• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

push/pull rad, worth it?

Joined
Mar 24, 2010
Messages
5,047 (0.99/day)
Location
Iberian Peninsula
Hi!
I have a triple rad and I wonder if adding 3 more fans below to it would make a difference.
Somewhere I read that pump speed is much, but very much more important than rad fans speed or quantity.
I have spare fans so it's not a cost problem, but it makes my build more complex.

I'd also appreciate recommendations regarding the fans powering. Should I just join 6 fans into one onboard fan header? Or better not? Or use 2? At the moment, I have 3 in 1 cables.

What do you think?
 

Durvelle27

Moderator
Staff member
Joined
Jul 10, 2012
Messages
6,678 (1.56/day)
Location
Memphis, TN
System Name Black Prometheus
Processor |AMD Ryzen 7 1700X
Motherboard ASRock B550M Pro4|MSI X370 Gaming PLUS
Cooling Thermalright PA120 SE | AMD Stock Cooler
Memory G.Skill 64GB(2x32GB) 3200MHz | 32GB(4x8GB) DDR4
Video Card(s) |AMD R9 290
Storage Sandisk X300 512GB + WD Black 6TB+WD Black 6TB
Display(s) LG Nanocell85 49" 4K 120Hz + ACER AOPEN 34" 3440x1440 144Hz
Case DeepCool Matrexx 55 V3 w/ 6x120mm Intake + 3x120mm Exhaust
Audio Device(s) LG Dolby Atmos 5.1
Power Supply Corsair RMX850 Fully Modular| EVGA 750W G2
Mouse Logitech Trackman
Keyboard Logitech K350
Software Windows 10 EDU x64
Yep


Edit: It’s safer to give each fan individual dedicated power.
 
Last edited:
Joined
Mar 24, 2010
Messages
5,047 (0.99/day)
Location
Iberian Peninsula
spaguetti cabling incoming! .... 2x 3x1 would be ok, still

but main question is if putting 6 fans on a rad will be a good idea
 
Joined
Oct 17, 2012
Messages
9,781 (2.34/day)
Location
Massachusetts
System Name Americas cure is the death of Social Justice & Political Correctness
Processor i7-11700K
Motherboard Asrock Z590 Extreme wifi 6E
Cooling Noctua NH-U12A
Memory 32GB Corsair RGB fancy boi 5000
Video Card(s) RTX 3090 Reference
Storage Samsung 970 Evo 1Tb + Samsung 970 Evo 500Gb
Display(s) Dell - 27" LED QHD G-SYNC x2
Case Fractal Design Meshify-C
Audio Device(s) on board
Power Supply Seasonic Focus+ Gold 1000 Watt
Mouse Logitech G502 spectrum
Keyboard AZIO MGK-1 RGB (Kaith Blue)
Software Win 10 Professional 64 bit
Benchmark Scores the MLGeesiest
I've never had a noticeable or justifiable difference in temperature from adding fans to the other side. I just leave it to pull or push but never both. I think people do it ,just do it more often than not ( and keep in mind I don't consider 3 to 5° worthwhile).

Purchase good fans & you won't need to do it. Unless of course you're already dead set on doing it. IMO, I think a lot of it is nit picking & placebo effect
 
Last edited:
Joined
Jul 2, 2008
Messages
8,068 (1.40/day)
Location
Hillsboro, OR
System Name Main/DC
Processor i7-3770K/i7-2600K
Motherboard MSI Z77A-GD55/GA-P67A-UD4-B3
Cooling Phanteks PH-TC14CS/H80
Memory Crucial Ballistix Sport 16GB (2 x 8GB) LP /4GB Kingston DDR3 1600
Video Card(s) Asus GTX 660 Ti/MSI HD7770
Storage Crucial MX100 256GB/120GB Samsung 830 & Seagate 2TB(died)
Display(s) Asus 24' LED/Samsung SyncMaster B1940
Case P100/Antec P280 It's huge!
Audio Device(s) on board
Power Supply SeaSonic SS-660XP2/Seasonic SS-760XP2
Software Win 7 Home Premiun 64 Bit

