Anyone done a Brocade 6610 fan mod?

Notice: Page may contain affiliate links for which we may earn a small commission through services like Amazon Affiliates or Skimlinks.

Dreece

Active Member
Jan 22, 2019
503
163
43
and the results?

An 'after' video would be great :D

ps I hear its not possible with a straight out rpm switch... but just wondering if someone has come up with an alternative solution?
 
  • Like
Reactions: Yunia

Dreece

Active Member
Jan 22, 2019
503
163
43
I'm thinking I might hang the switch in a small refrigerated cooler: o_O (there goes my marginally sound carbon footprint)
 
Last edited:
  • Like
Reactions: Originalus

Dreece

Active Member
Jan 22, 2019
503
163
43
Well most refrigerators aren't that quiet either
Total silence is but a dream, these things run super hot!

With all things being considered...

Turbo-jet hairdryer vs fridge? I think the fridge definitely wins. :cool:
 

Dreece

Active Member
Jan 22, 2019
503
163
43
ok I have come up with a spectacular solution, not only will it be kept cool, but noise should be brought right down to near silent, it is an old-skool method...

Will report back once it's done :D
 

nthu9280

Well-Known Member
Feb 3, 2016
1,630
504
113
San Antonio, TX
Love to hear what your solution is.

Do you have Rev A power supplies? They are known to be noisy. Rev C is the desired option. Even those are not going to be quiet by any means. Cooling ~120W during idle in a 1U space. Though the temperature tolerance is pretty high, It still has to keep things at manageable levels.
 

Dreece

Active Member
Jan 22, 2019
503
163
43
Soon, very soon!

Revision? I'm not sure, when I rip it out tomorrow I'll have a look. Unless we can tell from the shell?

The good thing about my one is that it was virtually brand new, I think it was ex-overstock at the medium size business, I collected it direct from their network engineer.
 

Skippy96

New Member
May 14, 2019
19
7
3
havent seen anything on this thread in a while. from reading the main brocade thread:

1. the main noise is coming form the power supplies
2. the rev A and B power supplies are louder than the rev Cs
3. if you have only one power supply, the fan will be louder in the single supply than if you have two plugged in
4. the 6610 has a built in monitor that checks fan speeds at boot, if it doesn't see what it likes it wont boot - VERY DIFFERENT THAN THE 6450

so it looks like the only way to swap fans is to fake the monitor into thinking the RPMs are the same.

Someone posted this way to create a fake fan but not sure if anyone has tried it:
Fake Fan RPM Sensor — www.TechIdiots.net - www.ElectronicInsanity.com

im on page 64 of 94 so hopefully someone read all 94 pages and can tell is if it works or if there are any other ideas please post!!
 

Skippy96

New Member
May 14, 2019
19
7
3
ok so I'm at my technical limits here guys so if anyone can help please reach out:

i got the pulse generator modules from Amazon and tested them on my pc and they work fine, they are outputting ~8,000RPM based on the pot setting.

i also took the power supply apart. the pinout for the fan connector seems pretty basic:

1.red - LED power
2.green - LED GND
3.yellow - FAN RPM sensor
4.brown - FAN PWM control
5.black - FAN GND
6.orange-FAN PWR

these are the tech specs of the fan i was able to find online (9GV0412J301):

https://www.mouser.com/datasheet/2/471/San_Ace_40GV28_E-1286218.pdf
https://www.sanyodenki.com/archive/document/product/cooling/DC_PWMcon_en.pdf

it looks like the fan "could" run at lower RPMs by sending a lower PWM control signal.

could i do this just by soldering a resistor in between the connections?

thanks all.
 

