test-bed setups, at all.Quote:
Originally Posted by bigjohns97
To get an indication of how well the XP90C cools compared to this wc setup, you would have to see the XP90C tested on the exact same testbed as this wc setup was tested on.
Amazingly awesome, ygpm. :)Quote:
Originally Posted by nikhsub1
.11 c/w is impossibly low. Probably they screwed up somewhere.Quote:
Originally Posted by bigjohns97
niksub1, Im curious if u dont mind telling me. how much did you have to pay to get those those fittings out from taiwan. If you do mind, forget i asked.
Is it just me or it looks like the radiator will interfere with DVI cables or any cables that comes from the expansion slots?
They worked out to be like $6 each... with shipping. I mean, $24 isnt going to make anyone poor, but, $24 for 4 barbs is steep. I had to have them though. :D
Oh and they seal with an oring... they are bad ass.
The thread that started it with better pics: http://www.ocforums.com/showthread.php?t=369254
oh thatsn ot too bad.. i thought u meant like 20 a barb..
Can someone here answer my questions above?Quote:
Originally Posted by Sgt_Strider
It could block some ports, but it depends on how flexible ur cables are, how wide your case is, and how you mount the radiator.
What is this a Pro forums meet up :)
Great water cooling kit
I see, I guess I"ll have to wait until someone receives this kit. Good thing I didn't order the H20 120 yet..Quote:
Originally Posted by MaxxxRacer
According to someone on procooling, its offset enough to miss the VGA + DVI connectors. Still, it looks like it could be close on a lot of cases.Quote:
Originally Posted by Sgt_Strider
nice looking setup, perfect for those who don't want to build their own custom wc loop. It's nice to see some true performance packages hit retail market.
looks good . a well performing kit with convenients and a good price tag .. not bad at all
not to mention that they should be flexible enough.. i cant see it really being like a big problemQuote:
Originally Posted by saratoga
i think i'll be going with this, instead of custom building one from scratch. unless a custom setup is really that better, and if nothing else better comes out by december.
How loud do you think this system will be? I am about to install my first WC and this kit looks nice. The number 1 reason I want to WC is for a silent PC. Currently I have a p180 case, dfi lanparty sli-dr, 3700+, and 7800GTX SLI setup. Think this system along with a couple of DD's GPU blocks will help cool things down and keep my system extremely quiet?
Given the radiator specs and system's target croud, i would say that Swiftech provides a good performing radiator, rather than a true low-noise one...Quote:
Originally Posted by Teecee
If you seek silence, consider the Black Ice Pro III or wait for new X-Flow series from HWLabs. IMHO.
I really really really really want to see where this radiator is going fall compared to a BIP and BIX (somewhere between the two, I'm expecting)....
Hmmm, what does this radiator look like (Swiftech's old one)?Quote:
Originally Posted by Xylo
http://www.swiftnets.com/products/MCR120.asp
Then we have the new radiator here, with lots of performance charts here to compare to their old one above:
http://www.swiftnets.com/products/MCR120-QP.asp
Is the old MCR120 a BIP? That's the question that needs answering.
Yea I got the old swiftech radiator right here and its just a regular skinny BIP.Quote:
Originally Posted by Cathar
Wonder if Bill can answer this?
I might be able to provide some insight into this.
I am not sure on this so dont take it as the word or god or anything, but CoolingWorks might be building these radaitors.
Cathar, Im pretty sure the old MCR120 wasnt a BIP, but was a copy of one that was manufactured in an overseas facility to keep costs down.
I am fairly sure the MCR120 was a BIP, and I 1st thought the new rad looked like a modded HW design (different tanks and lower fpi) but looking closer it is not quite the same construction on any of it.
Here is a quote from BillA on Rads (the new rads could be the design he was reffering to.
Quote:
Willie sent me a prototype which I have not yet tested, but it has some problems
Willie and I were unable to agree on design 'rights', so I am still looking for a mfgr to make one for Swiftech the right way
(I will not gift a design to be sold to my competitors)
Has anyone found any reviews of this kit yet as i cannot
the original MCR120 was a re-named BIP (the later version)
the present MCRxxx-QP series are a Swiftech design produced under contract
other than the materials - copper, brass, and solder - there is little in common with the HW Labs products
the HW Labs rads are good products, one does not appreciate how good until one attempts to make a better rad
this has been a 9 month project to 'get it right'
Will there be a triple fan design version coming out soon?
Bill would you be willing to comment on why brass was choosen for the flowtubes and not copper, and if the HW labs radiators are using brass or copper flowtubes.
