Is there a reason for the 25% limit for the minimum speed of the EON fan?

Hope someone from Argon will read this and can provide an answer (please).

The noisy fan of the EON is causing quite some discussion about replacements parts and so on, so there seems to be quite some desire to have a solution.
I would be glad to have an option, that the fan blows some fresh air all the time with low speed into the case. However the script controlling the fan will adjust the level of the fan to be not lower than 25%, which is very loud. The relevant part of the code can be found easily within the script, however the question for me is:
What is the reason for this limit? Will it cause any problems, in case i change it to 10% or even 5%?

I haven’t tested-but the voltage may simply be too low when below 25% to even have the fan turn? I know with ASUS motherboards when they try to tune the fans, they will check the threshold for the point that a fan will not spin.

1 Like

Thanks for the feedback @troyfontaine . I did a simple check already, modified the script and even with 10% the fan is working well, while on the other side it is also still noisy :frowning: This is maybe the most irritating part to me. The difference of noise level between 100% and 10% is not that much, at least not what I would expect…
To get noise down the easy way was releasing the side panels: Without panel the fan is almost silent. So the noise might come from the air streaming through the wholes in the case. The more important it would be to have a flow with minimal speed.

In that case, then the noise must come down to needing vents in the side panels, the EON has a certain aesthetic that doesn’t really lend itself to airflow. I have gone as far as to put mine in an unoccupied room because of the noise. Thanks for checking!

I upgraded my power supply & fan to 25mm thicker/deeper, so little a more noisier but in my opinion it gives a more efficient air flow, you can set fan profile based on drive bay temperature rather than CPU temp. I find it more functional this way. the CPU & GPU heat is dissipated through the two metallic pylons at the bottom, so theres no theoretical reason to use fan profile based on CPU sensing temp.

My fan even works with 1% (after modifying the scripts). And only on this value the noise is acceptable, but enough to cool the case down if only my SSD is running. YMMV,

I’ve done some modifications to the cooling setup of the EON, which makes fan noise fairly acceptable imho.

  • I’ve swapped the fan with a Sunon MF60101V3-1000U-A99
    This has lower RPM (@12V) as the original fan. Since its mounting holes are only 4mm, it sits on top of the fan mounting studs - that does not yet cause problems with 3 1/2" drives.
  • To increase air flow through the enclosure, I’ve made a simple air guide, which makes sure the fan exaust air actually is routed out of the EON:

    The fan sits in a cutout of the ABS sheet, which is fixed to the top cover by scotch tape.
  • The fan control itself suffers from a less-than-optimal design of the fan control circuit. This is fairly similar to the one being shown for the One v2 box. The most important difference is the fan operating voltage - the EON fan runs off 12V. Also the output capacitor seems to be ~7,5uF for the EON (measured with a cheap multimeter).
    The EON fan control is located in the display/button assembly.
    As shown in the linked One v2 post, the problem is the somehow inadequate averaging circuit consisting of a single capacitor being discharged very quickly by the MOSFET. Ideally, one would use an LC filter to get a really averaged fan voltage and reduced ripple. Since I had no suitable inductance (should be aroung 1mH), I just added a 10 Ohms resistor between the Drain of the MOSFET and the output capacitor/diode. For this, I’ve lifted the Drain pin off the circuit and attached a wire to it.

    (This modification does require some general understanding of electronics, a fine soldering iron and a steady hand)
    As a result, there’s less ripple on the fan voltage, and a wider range of pulse widths can be used. In my setup I measured:
PWM    100%   50%   20%   10%    9%    8%    7%    6%    5%    4%    3%    2%
U_fan  11.6V 11.1V 10.0V  8.5V  8.2V  7.9V  7.5V  7.0V  6,5V  5.8V  4.9V  3.9V

That’s far from ideal, but much better than with the original circuit.
With the 3% setting, the Sunon fan starts up reliably, although the data sheet specifies a minimum startup voltage of 6V. With 2%, startup does not work reliably (although the fan keeps running).

  • Finally, I updated the fan control configuration (/etc/argoneon.conf):
[CPUFan]
53.0 = 04
56.0 = 08
60.0 = 18
65.0 = 100

[HDDFan]
38.0 = 03
40.0 = 04
43.0 = 05
46.0 = 06
48.0 = 07
50.0 = 09
53.0 = 10
56.0 = 25
60.0 = 100

As a result, at almost 26°C room temperature, fan control toggles between ‘off’ and 3%. Noise level is noticeable in 1m distance, but certainly not annoying (for me). Btw.: the noise level is much lower with side panels removed (not noticeable at 3% or 4%) - not quite sure yet why these have such a huge (bad) effect…