Zalman Reserator 1 - Fanless Watercooling

[RoboTech]

Can you say direct die cooling, Reserator style...



This bad-boy may add new meaning to the term "tower" case !!!
(Sorry, I couldn't resist )

[RoboTech]

OK, seriously now...

I've been reading a lot of Reserator bashing and "opinions" in this thread so thought I would add my $.02 worth. I have been using a Reserator for three weeks.

XP-2400 @ 2,000 Mhz, 1.65 Vcore
Amb air = 24 C
Ave water = 33/37 C
CPU = 39/45 C
idle/load
CPU temp measured with a calibrated thermocouple on core

Open loop system flow ~ 28 GPH
with external E-1048 ~ 58 GPH

Of course this isn't a high-end, high-performance cooling system. Zalman specifically designed it as a fan-less water-cooling system to be silent.

"Let’s face it – a knowledgeable DIY type could put together a very quiet water-cooled system (or air-cooled for that matter that would provide as good or better performance for less money than the Reserator I – but it wouldn’t be silent. And you have to remember, the Zalman Reserator was not designed for, nor is it being marketed to the DIY enthusiast seeking a high-performance cooling solution."



This data was generated with a thermal die simulator...

IMHO the Reserator is a unique cooling solution - for those with modest cooling needs who want an easy to install kit that is silent and looks cool!

[Titan151]

A mobile CPU (say an XP 2600) puts out something like 45 watts (per newegg, sells em for $99). If you use that CPU and don't OC it will work much better then say with a standard XP2500 (puts out something in the range of 65 watts?) This give me an idea. By a big copper fanless heat sink and a mobile CPU and walla!! Problem solved. End of discussion. You get cool without getting ripped off!

And then when you come to your senses and are ready to go big, you can rip of the copper, hook up a pelt and 20 loud ass, high RPM fans and blow out your windows with a mean OC... Then you can use the gigantic copper thing for a paperweight. Ok so there is your solution and now this thread has reached the end!

[Groth]

RoboTech, is your graph for the stock Reserator 1, or with the external 1048?

[pHaestus]

RoboTech: C/W much lower on your die sim than I see on my test CPUs; perhaps die insulation isn't sufficient? That'd be consistent with attached graph too...

[spinky]

Originally Posted by gone_fishin

Naw, it's still junk for the money.

The itsy bitsy pump intakes at the bottom right next to the hot water entering the leaning tower of blue wasted space, sucking it right back through the loop. They did nothing to increase the water to metal contact area inside the before mentioned tower, except making it hideously tall (extend the same style fins to the interior next time anyway, but nobody could see how 'cool' that looks so don't get your hopes up).

believe me there are other things that are worst than the reserator for the money.
like buying a Geforce 2 Ultra for USD$399 when it came out, 6 months later the GF3 for USD$399... 6 months life cycle..when i GF4 Ti came out i bought it, played with it for 3 days and sold it before it devalued. least with the reserator, if all else fails you still have a resovoir that will never go outdated on you

Robotech > nice of you to join in, a user, that way the rest of them can stick their opinions up theirs; cause you own one and they don't

somebody correct me if i'm wrong.

die 52.5
avg. water temp 44.7
air 21.2
@ 100 watts
isn't that a C/W of 0.313?

[RoboTech]

Originally Posted by pHaestus

RoboTech: C/W much lower on your die sim than I see on my test CPUs; perhaps die insulation isn't sufficient? That'd be consistent with attached graph too...

Yes, the C/W values will be much different because my new sim has a larger die surface area... I went with this as I think it is more representative of current processors and what we will see in the future. The larger surface area is also more robust so should hold up better after lots of use.

"The thermal die consists of a copper block with a raised die area of 1,000 mm^2. (This is the approximate size of a P4 IHS - smaller than an Athlon 64 IHS but much larger than the exposed Athlon XP core.) Two Watlow cartridge heaters generate up to 300 watts of heat at 32 VDC. Waterblocks and heatsinks are mounted to a rigid phenolic top via threaded metal inserts arranged in Athlon XP, Athlon 64 and Pentium 4 mounting patterns. The die is well insulated with Cotronics ceramic fiber insulation."

Actaully the insulation is very good. "The thermal die simulator dissipates 1 watt for every 12.5ºC rise in temperature over ambient. At 100 watts input power this means that 2.2 W is being lost to the surrounding air and that 97.8 W is actually being absorbed by the waterblock." This statement is based on data from thermal loss calibration curves I ran on the sim.

