Question Air Cooling My Gaming Rig - New to CFM and Cooling

[philzee]

My goal is to optimize the cooling performance of your PC setup. I am absoultely new to CFMs and case cooling. I don't want to overclock anything. I just want a reliable net positive airflow and cooling. My PC's primary use will be gaming.

I’ve been working towards achieving a target of maintaining safe temperatures (CPU <85°C, GPU hotspot <80°C) while ensuring positive airflow (ideally +50 to +150 CFM) across all operating temperatures (40°C, 55°C, 65°C, 80°C) to enhance gaming performance and prevent dust buildup.

Recently, I’ve adjusted fan curves to reach +108.2 CFM at 65°C (at least this is what I think I am doing), and I'm now trying to refine CFM values to maintain consistent positive pressure, especially addressing the GPU hotspot that’s been around 89.0°C.

Here is a table I have been keeping track of what fans I currently and what fan curves I am currently running in the BIOS.

Fan Name	Model	Location	Temp Source	Direction	Temp Settings (°C)	% Speed Settings	Estimated RPM	Estimated CFM	Max RPM
PUMP_FAN1	Thermalright TL-C14C-S	Front Panel	System	Intake	[40, 55, 65, 80]	[80, 85, 100, 100]	[1200.0, 1275.0, 1500.0, 1500.0]	[60.64, 64.43, 75.8, 75.8]	1500
SYS_FAN2	Thermalright TL-B8	Side Panel (y=66)	System	Intake	[40, 55, 65, 80]	[80, 85, 100, 100]	[1760.0, 1870.0, 2200.0, 2200.0]	[27.2, 28.9, 34.0, 34.0]	2200
SYS_FAN3	Thermalright TL-B8	Side Panel (y=297)	System	Intake	[40, 55, 65, 80]	[80, 85, 100, 100]	[1760.0, 1870.0, 2200.0, 2200.0]	[27.2, 28.9, 34.0, 34.0]	2200
CPU_FAN1	Scythe Fuma 2 (Fan 1+2)	CPU Cooler (Top)	CPU Core	Exhaust	[40, 55, 65, 80]	[40, 45, 55, 60]	[600.0, 675.0, 825.0, 900.0]	[20.47, 23.03, 28.14, 30.7]	1500
SYS_FAN1	Super Flower 120mm	Rear Panel	System	Exhaust	[40, 55, 65, 80]	[40, 45, 50, 50]	[600.0, 675.0, 750.0, 750.0]	[26.87, 30.23, 33.59, 33.59]	1500
SYS_FAN4	Super Flower 120mm (3x)	Top Panel	System	Exhaust	[40, 55, 65, 80]	[40, 45, 50, 45]	[600.0, 675.0, 750.0, 675.0]	[26.87, 30.23, 33.59, 30.23]	1500
SYS_FAN5	Thermalright TL-C12C-S	Bottom/Side (PCI)	System	Exhaust	[40, 55, 65, 80]	[40, 45, 60, 70]	[600.0, 675.0, 900.0, 1050.0]	[26.87, 30.23, 40.3, 47.02]	1500

Total Intake vs. Total Exhaust by Temp Setting (for reference):

• 40°C: Intake 236.32 CFM vs. Exhaust 170.58 CFM, Net +65.74 CFM
• 55°C: Intake 251.09 CFM vs. Exhaust 191.91 CFM, Net +59.18 CFM
• 65°C: Intake 295.4 CFM vs. Exhaust 224.48 CFM, Net +70.92 CFM
• 80°C: Intake 295.4 CFM vs. Exhaust 225.65 CFM, Net +69.75 CFM

I have swapped out the stock fan in the FDP02 external GPU exhaust with a Thermalright TL-C12C-S.

I really don't know if I am doing any of this right. I have been consulting with AI on this. But now I am at a spot where I would like to know what I am doing and do it right. Thank you.

Here the case fan CFMs I have:
Thermalright TL-C14C-S 140mm 75.8CFM
Thermalright TL-C12C-S 120mm 67.17CFM
Thermalright TL-B8 80mm 34CFM
Scythe Fuma 2 fan 1 120 x 120 x 15 mm 39.44CFM
Scythe Fuma 2 fan 2 120 x 120 x 26 mm 51/17CFM