dorsetknob

"YOUR RMA REQUEST IS CON-REFUSED"
Joined
Mar 17, 2005
Messages
9,105 (1.31/day)
Location
Dorset where else eh? >>> Thats ENGLAND<<<
Buy dedicated fan controller to use
if you push/pull with 3+3 fans while the Fan header MAY handle the power draw its Safer in the long run to use a dedicated Fan controller
 

manofthem

WCG-TPU Team All-Star!
Joined
Jan 9, 2011
Messages
10,960 (2.27/day)
Location
Florida
Processor 3900X @ 4.0
Motherboard Asus ROG Strix X570-E
Cooling DeepCool Castle 360EX
Memory G Skill Trident Z Neo 32GB 3600
Video Card(s) RX 5700 XT Pulse
Storage Sabrent Rocket Q 1TB
Display(s) LG 34UC88
Case Thermaltake P3
Power Supply Super Flower Leadex III 750w
Mouse Logitech G900
Keyboard G Skill KM570 MX Silver
Software Windows 10 Pro
I have a triple rad and I wonder if adding 3 more fans below to it would make a difference.
Somewhere I read that pump speed is much, but very much more important than rad fans speed or quantity.
I have spare fans so it's not a cost problem, but it makes my build more complex.

I initially ran push/pull, but eventually opted to just push and the difference wasn't huge, pretty negligible. Later added extra rad space instead.

I'd also appreciate recommendations regarding the fans powering. Should I just join 6 fans into one onboard fan header? Or better not? Or use 2? At the moment, I have 3 in 1 cables.

What do you think?

I wired 3 fans together into one connection and run them off the board, but now they're plugged into a splitter, something along these lines; plugs into the psu connector. It's been running that way for a few years and going strong til today.


Also just wondering, what are you watercooling: cpu only or cpu and gpu?
 
Joined
Jan 31, 2012
Messages
2,452 (0.55/day)
Location
Bulgaria
System Name Sandfiller
Processor I5-10400
Motherboard MSI MPG Z490 GAMING PLUS
Cooling Noctua NH-L9i (92x25mm fan)
Memory 32GB Corsair LPX 2400 Mhz DDR4 CL14
Video Card(s) MSI RX 5700 XT GAMING X
Storage Intel 670P 512GB
Display(s) 2560x1080 LG 29" + 22" LG
Case SS RV02
Audio Device(s) Creative Sound Blaster Z
Power Supply Fractal Design IntegraM 650W
Mouse Logitech Triathlon
Keyboard REDRAGON MITRA
Software Windows 11 Home x 64
Temperature-wise, I think it benefits. I earned 2-3 degrees with push-pull on my Kraken, but it all depends if you really need this. Also, if you will have 10 fans in the PC, that might be a complication : noise-wise and powering them all. Driving 3-4 fans off one MB header-> please check your MB product page/manual. It should say the maximum power output. It says on mine at least, but I have a Lamptron fan controller anyway.
 

CAPSLOCKSTUCK

Spaced Out Lunar Tick
Joined
Feb 26, 2013
Messages
8,578 (2.12/day)
Location
llaregguB...WALES
System Name Party On
Processor Xeon w 3520
Motherboard DFI Lanparty
Cooling Big tower thing
Memory 6 gb Ballistix Tracer
Video Card(s) HD 7970
Case a plank of wood
Audio Device(s) seperate amp and 6 big speakers
Power Supply Corsair
Mouse cheap
Keyboard under going restoration
I use this

molex to fan header

 
Joined
Mar 24, 2010
Messages
5,047 (0.99/day)
Location
Iberian Peninsula
Also just wondering, what are you watercooling: cpu only or cpu and gpu?