[gone_fishin]

Originally Posted by spinky

believe me there are other things that are worst than the reserator for the money.
like buying a Geforce 2 Ultra for USD$399 when it came out, 6 months later the GF3 for USD$399... 6 months life cycle..when i GF4 Ti came out i bought it, played with it for 3 days and sold it before it devalued. least with the reserator, if all else fails you still have a resovoir that will never go outdated on you

Robotech > nice of you to join in, a user, that way the rest of them can stick their opinions up theirs; cause you own one and they don't

somebody correct me if i'm wrong.

die 52.5
avg. water temp 44.7
air 21.2
@ 100 watts
isn't that a C/W of 0.313?

Since when did one bad deal justify another?

I would argue that since Robotech went out and bought one, that his views may now be biased to defend the purchase.

[spinky]

Originally Posted by gone_fishin

Since when did one bad deal justify another?

I would argue that since Robotech went out and bought one, that his views may now be biased to defend the purchase.

cause a GF2Ultra is left on the shelf collecting dust & taking up space, therefore fullfilling the criteria set forth in the definition of "junk" whereas a reserator will still hold water and still function as a resovoir, thus its NOT a junk.

[RoboTech]

Originally Posted by spinky

somebody correct me if i'm wrong.

die 52.5
avg. water temp 44.7
air 21.2
@ 100 watts
isn't that a C/W of 0.313?

C/W is dT/W or (Die temp - water temp)/watts (look at pH's graph)
(52.5 - 44.7)/100 = 0.078 deg C per watt

C/W values are proportional to the area of contact between die and waterblock/HSF. Therefore C/W values can not be directly compared from one thermal die simulator to another unless they have the exact same surface area.

[RoboTech]

Originally Posted by gone_fishin

I would argue that since Robotech went out and bought one, that his views may now be biased to defend the purchase.

No, I try hard to be an un-biased reviewer... If it sucks, I'll say so!

[pHaestus]

Why do you show total Watts going in rather than watts going into the wb then? If you've done the correction then why not do it in the graph?

I don't believe using a large die is representative of modern CPUs at all (but we've had this discussion before). If it were a good proxy then mfgrs wouldn't be contemplating to switch to water cooling; they'd be using heatsinks with quiet fans and hovering in the mid50C range under load...

//edit extraneous comments removed

[bobkoure]

It would be interesting to have an external reservoir like this - looks like it might be able to shed enough heat that you could leave fans off when idling. Maybe hook the radiator fan voltage through one of those temp-based speed controllers (had a link to a schematic for one somewhere around there...) - and use water temp.
Anyone got connections in the world of aluminum extrusion? I did long ago (automatic door and supermarket storefront business) but my supplier of architectural aluminum has gone out of business (or changed name/location - not sure which). I would guess there's a usable cross-section shape already being made. Then it's just a matter of running through the appropriate anodizing path (type 3? As I remember, not used in the world or architectural aluminum extrusions...) speaking of which, the shapes we had to work with never had anodizing on the insides. I had assumed at the time that the anodizers were just saving money, but maybe there's a problem anodizing the hollow insides of long shapes, like you get with plating...?
Then there's the whole issue with having yet more aluminum in your system. It might be that the best external reservoir with idle-cooling capabilities might be a length of slant-fin radiator tubing.

[satanicoo]

So, after all, is this a good purchase or is it sufficient a large air cooler?

[HammerSandwich]

Originally Posted by gone_fishin

They did nothing to increase the water to metal contact area inside the before mentioned tower, except making it hideously tall (extend the same style fins to the interior next time anyway, but nobody could see how 'cool' that looks so don't get your hopes up).

I'll grant that the Reserator may have circulation problems, but is the hollow inside otherwise a disadvantage? Don't you need much more area on the air side because of the difference in specific heat between air and water?

[RoboTech]

Originally Posted by pHaestus

Why do you show total Watts going in rather than watts going into the wb then? If you've done the correction then why not do it in the graph?

Because I used data directly from testing and didn't adj for heat loss - as you point out, it could easily be done though...

A die sim is a good move, and low temperatures please mfgrs (see Bob Dyl). It'll make it problematic to do comparative wb testing though; I'd guess numbers about half as high with top end wb? Getting into "in socket thermistor" temp rise territory.