PCPartPicker Part List

CPU: AMD Ryzen 9 7950X3D 4.2 GHz 16-Core Processor ($685.00 @ Amazon)
CPU Cooler: Scythe FUMA 2 51.17 CFM CPU Cooler
Motherboard: MSI MAG X670E TOMAHAWK WIFI ATX AM5 Motherboard ($248.95 @ Amazon)
Memory: TEAMGROUP T-Force Delta RGB 32 GB (2 x 16 GB) DDR5-6000 CL28 Memory ($124.99 @ Newegg)
Storage: ADATA XPG SX8200 Pro 1 TB M.2-2280 PCIe 3.0 X4 NVME Solid State Drive ($62.99 @ Amazon)
Storage: Crucial T500 W/Heatsink 1 TB M.2-2280 PCIe 4.0 X4 NVME Solid State Drive ($79.98 @ Amazon)
Storage: Seagate Barracuda Compute 2 TB 3.5" 7200 RPM Internal Hard Drive ($64.98 @ Newegg)
Video Card: MSI Radeon RX 6800 XT Gaming X Trio 16G Radeon RX 6800 XT 16 GB Video Card ($1216.94 @ Amazon)
Power Supply: SeaSonic FOCUS GX 750 W 80+ Gold Certified Fully Modular ATX Power Supply ($175.26 @ Amazon)
Operating System: Microsoft Windows 11 Home OEM - DVD 64-bit ($109.99 @ Amazon)
Case Fan: Thermalright TL-B8 34 CFM 80 mm Fan ($7.79 @ Amazon)
Case Fan: Thermalright TL-B8 34 CFM 80 mm Fan ($7.79 @ Amazon)
Case Fan: Thermalright TL-C14C-S 75.8 CFM 140 mm Fan ($8.28 @ Amazon)
Case Fan: Thermalright TL-C14C-S 75.8 CFM 140 mm Fan ($8.28 @ Amazon)
Case Fan: Thermalright TL-C14C-S 75.8 CFM 140 mm Fan ($8.28 @ Amazon)
Case Fan: Thermalright TL-C12C-S 66.17 CFM 120 mm Fan ($7.90 @ Amazon)
Case Fan: Thermalright TL-C12C-S 66.17 CFM 120 mm Fan ($7.90 @ Amazon)
Case: Super Flower ZILLION M705
Case Fan : SilverStone FDP02 External GPU cooling Fan
ARGB Splitter Hub: Miicro Connectors 11-Port ARGB Splitter Hub for Fan and Light

 

[CountMike]

My goal is to optimize the cooling performance of your PC setup. I am absoultely new to CFMs and case cooling. I don't want to overclock anything. I just want a reliable net positive airflow and cooling. My PC's primary use will be gaming.

I’ve been working towards achieving a target of maintaining safe temperatures (CPU <85°C, GPU hotspot <80°C) while ensuring positive airflow (ideally +50 to +150 CFM) across all operating temperatures (40°C, 55°C, 65°C, 80°C) to enhance gaming performance and prevent dust buildup.

Recently, I’ve adjusted fan curves to reach +108.2 CFM at 65°C (at least this is what I think I am doing), and I'm now trying to refine CFM values to maintain consistent positive pressure, especially addressing the GPU hotspot that’s been around 89.0°C.

Here is a table I have been keeping track of what fans I currently and what fan curves I am currently running in the BIOS.

Fan Name	Model	Location	Temp Source	Direction	Temp Settings (°C)	% Speed Settings	Estimated RPM	Estimated CFM	Max RPM
PUMP_FAN1	Thermalright TL-C14C-S	Front Panel	System	Intake	[40, 55, 65, 80]	[80, 85, 100, 100]	[1200.0, 1275.0, 1500.0, 1500.0]	[60.64, 64.43, 75.8, 75.8]	1500
SYS_FAN2	Thermalright TL-B8	Side Panel (y=66)	System	Intake	[40, 55, 65, 80]	[80, 85, 100, 100]	[1760.0, 1870.0, 2200.0, 2200.0]	[27.2, 28.9, 34.0, 34.0]	2200
SYS_FAN3	Thermalright TL-B8	Side Panel (y=297)	System	Intake	[40, 55, 65, 80]	[80, 85, 100, 100]	[1760.0, 1870.0, 2200.0, 2200.0]	[27.2, 28.9, 34.0, 34.0]	2200
CPU_FAN1	Scythe Fuma 2 (Fan 1+2)	CPU Cooler (Top)	CPU Core	Exhaust	[40, 55, 65, 80]	[40, 45, 55, 60]	[600.0, 675.0, 825.0, 900.0]	[20.47, 23.03, 28.14, 30.7]	1500
SYS_FAN1	Super Flower 120mm	Rear Panel	System	Exhaust	[40, 55, 65, 80]	[40, 45, 50, 50]	[600.0, 675.0, 750.0, 750.0]	[26.87, 30.23, 33.59, 33.59]	1500
SYS_FAN4	Super Flower 120mm (3x)	Top Panel	System	Exhaust	[40, 55, 65, 80]	[40, 45, 50, 45]	[600.0, 675.0, 750.0, 675.0]	[26.87, 30.23, 33.59, 30.23]	1500
SYS_FAN5	Thermalright TL-C12C-S	Bottom/Side (PCI)	System	Exhaust	[40, 55, 65, 80]	[40, 45, 60, 70]	[600.0, 675.0, 900.0, 1050.0]	[26.87, 30.23, 40.3, 47.02]	1500