Good one! I run both in one loop, overclocked, and recently I have returned to a little benching. That is the only reason, maybe, for "over-engeneering" again :)
Maybe I should just fix some heat issues lowering voltages (went from 1,3 to 1,27 Vcore (1,29 SA 1,28 VCCIO ) and that solved one benchcrash due to heat... now passing RealBench at cpu=4800Mhz)
 
Joined
Feb 3, 2017
Messages
3,475 (1.33/day)
Processor R5 5600X
Motherboard ASUS ROG STRIX B550-I GAMING
Cooling Alpenföhn Black Ridge
Memory 2*16GB DDR4-2666 VLP @3800
Video Card(s) EVGA Geforce RTX 3080 XC3
Storage 1TB Samsung 970 Pro, 2TB Intel 660p
Display(s) ASUS PG279Q, Eizo EV2736W
Case Dan Cases A4-SFX
Power Supply Corsair SF600
Mouse Corsair Ironclaw Wireless RGB
Keyboard Corsair K60
VR HMD HTC Vive
temperature-wise, the difference is not that big but it does seem to allow lower fan speeds. which kind of makes sense when you think about it.
 
Joined
Oct 17, 2012
Messages
9,781 (2.34/day)
Location
Massachusetts
System Name Americas cure is the death of Social Justice & Political Correctness
Processor i7-11700K
Motherboard Asrock Z590 Extreme wifi 6E
Cooling Noctua NH-U12A
Memory 32GB Corsair RGB fancy boi 5000
Video Card(s) RTX 3090 Reference
Storage Samsung 970 Evo 1Tb + Samsung 970 Evo 500Gb
Display(s) Dell - 27" LED QHD G-SYNC x2
Case Fractal Design Meshify-C
Audio Device(s) on board
Power Supply Seasonic Focus+ Gold 1000 Watt
Mouse Logitech G502 spectrum
Keyboard AZIO MGK-1 RGB (Kaith Blue)
Software Win 10 Professional 64 bit
Benchmark Scores the MLGeesiest
i keep my PC cool for DAYS with elite pro hacker tactics.......

& for Larger server sized application.......

only Scrubs use fans....Pffft! :slap:
 

peche

Thermaltake fanboy
Joined
Nov 7, 2014
Messages
6,709 (1.96/day)
Location
San Jose, Costa Rica
System Name Athenna
Processor intel i7 3770 *Dellided*
Motherboard GIGABYTE GA-Z68X-UD3H-B3 Rev. 1.1
Cooling Thermaltake Water 3.0 Pro + Tt Riing12 x2 / Tt ThunderBlade / Gelid Slim 120UV fans
Memory 16GB DRR3 Kingoston with Custom Tt spreaders + HyperX Fan
Video Card(s) GeForce GTX 980 4GB Nvidia Sample
Storage Crucial M4 SSD 64GB's / Seagate Barracuda 2TB / Seagate Barracuda 320GB's
Display(s) 22" LG FLATRON 1920 x 1280p
Case Thermaltake Commander G42 Window
Audio Device(s) On-board Dolby 5.1+ Kingston HyperX Cloud 1
Power Supply Themaltake TR2 700W 80plus bronce & APC Pro backup 1000Va
Mouse Tt eSports Level 10M Rev 1.0 Diamond Black & Tt Conkor "L" mouse pad
Keyboard Tt eSports KNUCKER
Software windows 10x64Pro
Benchmark Scores well I've fried a 775' P4 12 years ago, that counts?
1. avoid using 1 header or connection for moar than 2 fans, use logic that will hurt something,
2. if you already have the fans somewhere there, use'em, if you have the idea to improve cooling, better add another rad to the loop, difference will be max 10C in the best scenario,
3. a great will item to use will be a molex / header spliter for moar fans, like the mention in post #9 , pretty useful indeed,


according your specs...This case has plenty space for an epic cable management...
according your specs...
 