I had a copper tube/fin rad made to test, no difference compared to brass tube with copper fin
- understand that rads are convection rate limited (fin to air) so the tube matl is insignificant wrt overall dissipation
all commercial products with flat tubes or corrugated plates (heater cores) utilize yellow brass (70% copper)
I'm finishing a 'white paper' on rads that I presume Gabe will release next week (??) addressing all rad types and their specific performance
- the usual but ever better accuracy and repeatibility
I second this question. :rolleyes:Quote:
Originally Posted by EnJoY
http://www.frozencpu.com/ex-wat-76.html?id=AhstPrrU
How is gonna buy one and test it.
Anyone have a lower price than FrozenCPU? The kit price is about the same as buing all the parts individually. :(
I have a question, since the tubing used is 7/16, what if you want to use a GPU block? AFAIK, don't they come in 1/2 and 3/8? How can you pair one up with this kit?
The 7/16" tubing is meant for 1/2" fittings. It fits more tightly over them without too much effort. The difference in ID makes no detectable difference in flow.
Is there any review out yet.???????????
This look like a killer setup too me :stick:
What do you guys think???
Bill, didnt even think about that, but it does make sense.. heat conductivity between water and metal is much greater than air to metal so ur ending up with the limiting factor being the air.. interesting.
I wonder waht it would take for the metal to air transfer to catch up to the water to metal transfer in a radiator..
More, thinner fins. That would create increased air resistance though, so more powerful fans would be needed. More tubes would help too as there would be greater tube-fin contact.
There's no easy answer. It's all about finding the right balance point.Quote:
Originally Posted by MaxxxRacer
with ~6 interactive dependent variables, that is a understatement of some magnitudeQuote:
Originally Posted by Cathar
More tubes would also reduce the surface to air ratio though.Quote:
Originally Posted by cartmanea
What is interesting Bill is that you have I have gone for two different design paths (excluding the PA160.1 work which holds some similarities) on our recent efforts.Quote:
Originally Posted by BillA
Will be interesting to observe the evolutionary paths at the end of the day.
Still, with 6 co-dependent variables what would be most interesting would be to plot a 6-dimensional graph (good luck conceptualising it though), and then writing an analysis tool to plot of path of least resistance over the 6d surface. I suspect that we'd find a rather large "plateau" of acceptable solutions and perhaps you and I are just descending the 6d slopes onto different regions of that same plateau.
its worse than that
specifically the fin design; flat, slit, louvered, corrugated, dimpled. etc
lots of fin designs out there, now compound that with the possible fin thicknesses
and there is yet more of which I will not speak
but yes, this is addressed through parametric modeling (testing) and analysis
while there are numerous 'solutions' to a given task, I have found that very small changes in the wrong area have disproportionate decreases in performance
- if an item (liquid/air heat exchanger) is to excel, it must be close to spot on in the tube/fin area
If anyone isn't going to use the rad. from the kit for whatever reason, I'll buy it from you... :)
Why do you think its that much better than lets say a BIX2??Quote:
Originally Posted by Xylo
Noise.Quote:
Originally Posted by Ferry82
You need higher flow fans for a BIX over this radiator, it would appear. As far as performance, I don't think we know yet.
Well, I already ordered a blue one today.Quote:
Originally Posted by Ferry82
To answer your question: no, but a) cheaper and b) I can use weaker fans...but I think it's gonna be better than a BIP.
Well here is my short review of this kit. This is my first install so I might not know wtf I am talking about so take that into consideration.
First this is truly a complete kit. All you need to add is distilled water and have a screw driver. The kit comes very well packaged and you can tell it is quality stuff. It has 40" of 7/16 tubing which will be a snug fit on the 1/2 barbs but it works and should be long enough for most CPU setups. It also comes with some stuff that helps prevent kinks in your tubing. I am not sure if it works or not but I threw it on there anyways.
The instructions were pretty good. A little vague in some places and not well laid it. But this was my first WC kit and it is working with no leaks so the instructions are good, just could use some improving.
The two 120mm fans on the rad I do not like. They are loud. Well loud compared to a Nexus fan and you cannot monitor their RPMs via the MB. I swapped those out with some Nexus fans and now this system is completly silent. I have my pump on 5, which I believe is the highest settings and I cannot hear anything. The loudest thing in my case now is my dvdrom when it is seeking and the chipset fan which spins around 3000-4000rpms. As soon as a waterblock for the chipset comes out I will fix that.
It did take me 4-6 hours to do all of this but I am a noob at this plus I do not think the Antec P180 was the ideal water cooling case.
Would I buy this kit again or recommend this to a friend? Yes I would.
I am very happy with the performance and quietness. Brought my 3700+ @2.75ghz load to 40C.
Here are a couple pics.
http://www.jdover.com/images/wc13.jpg
http://www.jdover.com/images/wc09.jpg
http://www.jdover.com/images/wc03.jpg
http://www.jdover.com/images/wc22.jpg