Martinm210

Pretty much there, just need to do some CFM measurements.

I updated my page with everything here:

http://www.martinsliquidlab.com/Ther..._3-Review.html



And for a different way to look at it. I'm still a bit fuzzy what to really use for actual heat loads in systems, but here is a comparison of fans on some assumed values. I suspect a good 10-20% or so is lost through heat on the PCB and in the air, also GPU only type blocks would have even less as the memory is air cooled, but it's a start if the chart above is hard to follow:



And I've attached a really quick estimator spreadsheet so you can plug your own heat value in there along with the air in and it'll return an estimated water temperature.

Any suggestions for format, I plan to do more with other subsequent radiators where I'll have something to compare at that point, but this is the first one...?

I plan to later measure actual CFM for the fan scenarios and maybe explore some shroud effects, but this was enough for a weekend's worth of work..

Attached Files

PA120_3EstimatorWeb.zip (7.5 KB, 61 views)

__________________
Flow Rate Estimator | Radiator Sizing Estimators

ira-k

Those Ultra Kaze brought it to life didn't they.......What are they 133cfm? If I remember right Bill A. could only get a 10C rad out to ambient with 103cfm Pana and a 6LPM flow at around 600W...So thats a good bump from a little higher cfm fan, they must have a better pressure...

Like you were saying in the other thread the fans have just about as much to do with good temp's as any of the rest of it, you have to match them to your heat load and rad style...It used to be more centered on flow with the older blocks, most of the newer blocks do fine with a 1.5gpm or so flow..Shrouds are about the cheapest few C that can be had on higher fin per inch or thicker rads to..

Looks really good Martin, you've done a fine job and I'm sure it will only get better with a little time..I'm glad you think its fun.....

__________________
...***Beware all ye Men born of Woman, beware the Hemogoblins***...
**E8600 @ 5751Mhz ***OCN Water Cooling Club And Picture Gallery ***150 Opty @ 3242Mhz **

Martinm210

Yeah 136cfm (open air) or so, I'm hoping a graph like this will help show the fan affects a little better. I don't know how many times I heard a lengthy discussion about which radiator and the fan was never mentioned, yet there's 3X the performance difference depending on the fan.

Although I don't know how you put up with those 100+ CFM fans, I thought my testing bench was going to take off!!.. Turning those down to 2000 RPM was much nicer on sound and still perform pretty well. I was thinking of trying them at 1000 RPM also, but for some reason they started loosing RPM sensing ability at around 1400.

I'll also have to try some shroud experiements too, just wanted to get this out of the way for some others.. I've got a few blocks and pumps to do too, but one radiator down..

__________________
Flow Rate Estimator | Radiator Sizing Estimators

DaCrusader

Woh! thats amazing.... are they the dual motor fans? What was ambient?

__________________

Martinm210

No, just the regular Scythe 120 x 38mm fans, 3000RPM, 136cfm. I picked these for my high speed because I'm an old Scythe "Fan".. I had a Scythe infinity before water cooling infected me..

These ones:
http://www.ncix.com/products/index.p...facture=Scythe

This took several days worth of testing, and logging temperatures over half an hour periods of time, so ambient was changing all the time. I measure it with 6 air in sensors and logged it every second. So 6 full tests at 1/2hour x 6 sensor, so which of the 100,000+ readings would you like..

Here is just one of 12 tests run to complete that chart:


In this particular one it was in the 26 to 25C range, probably after the woodstove was melting the place. My house fluctuates alot burning wood, that's part of the reason I'm so particular about all the sensors and logging.

__________________
Flow Rate Estimator | Radiator Sizing Estimators

Sistum Id

[QUOTE=Martinm210;3743485]No, just the regular Scythe 120 x 38mm fans, 3000RPM, 136cfm. I picked these for my high speed because I'm an old Scythe "Fan".. I had a Scythe infinity before water cooling infected me..

These ones:
http://www.ncix.com/products/index.p...facture=Scythe [QUOTE]

Nice info

How loud are those Scythe @ 3k RPM? They are rated at 45dba but how would you rate the sound levels for 3 of them @ 3k RPM. A whine, hum, whoosh.

__________________
FYATT

ira-k

I know what you mean, I've seen guys running little low cfm 120x25mm fans on a BIX and complaining because its not working well......Kind of the same thing with the PA's, guys will have it loaded down and running little fans and wondering why its not cooling well, they think they can get that 500W with them and they cant...They just work OK with low cfm fans because they have a low fin per inch count....

The 115's and 108's don't really bother me, its just a low hum kind of noise, almost a white noise after you get used to them, I have to stop and actually listen for them sometimes...

I know how you feel though, a friend of mine runs those 190 or 200cfm Deltas and I cant stand them, I get an almost instant head-ache, just something about the tone of them I guess....

What about the 86 Pana Martin, do those bother you? Thats about as small as I would really want to go personally on any rad....

Man you did great for your first whack at this......You'll figure out more tweaks as you go along I'm sure, take your time and get as much fun out of it as you can....I love this thermal testing....

__________________
...***Beware all ye Men born of Woman, beware the Hemogoblins***...
**E8600 @ 5751Mhz ***OCN Water Cooling Club And Picture Gallery ***150 Opty @ 3242Mhz **

Martinm210

Yeah, I could see using the fans in a completely enclosed radiator box, something with some sound deadening material and inlet/outlet ports with a house filter on them or something to quiet them down.

They really were not as bad as I thought they would be, I'm just used to the slow speed yates in my system which hides down under the desk, so when I had these going up on the test bench it was louder than I'd like. The 86cfm panas are also a great range, if I actually had a heat load, I'd probably be using them. But considering I probably only have 125watts in my CPU loop, that's less than a degree difference.

And the CrystalFontz was perfect for this. It's amazing how sensetive the rig was to changes in the environment. Even though I don't have any heater vents anywhere in my little shoebox of an office, I could see a little blip ever time the furnace in the house kicked on. After figuring that out, I shut off the furnace. It was really nice to watch the results plotted over a 30 minute period of time, it was really easy to spot any errors that way.

I learned alot on this first test and feel pretty good about my testing method now. Ready for more. I'm going to take a little break and wrap up a few block reviews I have on hand, but then I'm putting the 480 on the bench....

__________________
Flow Rate Estimator | Radiator Sizing Estimators

DavidNJ

I've been listening with a stethoscope around the fans, and it seems a lot of the noise emanates from the side of the housing. I'm thinking a sound deadening wrap around the fans and shroud should really quiet it down.

To make the shroud you mentioned to Ira, I just use these spacers, long screws, and duct tape:

Spart

Thermochill For Life!

__________________