Last edited:
Joined
Apr 24, 2012
Messages
1,589 (0.36/day)
Location
Northamptonshire, UK
System Name Main / HTPC
Processor Ryzen 9 5900X / Ryzen 7 2700
Motherboard Strix B550i / B450i Aorus Pro
Cooling Lian-Li Galahad 360 / Wraith Spire
Memory Corsair LPX 2x16 3600MHz / HyperX Predator 2x8GB 3200MHz
Video Card(s) RTX 3080 FE / ARC A380
Storage WD Black SN770 1TB / Sabrent Rocket 256GB
Display(s) Acer Z301c / 39" Panasonic HDTV
Case Corsair 2000D / Cougar QBX
Audio Device(s) Yamaha RX-V379 / Realtek ALC1220
Power Supply Corsair SF600 / BeQuiet SFX Power 2 450W
Mouse Logitech G900
Keyboard Drop Sense75 with WQ Studio Morandi's
VR HMD Rift S
Software Win 11 Pro 64Bit
How thick is the rad? If its a slim one, there will be no difference, you're just adding more noise to your system.
 
Joined
Dec 31, 2009
Messages
19,366 (3.72/day)
Benchmark Scores Faster than yours... I'd bet on it. :)
Hi!
I have a triple rad and I wonder if adding 3 more fans below to it would make a difference.
Somewhere I read that pump speed is much, but very much more important than rad fans speed or quantity.
I have spare fans so it's not a cost problem, but it makes my build more complex.

I'd also appreciate recommendations regarding the fans powering. Should I just join 6 fans into one onboard fan header? Or better not? Or use 2? At the moment, I have 3 in 1 cables.

What do you think?
Where to start....

Are your current temps OK? If so, I wouldn't bother... its just extra noise for a couple degrees C different.

Pump speed isn't terribly important. You need to keep around 1-1.5GPM flow for optimum results. Faster it is diminishing returns, slower, you lose a bit of performance.

DO NOT place 6 fans on to one header unless you know how much amperage it puts out. Some headers are high amp headers and can support 2A(24W) and can piggy back a few fans. Most are around 1A(12W) however. The stick on the fan hub will show how much power each uses. I am sure you can do the math at that point.
 
Joined
Mar 24, 2010
Messages
5,047 (0.99/day)
Location
Iberian Peninsula
Thanks all, for your useful comments. My rad is the EK Extreme 360, a fatty. So far I have gone the voltage reducing route and that has help a lot. Next week i will also install a twice as big reservoir and then I will see if anything like pull/push fans is still necessary.

My temps now: during gaming they stay under 60 ºC. Benching they crawl up to 80 ºC. (CPU). Did a long run with RealBench and the lower voltages and it went flawless staying at said 80ºC. As I mentioned somewhere above, my question started when at 1,32v core it crashed, the temps were close to 90ºC. So now it seems all solid. (CPU@4800Mhz all 6 cores.)

Thanks again! Will upload a pic when the new parts are added. ;)
 
Joined
Dec 31, 2009
Messages
19,366 (3.72/day)
Benchmark Scores Faster than yours... I'd bet on it. :)
What does a resivoir have to do with temps?

I think you would see better results from delidding than adding more fans. It isn't so much that you don't have enough rad or fans, but the heat isn't being transfered well from the CPU to the block due to the TIM used.
 
Joined
Mar 24, 2010
Messages
5,047 (0.99/day)
Location
Iberian Peninsula
Moar liquid = moar cooler. The stock reservoir is very small (12cm).

Delidding, ok ok ok, during holidays, maybe :eek: :rockout:
 
Joined
Dec 31, 2009
Messages
19,366 (3.72/day)
Benchmark Scores Faster than yours... I'd bet on it. :)
No... it means it takes just a bit longer to warm up/normalize. It does NOT mean more cooling. :)

Now, if you used a POOL, I would agree, but a few ounces of water isn't going to do much of anything, sorry.
 
