Silencing PWS-741P-1R power supply

Arseni Mourzenko
Founder and lead developer
March 13, 2021
Tags: hardware 10 short 48

Disclaimer: the following article explains an unauthorized modification of an electrical appliance which runs at high voltage and contains a lot of chemical stuff which doesn't play nicely at high temperatures. The manipulation explained in the article voids the warranty and modifies the product in a way which wasn't tested by the manufacturer. Unless you have enough experience, DO NOT ATTEMPT TO DO THE SAME MANIPULATION WITH A SAME OR A SIMILAR DEVICE. Not only you'll void the warranty, but you would be at risk of permanently damaging the product or compromising your own safety.

I have a bunch of Su­per­Mi­cro 2U chas­sis 825TQ that I use for the tasks which don't re­quire a lot of CPU pow­er.

The orig­i­nal chas­sis makes a lot of noise. When I say a lot, I mean it. You can't pos­si­bly stay in the room when this thing is on. I orig­i­nal­ly re­placed the three 80×80 fans, which made the chas­sis way more qui­et that it used to be. Once I did it, the next loud­est source of noise be­came the small 80×80×28 fans of the re­dun­dant pow­er sup­ply. So it was time to re­place those ones as well.

Note that:

It is for those rea­sons that I took the risk to re­place the fan: if some of the fac­tors were dif­fer­ent, I be­lieve that tak­ing the risk of tam­per­ing with a pow­er sup­ply wouldn't be a good idea. It should be not­ed that the orig­i­nal fan is rat­ed at 40 m³/h, while the one I will be us­ing is rat­ed 9,4 m³/h, which is a sub­stan­tial dif­fer­ence.

Choos­ing the fan

Un­like with the re­place­ment fan for an UPS, I didn't have much choice here in terms of the fans: all fans with high sta­t­ic pres­sure are noisy, so for qui­et op­tions, Noc­tua NF-A4x20 PWM seems to be the only op­tion. I took the 4-wire (PWM) vari­ant, be­cause the orig­i­nal AVC DB04028B12U-128 fan is a PWM fan.

Fig­ure 1 The video show­ing the noise made by the two fans.

Wiring the fan

Fig­ure 2 The orig­i­nal AVC DB04028B12U-128 fan.

Fig­ure 3 The fan con­nec­tor.

The orig­i­nal fan uses a cus­tom con­nec­tor, so some sol­der­ing was nec­es­sary. By the way, the col­ors of the wires were all wrong: usu­al­ly, PWM fans use black-yel­low-green-blue wiring (A.), or black-red-yel­low-blue vari­ant (B.), there­fore the blue wire is al­ways the con­trol one; this fan, how­ev­er, was us­ing black-red-blue-yel­low col­or­ing (C.), that is, the blue wire was a tachome­ter sig­nal one.

Fig­ure 4 The stan­dard col­or codes when wiring the fans.

I didn't know that, so I sol­dered the wrong wires to­geth­er, and was wel­comed with an am­ber LED light of the pow­er sup­ply, in­di­cat­ing that a fault pre­vents it from start­ing. Found the is­sue; sol­dered the cor­rect ones, but this time, the fan was start­ing for a frac­tion of a sec­ond, then stop­ping, and the pow­er sup­ply LED would re­main am­ber.

Fi­nal­ly, I fig­ured out that the prob­lem was the dif­fer­ence in the speeds of the two fans. The orig­i­nal one is rat­ed 13.000 RPM, while Noc­tua's is 5.000 RPM. Giv­en that the pow­er sup­ply is un­der­used, it en­sures the fan turns at its min­i­mum speed: the orig­i­nal fan did ex­act­ly that, while the new one just stopped, which alert­ed the pow­er sup­ply through the tachome­ter that some­thing went wrong.

The so­lu­tion was to dis­con­nect the PWM ca­ble. This means that the Noc­tua fan is turn­ing now at its max­i­mum speed, but this shouldn't be a prob­lem: at max­i­mum speed, it is still much qui­eter than the orig­i­nal fan at its min­i­mum.

Fig­ure 5 The match­ing of cus­tom col­ors with the stan­dard ones.

Fig­ure 6 Noc­tua NA-EC2 ex­ten­sion ca­ble cut to be sol­dered with the orig­i­nal ca­ble.

Fig­ure 7 NA-EC2 sol­dered to the orig­i­nal ca­ble.

Run­ning it

Af­ter I did the change for one of the two PSUs, I let it run for a few days, just to make sure it works cor­rect­ly. The PSU didn't com­plain, and the ex­haust air tem­per­a­ture (mea­sured with a tem­per­a­ture probe po­si­tioned at the back of the serv­er) didn't show any­thing sus­pi­cious. I then re­placed the fan in the sec­ond PSU.

Fig­ure 8 Con­nect­ing the fan to NA-EC2.

Fig­ure 9 Noc­tua NF-A4x20 PWM.

Fig­ure 10 The fan, with cus­tom screws.

What's nice is that this 2U chas­sis is now one of the qui­etest de­vices out there. I can't hear it when I turn it on, giv­en the am­bi­ent noise pro­duced by the pri­ma­ry serv­er (with its sev­en fans), the UPS de­vices, and the switch.