Loading

Fan Performance Testing (Part 1)

Published on 5th February, 2017

LZ7 Side Power Panel with 140mm Fan and Finger Guard

This is the first of a 2 part article investigating fan performance and how it is affected by case design.  The research is being used to improve and develop current and future Lazer3D cases.

[nextpage title=”Introduction” ]

INTRODUCTION

I think most people are aware that fan performance is affected by vent design and close proximity to flat surfaces, but the big question is by how much?

The aim of this article is to try and apply a bit of scientific experimentation to measure the percentage loss of airflow in different scenarios.  This first part of the article will focus on how well a PC fan can cope when placed near to a flat surface on its intake side.  The second part (to follow) will focus on vent design efficiency (difference between perforation, slots and dust filters, etc).

 

BACKGROUND

When people build a PC in a console style case such as the Silverstone RVZ01 or Node 202, they usually find that they get better thermal performance standing the case upright rather than lying it flat.

The reason for this is because the GPU fans are in close proximity and parallel to the flat surface of the desktop the case is sat on.  Even though the desk appears to be a good distance away from the GPU fans (20 – 30mm in most cases), this distance still has a significant affect on fan efficiency.

So just how much of a reduction in fan performance is the desktop surface causing? Let’s find out:

[/nextpage]

[nextpage title=”Test Setup” ]

TEST SETUP

To measure the fan performance a Wind Speed Meter was used, placed parallel with the fan at a distance of 5cm from the fans exhaust side, the resulting airflow produced by the fan is displayed on the Wind Speed Meter in km/h.

The Wind Speed Meter used:

PC Fan Testing Wind Speed Meter Anemometer

 

A selection of fans were used for testing, including:

  • Silverstone AR05/6 Fan (92x15mm)
  • Silverstone SST-FW121 (120x15mm)
  • Silverstone FQ141 (140x25mm)
  • Prolimatech Ultra Sleek Vortex 14 (140x15mm)

For the purposes of this test all the fans were set to a constant 80% power level.

The following photo shows the test setup using the Prolimatech Ultra Sleek Vortex 14 fan without any airflow restrictions:

PC Fan Performance Testing - Prolimatech Ultra Sleek Vortex 14 - Free Air Configuration

 

A large A4 sized block of foam positioned parallel to the fan was used to simulate a solid flat surface, this was moved towards the intake side of the fan at set distance intervals.

Here you can see the large foam block positioned 10cm (4″) from the fan:

PC Fan Performance Testing - Prolimatech Ultra Sleek Vortex 14 - 10cm Distance to flat surface

 

Here is the same fan set at the same 80% power, but notice the drop in wind speed measured when the surface is moved closer, shown at 6cm:

PC Fan Performance Testing - Prolimatech Ultra Sleek Vortex 14 - 6cm Distance to flat surface

[/nextpage]

[nextpage title=”Results” ]

RESULTS

The chart below shows the drop in Wind Speed detected by the Meter as the flat surface gets closer to the fan intake opening:

 

PC Fan Performance Testing - When Intake Side is Close to a Flat Surface

*CLICK THE IMAGE TO ENLARGE

 

As you can see three out of the four fans tested start to see a noticeable reduction in airflow at a distance as far as 15cm (6″), the Silverstone 120mm slim fan fared best maintaining full performance until 8cm.

The chart shows all four fans see a steady decline in performance as they get closer to a flat surface with a sudden and dramatic fall in airflow after a certain distance, this drop off point is different for each fan.

Out of all 4 fans tested the 92mm fan worked best at close distances less than 3cm retaining up to 80% of its performance.

Below is a table (Click to Enlarge) showing what efficiency each fan is performing at set distances:

 

PC Fan Efficiency Drop Comparison Against a FlatSurface

 

The above chart once again shows that the smaller the fan the better it fairs at closer gaps to a flat surface, with the 92mm Silverstone fan retaining 83% of its original airflow at a 3cm gap, whereas the Prolimatech 140mm Slim Fan did not register any airflow on the meter at this same 3cm gap.

It is worth noting that even though the 140mm Ultra Sleek Vortex 14 was not registering any airflow on the meter reading at a distance of 3cm, you could still feel a gentle breeze with your hand, but not the same typhoon of air it usually produces when unrestricted.  The Wind Speed Meter is only capable of reading airflow above 1.4km/h, so once the fan is producing less than 1.4km/h the meter reads zero.

[/nextpage]

[nextpage title=”Conclusion” ]

CONCLUSION

Smaller fans perform more efficiently in close proximity to flat surfaces.

92mm fans retain around 80% of their performance at a distance of 3cm from a flat surface, but lose nearly all their airflow with a gap of less than 2cm.

120mm slim fans require at least 4cm of free space to work effectively and should be used with caution if the gap is less than 3cm.

140mm fans need considerable clearance to work efficiently, ideally above 5cm to retain at least 60% efficiency.  A gap of less than 3cm will result in very little airflow.

Fan clearance and output efficiency is only part of the full picture, you also need to factor in how restrictive the vent that the air is being pulled through is.  The second part of this article will go into more detail about this exploring common vent designs.

 

CLOSING THOUGHTS

A lot Graphics Cards use 92mm slim fans for their cooling solution, meaning that ideally most cards will need at least 3cm of distance between the fan and the desktop to have the best chance of not throttling during a gaming session.

If you are using an oversized GPU with larger fans such as 120mm slim fans, then you will need more distance between the fan and the desktop for optimal performance, ideally a minimum of 5cm.

If you are using a dual slot GPU in the Silverstone RVZ02 then I estimate the gap from your GPU fan to the desktop (if you are using the included rubber feet) is about 25mm (1″).  This explains why many people see a large reduction in GPU performance when the case is lay on its side.  If you are using the case in this orientation then you will best suited to select a GPU with 3x 92mm fans (or smaller) rather than 2x 120mm fans.  Or alternatively flip the case upside down so the GPU is facing upwards, and use a CPU cooler less than 40mm in height.

The Node 202 should fair better than the RVZ02 as the GPU sits higher in the 202, this extra few cm will be enough to allow the GPU to breath much more effectively when the case is lay down sideways.

Some tower cases have fan mounting positions on the underside of the case for large 140mm fans.  Unless the case feet are taller than 5cm then you may not be getting much airflow out of these fans if used as an intake vent.  These bottom mounted vents may be more useful if used as exhaust vents rather than intake vents.

Other scenarios where this research may help to improve system performance would be in situations where the CPU fan is drawing in air close to an internal surface of the chassis that isn’t ventilated, such as the side or top panel of the case.  If you find you have less than 5cm of clearance for your fan then you may actually get better cooling if you use a heatsink with a 92mm or smaller fan rather than large heatsinks with 120mm or larger fans, or use a tower cooler with a side mounted fan.

 

WHY DO THE SMALLER FANS SHOW FASTER AIRFLOW?

You may be thinking that the chart seems a little misleading as it shows the 92mm fan has a higher wind speed than the the larger 140mm fans.  This is correct and partly to do with higher RPM of smaller fans, however it is not the complete picture.

The Wind Meter only captures the airflow in a circular area of 80mm diameter, so when sat next to a 140mm fan it is only in the path of about 30 – 40% of that air, compared to about 90% coverage of the 92mm fan.  Therefore even though the 140mm fan is producing slower air, it’s still producing a higher volume of air overall once you factor in the other 60 – 70% of air that the sensor is not measuring.

The chart should not be looked at from an ‘air speed’ point of view, but rather a comparison of how specific distances affects the fan’s ‘efficiency‘.

[/nextpage]