Joined
Mar 18, 2015
Messages
2,960 (0.90/day)
Location
Long Island
1. Push / pull will add about 30% more cooling ability. Push has the edge < around 1750 rpm ... pull does little better > 1750 rpm. This won't be true for every rad and every fan but "on average". See articles at MartinsLiquidLab ... or d/l the rad info here

http://www.overclock.net/t/1457426/radiator-size-estimator

2, Fan headers are typically rated at 1 amp... you want to be a bit below that to allow for inrush current. For 1250 rpm 140mm fans, I will do up to 6 ... w/ 1500 rpm up to 5, w/ 1800 up to 4. I recommend the Phanteks Fan Hub as its versatile... for PWM fans on PWM header Swiftech works well... for DCV headers w/ DCV fans, can also use the PCB hubs from ModMyToys.

3. There's no real advantage anymore to spending twice as much for PWM fans ... most DCV fan designs today go down to 25% speed w/o an issue and don't exhibit the low speed humming / clicking that sometimes occurs w/ PWM. The Phanteks Hub can be used w/ up to 11 fans on PWM header w/ the SATA power cable. If ya use power cable on DCV header, you lose speed control so limit number of fans per Item 2 above (< 1.0 amps). I have 4 channels ..

a. CPU & CPU-OPT (PWM) => twin pumps
b. CHA 1 (DCV) => Hub No. 1 => (6) 1250 rpm fans on 420mm rad
c. CHA 2 (DCV) => Hub No. 2 => (4) 1250 rpm fans om 280mm rad
d. CHA 3 (DCV) => Hub No. 3 => 6) 1250 rpm case fans

All fans range from 350 - 850 rpm under fan curve, rad fans ran ramp down after temps drop over 60 seconds to drop coolant temps. Fans are set to turn off when temps would call for flows below 325.

4. Running or PSU (Molex) means no speed control and noise.

5. There is an advantage to varying pump speed but after 1.00 it starts to tail off and once ya get over 1.25 gpm, it's has sharply diminishing returns. Pump (35x@ twin pumps) speed curve on my box typically varies from about 35% to 65% of full speed but it will go to max under Furmark + RoG Real Bench when run together. CPU will get to mid 70s under ROG Real bench ... the GPUs are at 44c ... they drop to 29C aw/ fans at full speed but fans audible over 850 rpm

6. A reservoir is a valuable tool as it adds thermal mass ... when exposed to varying loads, your system can chase it 's tail so to speak as fans ramp up during short heave loads and then ramp down when load drops... but "in between" your coolant didn't drop because ya fans ramped down as soon as load dropped. Result is a fans ramping up and down repeatedly leaving you listening to the whirring noise each time they ramp up. Use the fan utility in your MoBo package to set ramp down times once you add a reservoir. Another function of the reservoir is to trap any "off-gassing". You can bleed your loop, after installing everything and soon after you will see or hear gas bubbles in the loop. Ignore it and ya can lose a pump. Ya still have to pay attention as air can be trapped up ant highest point in ya loop but res will trap most of it.

7. For me, it';s just as much about noise as temps ... if i can tell PC is running when I walk in room and sit down in front of KB, (screen asleep) ... then poppa ain't happy. It's not quiet I'm looking for, it's dead silence. Fans matter. I prefer more fans to higher rpm fans. Many of the water cooling guru choices from days gone by , don't really work so well on the low - medium fpi rads that dominate the marketplace. We don't need the SP that for example the ole he old Nidec (Scythe) GTs provided. I was shocked when in testing, taking the Noctua fans off a Noctua cooler dropped CPU temps 6C at the same 1200 rpm speed.

http://www.overclockersclub.com/reviews/phenteks_f140/3.htm

If ya have a Phanteks case, they are golden ... in another case I will use the Phanteks or new Silent Wings 3 which came out after this test

http://www.silentpcreview.com/article1345-page7.html

8. One last note if using PWM fans, don't gang more than 4 or 5 Corsair fans on a single channel as they lose speed control

http://www.overclock.net/t/1506812/the-best-pwm-rad-fans-going-to-be-very-exact-spec-needs/30

9. Not a fan of fan controllers... the idea of manually adjusting fan speeds to varying loads is just not something I wanan deal with.