Attachments

Last edited:
  • Like
Reactions: MarcoP82

lambda

Member
Oct 27, 2018
34
12
8
ok so I'm at my technical limits here guys so if anyone can help please reach out:

i got the pulse generator modules from Amazon and tested them on my pc and they work fine, they are outputting ~8,000RPM based on the pot setting.

i also took the power supply apart. the pinout for the fan connector seems pretty basic:

1.red - LED power
2.green - LED GND
3.yellow - FAN RPM sensor
4.brown - FAN PWM control
5.black - FAN GND
6.orange-FAN PWR

these are the tech specs of the fan i was able to find online (9GV0412J301):

https://www.mouser.com/datasheet/2/471/San_Ace_40GV28_E-1286218.pdf
https://www.sanyodenki.com/archive/document/product/cooling/DC_PWMcon_en.pdf

it looks like the fan "could" run at lower RPMs by sending a lower PWM control signal.

could i do this just by soldering a resistor in between the connections?

thanks all.
I don't know whether this will work but you could try to connect your pulse generator output to the PWM input of you fan. Assuming your pulse generator is generating a square wave with 50% duty cycle. your fan should spin at the speed close to the midpoint between the maximum speed and the minimum speed of you fan profile.
 

Skippy96

New Member
May 14, 2019
19
7
3
I don't know whether this will work but you could try to connect your pulse generator output to the PWM input of you fan. Assuming your pulse generator is generating a square wave with 50% duty cycle. your fan should spin at the speed close to the midpoint between the maximum speed and the minimum speed of you fan profile.
So are you saying by doing this I would show the power supply a fan RPM speed of 8,000 rpm but have the fan running at 4,000 RPM?
 
  • Like
Reactions: Airz

lambda

Member
Oct 27, 2018
34
12
8
So are you saying by doing this I would show the power supply a fan RPM speed of 8,000 rpm but have the fan running at 4,000 RPM?
rpm = (max_rpm - min_rpm) * duty_cycle + min_rpm
If we assume the following:
max_rpm = 14700
min_rpm = 2650
duty_cycle = 0.5

The rpm should be:
rpm = (14700 - 2650) * 0.5 + 2650 = 8675 rpm

You could also try to set the duty cycle to 0% by just connecting the fan PWM pin to ground:
rpm = (14700 - 2650)*0 + 2650 = 2650 rpm
 
Last edited:
  • Like
Reactions: MarcoP82

bkvamme

New Member
Sep 23, 2018
11
5
3
Hi,
I doubt that it would work (could be wrong). The PWM signal to the fan is sent at a frequency of 25kHz and have a voltage between 0-5V, and from what I can see of a typical RPM signal it is pulled low two times per revolution (8000 RPM = 16000 pull low/min = 267 / second), which is far from sufficient to generate the required PWM signal.

An alternative for generating the PWM signal is to set up an Arduino to do it: Set PWM frequency to 25 kHz

Easiest and probably cheapest way would be to get one of these: https://www.amazon.com/Noctua-NA-FC1-4-pin-PWM-Controller/dp/B072M2HKSN.

Alternative: https://www.amazon.com/Wire-PWM-Cas...ntroller&qid=1561112914&s=electronics&sr=1-18

edit: Alternatively, you could replace all of the fans with ones that run a bit quiter, but that will cost a wee bit.
 
Last edited:

kur1j

New Member
Aug 12, 2019
10
3
3
Hi,
I doubt that it would work (could be wrong). The PWM signal to the fan is sent at a frequency of 25kHz and have a voltage between 0-5V, and from what I can see of a typical RPM signal it is pulled low two times per revolution (8000 RPM = 16000 pull low/min = 267 / second), which is far from sufficient to generate the required PWM signal.

An alternative for generating the PWM signal is to set up an Arduino to do it: Set PWM frequency to 25 kHz

Easiest and probably cheapest way would be to get one of these: https://www.amazon.com/Noctua-NA-FC1-4-pin-PWM-Controller/dp/B072M2HKSN.

Alternative: https://www.amazon.com/Wire-PWM-Cas...ntroller&qid=1561112914&s=electronics&sr=1-18

edit: Alternatively, you could replace all of the fans with ones that run a bit quiter, but that will cost a wee bit.


I assume I would need one of the Noctuas for each fan so a total of 3 in my case? In my 6610 I have 3 fans. 1 for the PSU and I assume two generic fans.