New radiators: airplex modularity system

[Waterlogged]

Mind confirming what the tubes are actually made of?
Also, will be interesting to see performance numbers, flat tubes tend to perform considerably better if I'm not mistaken.

I was starting to think I was the only one to have "issues" with tube design. At best, I see these matching what we have now.

They do look great though...especially the all copper version.

[CedricFP]

I thought the whole point of these was modularity rather than performance?

And, unless I am mistaken, it seems this kind of modularity/flexibility really only pays for itself when you're constantly switching rad configurations.

For example, if you were a tester or just kept rebuilding your loop with different cases etc.

But it does look extremely nice. That all copper version... man, where do I insert my...

[mav2000]

Looking good and seems to be a lot for the price if I may add. But performance is key, so will wait for some performance figures.

[RacingTurtle]

Loving the innovation.
Typical these sorts of things always get released just after I complete a build!

[Church]

Hmm, i'd prefer flattened tube waterpaths to round ones aswell, but otherwise VERY interesting rad/module line. Especially interesting to me seems that integrated module with flow sensor/fan controller and so on .. Shoggy: can you elaborate on it's details?
All-copper rad core also might have extra win (I recall magicool's plexi rads .. while they were notorious for cracking of their plexi ends, pure cooling performance wise they outperformed common painted rads like RX & PA while being thinner then them, both at quiet and medium fan rpm ranges).

CedricFP: imho modularity might help for when moving LC gear to other case, where because of size you might wish to add or remove some rad section.

[defect9]

I'm wondering if the cooling pipes are just boring round pipes, or if there's anything inside to effectively increase surface area, like fins or something?

[Serpentarius]

anyone spot louvered fins in there?

[Aquatuning]

Nice one AC

[st0ned]

Nice one AC

Now work work to put them around europe

[Defoler]

These looks really amazing

But I have to point out maybe a small design flaw.

People tend to use a big longer screws and tend to push them too much in.
On one side, your tubes have a long way from the screws holes, so even if someone screw too much, it will not puncture anything.
But on the other side, he might.

On the GTX radiators for example, there is a small space, so even if you tighten the screw too much, it will not puncture any tubing.

This is just something to point out to, since these radiators are not "oops" proof.

But the copper radiators... lovely!

[MegadetHCl]

These radiators remind me of Lytron heat exchangers that use stainless steel tubing and copper fins. I've seen the Lytron units mentioned in laser cooling, and they are definetely not cheap.

[[XC] Synthetickiller]

Looks like all these high RPM fans will come in handy when stacking rads.

Can't wait to see reviews of noise and temps w/ 1 rad vs stacking 2.

[antiacid]

I like the industrial look to it. I'm eager to read about the performance numbers, both for single and stacked solutions.

[CedricFP]

Presumably the stacked rads will be run in parallell so you won't run into the performance issues the in-series MCR stack rads had?

[Church]

CedricFP: poor performance of stacked rads doesn't come as much from doubled flow restriction (double of little is not much either), but rather from increased airflow restriction and cooling 2nd rad with preheated air. Or under parallel you meant uniting them side by side not stacked one on top of another?

[CedricFP]

CedricFP: poor performance of stacked rads doesn't come as much from doubled flow restriction (double of little is not much either), but rather from increased airflow restriction and cooling 2nd rad with preheated air. Or under parallel you meant uniting them side by side not stacked one on top of another?

No... what i meant was the swiftech rads stacked and water flowed through them in series. That means they flowed through rad 1, were cooled substantially, and then flowed through rad 2 which receives hot air and therefore they are not cooled substantially anymore. Extreme diminishing return slope.

If you have flow going into 2 stacked radiators as paralell, it means that both rads receive heated water. You still run into diminishing returns, but not to the same extent.

Think of it as this - stacking 2 radiators and then having water flow through them in parallell can be compared to having one really damn thick radiator.

At least, I think that's the case. In my mind it is.

[chatch15117]

I'd like to see someone have a 1800mm rad mounted up the wall LOL

[Utnorris]

I am just curious if this will fit in a case.

[Church]

Utnorris: that .. or "will it blend"?
CedricFP: don't mix water flow restriction and air one. As i wrote above, rads usually are not very resistive relative to waterblocks, so waterflow connection in serial or in parallel shouldn't impact overall flow in loop, also only the very biggest rads may see noticeable temp drop after one pass through one rad, but most probably temps won't drop even one degree. It's not like water immediately heats up in block, and immediately cools down in rad.
It's different story with air, as stacked one onto other rad impacts a lot air flow, especially with quiet fans, so stacked rads benefit more from powerful fans with high air pressure to counter that (just like high FPI rads) and air's volumetric heat capacity is much lower then that of water's, so air heats up much quicker and for several degrees, so again to counter that you need more powerful (and more noisy) fans to blow air quick enough to not heat up too much in 1st rad to significantly impact 2nd rad performance in airflow path.
Thus doesn't matter much if two rads with separate airflow are connected serially or in parallel (they might even cool a bit worse in parallel because of halved water flow speed) waterflow-wise, but it matters a lot, if rads have separate airflow (eg. side by side) or are stacked one onto other, and even more so if fans used are quiet and slow (thus with little cfm and air pressure).

[CedricFP]

Churchy: I am not talking about water flow restriction....

...all I am saying is that running 2 rads stacked with parallel flow (in to both, out from both simultaneously) is more similar to having a very thick radiator than the Swiftech stacked rads ever were.

Airflow is certainly a concern. You are right that since temperatures tend to equalise in a loop, airflow becomes the largest limiting factor in stacked rad performance. Assuming you use the same RPM fans, would this method of parallel flow stacking not be beneficial over Swiftechs method?

[Church]

You don't get it. Your described rads side by side with separate airflows for each will act/perform/cool just like separate rads placed anywhere else. Only difference - they will be connected by those end modules instead of tubing. No need to worry about such scenario performance wise, only about mount possibilities/specifics, afterall, rad package side by side gets very wide.
Stacking drawbacks arise when you .. well, stack one onto other on top, with shared airflow for both rads, not with separate airflow with rads side by side.

[CedricFP]

You don't get it. Your described rads side by side with separate airflows for each will act/perform/cool just like separate rads placed anywhere else. Only difference - they will be connected by those end modules instead of tubing. No need to worry about such scenario performance wise, only about mount possibilities/specifics, afterall, rad package side by side gets very wide.
Stacking drawbacks arise when you .. well, stack one onto other on top, with shared airflow for both rads, not with separate airflow with rads side by side.

What? I'm not talking about side by side rads.



There's an image showing two rads stacked on top of eachother with a piece on top that provides what i assume is paralell flow to both rads. Just look at what Shoggy wrote:

The flow path in horizontal setups is in series while it is parallel for vertical setups.

That's why I said its more similar to a thicker single rad than the method Swiftech used...

I don't call "airflow" "flow". Flow in my posts refers only to water.

Would this setup not perform better than Swiftech's implementation? Granted, good fans will be needed. Push pull preferred.

EDIT: Perhaps not, though, as I just looked at some Swiftech stack images and it appears you could configure the flow to be parallel between both those rads as well.

[stoner]

i love the idea... the delrin part with+flowmeter+sensor it's great

[skinnee]

Holy buckets, thats a lot of testing...


Yes, this thread brought me out of my self-demanded mental vacation.

[Crankyhobo]

We love you though. Pretty please!

Holy buckets, thats a lot of testing...


Yes, this thread brought me out of my self-demanded mental vacation.