So is push / pull worth it ? If ya not happy with ya temps and no more mounts for rads, based upon martins testing, you can expect an improvement of about 30% or so ... if it's a cheap aluminum rad or a CLC, not worth the time and effort.
 
Last edited:
Joined
Feb 3, 2017
Messages
3,475 (1.33/day)
Processor R5 5600X
Motherboard ASUS ROG STRIX B550-I GAMING
Cooling Alpenföhn Black Ridge
Memory 2*16GB DDR4-2666 VLP @3800
Video Card(s) EVGA Geforce RTX 3080 XC3
Storage 1TB Samsung 970 Pro, 2TB Intel 660p
Display(s) ASUS PG279Q, Eizo EV2736W
Case Dan Cases A4-SFX
Power Supply Corsair SF600
Mouse Corsair Ironclaw Wireless RGB
Keyboard Corsair K60
VR HMD HTC Vive
9. Not a fan of fan controllers... the idea of manually adjusting fan speeds to varying loads is just not something I wanan deal with.
I dislike fan controllers for the same reason. However, you can get best from both worlds, for a price - configurable hardware controllers (with hardware sensors).
There are others but I am currently using Aquaero (5 LT): https://shop.aquacomputer.de/index.php?cPath=62_63_64
 
Joined
Dec 31, 2009
Messages
19,366 (3.72/day)
Benchmark Scores Faster than yours... I'd bet on it. :)
#6.... no. Adding a couple oz of water isnt going to help much at all.

#9. Fans arent/cant be set it and forget it? Whos running on the ragged edge of temps for normal activities that cranking a set of fans, lowering temps a few c, will resolve? There are also fan controllers which will work based on temps. ;)
 
Joined
Mar 24, 2010
Messages
5,047 (0.99/day)
Location
Iberian Peninsula
This is the final build for now. I am quite happy because I pass gloriously all stress tests.
For now no added fans. The upper 3 fans are off most of the time, and only start when temps are high.
I am not looking for records, but I am doing just fine in 3DMark, RealBench, Aida64 etc. And gaming at 4K...
Of course sugestions are welcome.

myrig.jpg


And for a more ambience impression, this contextual one.

myrig2.jpg
 
Last edited:
Joined
Mar 18, 2015
Messages
2,960 (0.90/day)
Location
Long Island
#6.... no. Adding a couple oz of water isnt going to help much at all.

#9. Fans arent/cant be set it and forget it? Whos running on the ragged edge of temps for normal activities that cranking a set of fans, lowering temps a few c, will resolve? There are also fan controllers which will work based on temps. ;)

On # 6 . . . Please read again.... nothing was said about coolant temps..... it's mostkly about noise and you can certainly observe fan speeds and record as we have. And it's not a couple of ounces of water, it's 10 ounces and a surface a surface area of 16+ square inches. Why does surface area matter ? Think about the 90 - 130 watt CPU that water block is only able to keep at 75C with its tiny surface area versus the 300 watt GPU that water block is able to keep < 40C.

1. You have to be using the proper tools. If using P95 which is virtually useless since Sandy Bridger. Yes it doesn't matter significantly as P95 will maintain a relatively constant load and yes a reservoir will have limited impact. But remember, as was said, the radiators remove ONLY about 60% of the calculated heat load. The other 40% comes from the fact that a) your system never sees the max heat load because **in the real world** and **using real applications / games** the heat load is not constant. A large 250 ml reservoir has a substantiall surface area and it acts just like an air cooler radiating heat from its surface, as is your tubing, radiator shroud, block surface area and everything else that is warm to the touch inside the PC ... the PC which has the interior case air volume exchanged with fresh cool air several times per second.