Total Intake vs. Total Exhaust by Temp Setting (for reference):

• 40°C: Intake 236.32 CFM vs. Exhaust 170.58 CFM, Net +65.74 CFM
• 55°C: Intake 251.09 CFM vs. Exhaust 191.91 CFM, Net +59.18 CFM
• 65°C: Intake 295.4 CFM vs. Exhaust 224.48 CFM, Net +70.92 CFM
• 80°C: Intake 295.4 CFM vs. Exhaust 225.65 CFM, Net +69.75 CFM

I have swapped out the stock fan in the FDP02 external GPU exhaust with a Thermalright TL-C12C-S.

I really don't know if I am doing any of this right. I have been consulting with AI on this. But now I am at a spot where I would like to know what I am doing and do it right. Thank you.

Here the case fan CFMs I have:
Thermalright TL-C14C-S 140mm 75.8CFM
Thermalright TL-C12C-S 120mm 67.17CFM
Thermalright TL-B8 80mm 34CFM
Scythe Fuma 2 fan 1 120 x 120 x 15 mm 39.44CFM
Scythe Fuma 2 fan 2 120 x 120 x 26 mm 51/17CFM

PCPartPicker Part List

CPU: AMD Ryzen 9 7950X3D 4.2 GHz 16-Core Processor ($685.00 @ Amazon)
CPU Cooler: Scythe FUMA 2 51.17 CFM CPU Cooler
Motherboard: MSI MAG X670E TOMAHAWK WIFI ATX AM5 Motherboard ($248.95 @ Amazon)
Memory: TEAMGROUP T-Force Delta RGB 32 GB (2 x 16 GB) DDR5-6000 CL28 Memory ($124.99 @ Newegg)
Storage: ADATA XPG SX8200 Pro 1 TB M.2-2280 PCIe 3.0 X4 NVME Solid State Drive ($62.99 @ Amazon)
Storage: Crucial T500 W/Heatsink 1 TB M.2-2280 PCIe 4.0 X4 NVME Solid State Drive ($79.98 @ Amazon)
Storage: Seagate Barracuda Compute 2 TB 3.5" 7200 RPM Internal Hard Drive ($64.98 @ Newegg)
Video Card: MSI Radeon RX 6800 XT Gaming X Trio 16G Radeon RX 6800 XT 16 GB Video Card ($1216.94 @ Amazon)
Power Supply: SeaSonic FOCUS GX 750 W 80+ Gold Certified Fully Modular ATX Power Supply ($175.26 @ Amazon)
Operating System: Microsoft Windows 11 Home OEM - DVD 64-bit ($109.99 @ Amazon)
Case Fan: Thermalright TL-B8 34 CFM 80 mm Fan ($7.79 @ Amazon)
Case Fan: Thermalright TL-B8 34 CFM 80 mm Fan ($7.79 @ Amazon)
Case Fan: Thermalright TL-C14C-S 75.8 CFM 140 mm Fan ($8.28 @ Amazon)
Case Fan: Thermalright TL-C14C-S 75.8 CFM 140 mm Fan ($8.28 @ Amazon)
Case Fan: Thermalright TL-C14C-S 75.8 CFM 140 mm Fan ($8.28 @ Amazon)
Case Fan: Thermalright TL-C12C-S 66.17 CFM 120 mm Fan ($7.90 @ Amazon)
Case Fan: Thermalright TL-C12C-S 66.17 CFM 120 mm Fan ($7.90 @ Amazon)
Case: Super Flower ZILLION M705
Case Fan : SilverStone FDP02 External GPU cooling Fan
ARGB Splitter Hub: Miicro Connectors 11-Port ARGB Splitter Hub for Fan and Light

Honest opinion, switch to a good 360 AIO liquid cooler and install as front intake and have same number of case fans as exhaust.
You haven't mentioned which case you have, it can make huge difference in cooling.

 

[philzee]

@CountMike

Honest opinion, switch to a good 360 AIO liquid cooler and install as front intake

This does have my interest. I will have to look into these. Is this something you would normally connect to the pump_fan head on the motherboard?
I found this article: Best AIO Coolers 2025

and have same number of case fans as exhaust.

Currently, I have 3x 140mm fans in the front intake and 2x 80mm in the side intake.
And then I have 3x 120mm in the top exhaust, 1x 120mm in the top rear exhaust, and 1x 120mm in the external rear exhaust (FDP02).
after installing these fans, there are a couple of schools of thought on whether to count the CPU fans as part of determining the net airflow. I erred on the side of caution and included them in the equations.

You haven't mentioned which case you have, it can make huge difference in cooling.