What low temperatures? (do you mean low C/W values) IMO the size of the die is pretty much irrelevant - having a larger copper surface that is not as succeptable to physical damage over time was more important to me than staying with "traditional" (Athlon XP) core size from years past. And what does B.D. have to do with anything???

bobkoure,

I don't think a big passive reservoir would help much in a standard water-cooling system. In most performance systems the ave water temp is only a few deg over ambient air (hopefully) so there is little dT driver to support passive cooling. The Reserator operates at a much higher ave water temp (8~12 C) which creates a more favorable dT but also causes the CPU to run hotter.

[spinky]

0.08 C/W for the reserator? is these figures comparable with other C/W figures?
like these -> http://www.overclockers.com/articles373/waterkit.asp

if the reserator is indeed 0.08 C/W it even beats the corsair... uh.
/me confused

[bobkoure]

Originally Posted by satanicoo

...is this a good purchase...?

Depends on whether fanless cooling is one of your priorities, doesn't it?
BTW, Innovatek makes a vaguely similar looking radiator

which I would guess, given the additional surface area (both water and air) would be more effective than the reserator. However, it does not include a pump or waterblock.
There's also a couple of attach-to-the-outside-of-your-case radiators. Again, state-side at least, see highspeed PC. And if I haven't made this clear, my only connection with these folks is as a customer - not trying to "shill" anything, but they seem to be the only US Innovatek supplier.
I assume that there are other European silent-cooling solutions that I don't know about, but they don't seem to have made it to the states (and I don't see comments about 'em in the Silent PC Review - watercooling forum

[RoboTech]

Originally Posted by HammerSandwich

Don't you need much more area on the air side because of the difference in specific heat between air and water?

Yes... Not to mention the difference in convective cooling between a liquid/metal interface and a meta/air interface! FWIW, the return water gets directed straight up into the main body of the Reserator, which produces good mixing (you can actually see the water surface swirl slightly) so there is no "short circuit" of warm return water being sucked right into the pump.

[bobkoure]

Originally Posted by RoboTech

I don't think a big passive reservoir would help much in a standard water-cooling system...The Reserator operates at a much higher ave water temp (8~12 C)

Duh (blush) - should have realized that. I plan on higher temps when putting together a quietly-fan-cooled system. Just wasn't thinking - sorry

[pHaestus]

I mean that in my test system you'd probably see die temperature in the 60+C range using a TBredB processor. Mfgrs don't really care for that much; they'd rather fall at the back of the pack at 40C than midrange at 65C. //edit extraneous comments removed.....

[RoboTech]

Originally Posted by spinky

0.08 C/W for the reserator? is these figures comparable with other C/W figures?
like these -> http://www.overclockers.com/articles373/waterkit.asp

if the reserator is indeed 0.08 C/W it even beats the corsair... uh.
/me confused

NO...!!!...
The reason I posted the graph was to primarily illustarte how the Reserator water temps responded to input power. Yes, you "can" calculate C/W values from the data presented but that was not my intent. C/W values will vary greatly from thermal die sim to sim mainly do to different die surface areas, insulation and mounting techniques. Both Joe (Overclockers.com), pH and I believe BillA use a much smaller die size (modelled on the Athlon cores of several years ago).

So NO, the C/W values are NOT comparable and no, the Reserator does not perform better than the HydroCool 200 (~4 deg C lower on XP-2400 in Whisper mode).

[satanicoo]

There are fanless heatsinks.
Probably they dont work as good as this unit, but since these are aimed at the non-overclockers, it should be an at-least better transportable system.

[gone_fishin]

Originally Posted by RoboTech

Yes... Not to mention the difference in convective cooling between a liquid/metal interface and a meta/air interface! FWIW, the return water gets directed straight up into the main body of the Reserator, which produces good mixing (you can actually see the water surface swirl slightly) so there is no "short circuit" of warm return water being sucked right into the pump.

NO....... you do not like water to metal contact ratio increased?
And the suction ( which is right next to return water inlet) does not siphon any of this straight off? No matter, it just causes it to reach it's HOT equalibrium point faster.

You are even defending it's most obvious shortcomings.

Someone explain to me... why a $200+ cooling system, that doesn't cool any better than stock, is not junk because you can still use the resevoir? That is like saying your new expensive color television lost it's color but you can still use it for a lampstand.

[pHaestus]

I should clarify something. RoboTech I wasn't trying to insinuate you were dishonest when I made a comment about mfgrs loving low temperatures. Joe and I have had some issues in the past where companies were quite upset by how high temperatures we reported were. //edit extraneous comments removed