2. Again in the real world, your system sees periods of light and heavy loads. Watch your CPU temps under RoG Real Bench ... they vary significantly. Run Furmark and watch the curve .... mine starts out at about 26C GPU temp and rises slowly to 44C at which point, under the synthetic non-real world load, it reaches equilibrium. But in the real world, when paying a game, the load varies substantially. You can see the same thing with Intel XTU load curve, the CPU temps follow up down up down up down. With no reservoir capacity, the coolant temps change in response to CPU / GPU temps very quickly and the temp curve is very jagged. With a large reservoir, those peaks and valleys are smoothed out resulting in smaller variations in CPU / GPU temps. As a result, the fans whos speed is controlled automatically (a point you seem to have missed) changes very little

No reservoir ...... 65-75-68-72-70-70-64-76
w/ reservoirs.... 68-72-70-70-70-70-67-73

In both cases the average is 70C (ignoring thermal radiation from reservoir surface) but in one case the changes in fan speed can be heard, in the other not at all.

In addition, a reservoir acts as an air trap for any off gassing within the loop, an air bubble trapped in a loop can otherwise significantly impact performance.

On # 9. this is 2107 not 1991. Been building PC's for 25 years and every box that leaves here has been set up to "set it and forget it". The fans in our test box were set on the day I built the box and 3 years later, I have not touched them since. Cooling is not just about not frying the PC components, it's also about making the user environment bearable. And that involves minimizing sound. Siting at the desk, using your ears, you can not tell the system is on. Using the fan control utility that comes with the motherboard, I have the system set up as follows:

CPU & CPU OPT Header => Pumps 1 and 2 .
Pumps are set to run from about 40% to 100% of full speed. I set up a temp / speed curve using the software provided and forgot about it.

CHA_1 header => Fan Control PCB No. 1 => (6) fans on 420 radiator
Fans are set to run from 350 rpm to 1250 rpm. I set them to shut off when temps are < 30C, they climb thru various speed / temp setpoints to 75C and > 75 they go to 100%. In response to temp changes, they ramp up to curve speeds over 12 seconds, and ramp down over 100 seconds to expel heat from the coolant after load disappears. I set up a temp / speed curve using the software provided and forgot about it.

CHA_2 header => Fan Control PCB No. 2 => (4) fans on 280 radiator
Fans are set to run from 350 rpm to 1250 rpm. I set them to shut off when temps are < 30C, they climb thru various speed / temp setpoints to 75C and > 75 they go to 100%. In response to temp changes, they ramp up to curve speeds over 12 seconds, and ramp down over 100 seconds to expel heat from the coolant after load disappears. I set up a temp / speed curve using the software provided and forgot about it.

CHA_3 header => Fan Control PCB No. 3 => (6) Case fans
Fans are set to run from 350 rpm to 1250 rpm. I set them to shut off when temps are < 35C, they climb thru various speed / temp set points to 70C and > 80C they go to 100%. In response to temp changes, they ramp up / down to curve speeds over 12 seconds.I set up a temp / speed curve using the software provided and forgot about it for about a year until I made a tweak, not touched it since.

At idle, all my fans shut off... and I don't have to do anything ... when system is under load, fan speed is raised in proportion to the load. This is simple stuff, and I don't know why this is being seen as a foreign concept. I don't care about the "ragged edge of temps" because my system never sees it. What I care about is silence. There's no realistic need for a "fan controller" when superior technology exists that has more options, is free and which most certainly can be "set once and forgotten".

The system uses (6) thermal sensors, accurate to 0.1C, which detect:

coolant temps in / out of 420mm rad
coolant temps in / out of 280mm rad
Ambient and interior case air temps

and displays them on a 6 channel digital display in a 5.25" bay at front of PC.

Can take CPU temps down to , 70C at 4.7 Ghz and the twin GPU temps down to 39C, but at that point fan speeds under stress testing exceed 850 rpm and I can hear the fans... So to avoid buying 30 foot cables and stikcing the PC in another room where I don't have to hear it, I am quite happy with 75C and 44C. All accomplished without a thought or lifting a finger since day # 1
 
Last edited:
Top