You are awesome.Quote:
Originally Posted by Martinm210
Printable View
You are awesome.Quote:
Originally Posted by Martinm210
Martin: imho AP-29s loose more (at least pressure, if not flow) from having less blades, then from having strengthening ring.
That's true, forgot about the fewer numbers of blades. They do pretty well on noise/CFM though despite having the blade change. I did a few videos of them a week or so ago, but didn't feel the test was publish worthy with my anemometer going flaky on me lately. I think I may have dust or something on the hot wire causing inconsistent readings. That and I'm afraid of opening the fans door again as I'd probably be flooded with more fan testing requests while I'm cleaning out my testing desk...:)Quote:
Originally Posted by churchy
Alphacool Xtreme III R2 added..
Hi Martin, thanks for all the effort and time you've put in for the testing! Will you be testing the EX360? I ask because you've tested the RS and RX, wouldn't be nice to leave out one of their brethren now would it? =P
Added the UT60 results since it did so well, but still working on the review details. I hope to get most of these done in the next few weeks as my last contribution for the summer, but I am burning out pretty fast...it was kind of fun for about the first four...now it has become WORK!...:eek:..:D
UT60 review is done:
Alphacool NexXxos UT60 360
More to come...
Interesting - is this the only full copper rad so far?
Do you have one of the AMS full copper rads to test?
It's greatly appreciated, though the watt dissipation seems so low. I'm kind of curious as to the physics of the slim vs thick rads. I'm not sure if you are familiar with the Corsair etc closed loop CPU coolers, but the double thick radiators greatly outperform the slim radiators. The H80 double thick 120x1 is actually a close performer to the H100 120x2 slim model. Both of those ship with the same fans and pump, though the H80 is default push/pullQuote:
Originally Posted by Martinm210
HAHA wow I bought a UT60 120mm version due to cost LOL looks like that was a good move haha.
thanks for this great shootout, amazing job Martin.
that UT60 is a killer Rad, it time to change my good old gtx360's !!!
I wonder if the UT60 preforms just as well with higher RPM fans like Delta's? Traditionally once you hit +2000RPM the GTX series begin to shine.Quote:
Originally Posted by quattro_
I think you can surmise that the rads with the straight line graphs have a good chance of continuing that line into the upper fan speed region
Wait until more watts and higher rpm fans are used. I willing to bet the GTX will be one of the top performers. Honestly though, not many people are going to be using fans with > than 2600 rpms.
True, but we cannot ignore the fact that some radiators preform better with lower RPM fans while others are the complete opposite. Note the GTX360's slope in comparison to the rest.Quote:
Originally Posted by OldChap
http://skinneelabs.com/assets/images...atdiss-5dT.jpg
Agreed. So far, I would say that for fan speeds over 2750, the likely order would be GTX360, MCR 320-XP, ut60.
With some fans capable of over 5000 rpm I see though that Martin has to draw the line somewhere (sorry, couldn't resist)
Yeah..I think the retest with higher speed and higher pressure fans will favor the gtx. At some point that double thickness high density will prevail. I just don't think it shows as well with these titans and with this fixed pumping power method. The titans are more average in cfm/rpm where skinnee was testing with GT15s whichmakes nearly a 400 rpm change to the rpm scale at the right. In addition this fixed pumping power method more simulates actual gains and losses from restriction. The GTX is higher in restriction than average so it is testing at a slightly lower flow rate than the lower restriction radiators. Skinnee was testing with fixed flow rate which ignores restriction which is how I did it back on my V1 bench, but that gives a slightly unfair advantage to the more restrictive radiators.
We will see in the folllow up high speed fan test. I have push plus pull Deltas and San Aces at the ready for that retest on the high speed leaders as I am curious as well. I am also looking to do some alternate inlet port type work at the end. some rads have odd numbers of tubes and some rads like the gtx have front to back core flow and there has always been questions about that but never tested. You would think the manufacturers would provide instructions if it did matter, but none do..we will see.
Higher rpms/higher pressure probably is easiest to replace with push-pull, thus reducing rpm point at which gtx's extra fin area starts to play in.
Iv been meaning to ask this question for a long time :Quote:
Originally Posted by churchy
what rpm fans in push will give the same performance as a gtx360 with 1000rpm fans in push/pull ?
Sent from my GT-N7000 using Tapatalk 2
I may be totally of... but is it possible that your testing rig is skewing the higher rpm results? It looks abit restrictive, and comparing to Skinnee`s tests the wattage dissapated is much lower, indicating this? The fans used vs GTs isnt THAT much worse...
I think the bigger difference is the air in sensor location, but I did check that restriction question. I tried one test on the RS360 with the top open and got results that were within a few percent, so there is some restriction, but nothing too major. Also in a case you will get some restriction, so I think it's more accurate than open air testing.Quote:
Originally Posted by eXa
But testing with the sensors near the radiators will unfortunately get radiant heat from the radiator as it warms up making the net water/air delta much lower because the air in sensors are heated up as well.
You can see that in my old V1 test where I was attempting to make a quad rad go passive. That red line is my air in sensors where the blue is the actual ambient.
http://martinsliquidlab.files.wordpr...led1.png?w=614
My Air IN sensors on my V2 bench are located 6" away and in front of the fans pushing, so there is no radiant heat error like V1 or others that are measuring right at the radiator. That and I'm sure there are losses in the tubing and other parts when you don't insulate it. I have at least one retest where I was able to get a 4-5% change simply by turning the ceiling fan on in the room and getting more circulation around the test setup. These are all errors of coarse and why I specifically built the test rig enclosed and with the air in sensors far far away..:)
Also as I noticed in picking up radiant heat in my V1 bench. If you are picking up radiant heat on your air in sensors, the more dense the radiator the more radiant heat = the more skewed the result so the error will vary depending on the radiator.
I also forgot to mention the difference in flow rate testing methods. Fixed flow rate testing will favor the more restrictive radiators slightly because they won't suffer any consequences of the higher restriction. That is how Skinnee was doing it and how I did it in my V1 bench.
In the V2 bench, I have switched to a fixed pumping power method where I have a fixed amount of test bed restriction and pump power in the 1.5GPM "neighborhood". However because each radiator has different amounts of restriction the lower restriction radiators will net a slightly high flow rate than those with more restriction as you would actually get in a real loop. I'm guessing the flow rate may vary from about 1.3GPM to 1.7GPM depending on the rad restriction and I think that better reflects actual conditions better than fixed flow testing.
FYI, added the Phobya G-Changer in the mix..
Aha, never mind me then :p:
Excellent work as always :up:
Just sitting here chuckling to myself at the the ppl going "ZOMG, got to change rad!". :hehe:
Keep up the good work Martin. :up:
http://img399.imageshack.us/img399/3017/popcorn2wf2.gif
great review!! cant wait to see how the xspc ex radiator performs
Got #10...the UT45 up and moving on to the EK rads. I did find it interesting that the copper tubes seemed to make at least some improvement.
Results are really tight on the slow speed stuff though, many are well within testing error.