Swiftech Storm Review @ OC.com
 

http://www.overclockers.com/articles1261/

Something's terribly wrong though. The pressure drop is drastically high according to Joe.

According to Swiftech's own data here:

http://www.swiftnets.com/assets/imag...%20vs%20fr.gif

The pressure drop at 1gpm is ~1.65PSI.

Robotech @ SystemCooling.com reviewed the block and measured the pressure drop to be as follows:

http://www.systemcooling.com/images/...image21big.gif

or around 1.5PSI @ 1gpm.

My own data puts the pressure-drop of the Swiftech Storm at ~1.53PSI @ 1gpm.

Robotech and I were measuring with 1/2" OD barbs. I suspect that Swiftech's values are perhaps taken with 3/8" OD barbs.

Joe's data puts the block's pressure drop at a staggering 2.74PSI @ 1gpm, and WAY out of line with 3 other measurements.

Was JoeC's Storm block heavily clogged during his testing? His results for it were a little disappointing to me, even though it handily took the lead, and I strongly suspect that cloggage may have something to do with that?

BTW - I had already emailed JoeC, but wanted to raise this here too as I think it's a valid concern.

Agree.
Looks more like a full system pressure drop

http://www.jr001b4751.pwp.blueyonder.co.uk/storm.jpg

Dunno.
Using MCW6002 as reference, there maybe no problems:

http://www.jr001b4751.pwp.blueyonder.co.uk/storm1.jpg

If so, then that is not consistent with other results reported by JoeC. Take, for example, the Maze4's reported pressure drop at 1gpm, which is 0.12PSI @ 1gpm. If the Maze4's results were that of a full system, then the full system would have to have a pressure drop of no more than 0.12PSI, and hence this would not account for the difference between other measured pressure drops and JoeC's measured pressure drops of the Swiftech STORM.


I can tell you right now that none of Swiftech's reported results are done with the 100mm^2 test die that BillA used to use (totally different measurement mechanism). It has been shown that the differences get larger as the dies get smaller.

I'm not saying that JoeC's thermal measurements are WAY out of whack, just perhaps up to 0.01C/W out, which could be accounted for by way of clogging causing the increased pressure drop.

Was solely indicating that JoeC's 2.74psi should be about right for the an Apex system @1gpm


The 100 sq mm is based on Bills's post. Noted the careful wording and so probed here. Cannot do much more to ensure information is correct.

Be that as it may, the results published at Swiftech are not achieved on a 100mm² test die. This is revealed in the wording at Swiftech's site which states the results are achieved using an industry standard thermal validation mechanism. I would be very much surprised if BillA's old test-die was categorised as such.

Uses an "industry standard heat die"(link) Failed to find such an animal at JEDEC perhaps most pertinent was this