Super Flower ZILLION M705

I should also mention that I discovered that I am getting 0 RPM on pump_fan1 even though the front fans are spinning. There might be a compatibility issue with the either the fans or the splitter and the pump_fan1 head. I will be testing out my free sys_fan6 head to see if I get a reading. if I do, I will be swapping one of my 80mm fans to the sys_fan6 head and plugging in the front fans to the head that I freed up from moving the 80mm fan.

 

[CountMike]

In no particular order
That's well ventilated case with a lot of mesh, I doubt you would be able to make positive pressure in it.
The main purpose of cooling is to take heat away from it's source, just pumping a lot of air to have positive pressure is not going to help much unless most of that air is evacuated. There are 2 instances where some positive pressure helps. Keeping dust away and enabling air to reach all places but it also means reduced airflow.
Ventilators on the air cooler influence only air flow inside the case by channeling it. Too much and it can take air away from other parts.
Unless there is some evaporation involved, just blowing air at an object doesn't actually cool it. Depending on temperature delta (difference between temperatures) it may even warm it. It can never cool it bellow air temperature.
Liquid coolers have a lot more mass than air coolers so they are better at keeping temperatures even under prolonged loads ad also have more surface to transfer heat and their fans directly count as part of case cooling and therefore can cut down on fan numbers needed for same effect.
Pumps on AIOs can be connected to any header that can be set to react to CPU temps or in some sets they are controlled together with fans. Arctic Liquid Freezer III Coolers for instance come with cables for either configuration plus have a small fan mounted on pump to blow air at VRM heat spreaders and M.2 socket and help their cooling (somewhat).

 

[Zerk2012]

89C for the GPU hotspot is not really bad.

 

[philzee]

89C for the GPU hotspot is not really bad.

I like hearing this. I am aiming for lower. I don't expect the moon or anything, just lower temps under load.

 

[philzee]

@CountMike
With my current setup I could stop targeting positive pressure. but that leaves me wondering if I should target just a little bit of negative pressure and how much is too much negative pressure.

 

[Paperdoc]

A couple suggestions, but start with an FYI. Every fan generates a speed signal consisting of a train of pulses (2 per evolution, 5 VDC amplitude) sent back to pin #3 of the host header on the mobo for counting to allow display of that speed. This speed info is NOT used for control of fan speed, but it is monitored for possible fan failure (that is, NO speed signal or sometimes a signal too slow). Failure normally causes a message to pop up on your screen so you can take actions as needed. The CPU_FAN header may take extra actions on failure detection, including shutting down your system without even waiting for high CPU temperature detection, as a way to prevent real damage to an uncooled CPU chip. Because the speed signal is pulses to count, each header can deal with such signals from only ONE fan source. So when you use a Splitter or a Hub to connect more than one fan to a header, that device will send back to the header the speed of only ONE of its fans and will ignore all others. Those "other" speeds will never be "seen" anywhere. For a simple Splitter or Hub that looks like a collection of cable "arms", only ONE output will have all four pins in it and the others will be missing Pin #3 so those fans' speed signals cannot be connected back. Thus you MUST connect one fan to the only output with all four pins. If you do not do this, the host header will receive NO speed signal to dsiplay and monitor. On other forms of Splitter or Hub ONE output port will be marked in some way to indicate which can send its speed signal back to the host header.

With an air cooler on the CPU that fan has NO impact on net air flow into and out of the case and you do not count it in calculations. When using an AIO system the RAD FANS do impact case air flow directly and need to be taken into account. However, because the rad fins are a real impediment to air flow, the actual air flow into (or out of) the case is significantly LESS that the specified air flow of those fans. (The max air flow spec for a fan is at NO back pressure or flow resistance.) Similarly (but to a smaller extent), any air INTAKE fan should have a dust filter in front of it and that filter reduces the air flow somewhat.

Given thoee impacts on air flow I do not know how you can estimate the air flow from a fan, but I doubt you can MEASURE it. For your purpose of determining whether or not you have a small positive air pressure inside the case to prevent dust accumulation there is a simpler way - the smoke tracer technique. For this you need a small source of smoke, like an incense stick or a smoldering cigarette. With the system running at some workload you move the smoke source around the outside of your case near any small air leakage points. Observe the smoke plume movement. If it drifts away from the case, you have a small positive pressure inside - success! If the smoke is sucked into the case you have negative pressure and dusty air can enter there. If the smoke moves very rapidly either way the pressure is higher than needed and you can adjust to reduce. Do this process for a few workloads from idle to heavy, and adjust your fan curves for each scenario until all show small positive internal pressure.

 

[geofelt]

You have the wherewithal to do some experimenting.
I would try the 3 140mm front intakes.
Use a single 120mm rear exhaust to direct the airflow past the cpu cooler and gpu.
Side 80mm fans disrupt the airflow and top exhaust fans divert the airflow up and out of the case. Abandon them.