Overclock.net › Forums › Case Mods & Cases › Other Hardware Mods › DYI fan controller
New Posts  All Forums:Forum Nav:

DYI fan controller - Page 3

post #21 of 40
Quote:
Originally Posted by ARandomOWl View Post

I started making my own USB PWM fan controller a long time ago. I haven't got round to finishing it off yet though.

You should start a thread on it. (if you haven't already smile.gif )
post #22 of 40
Thread Starter 
Quote:
Originally Posted by jvjessen View Post

Good advice, 100nF on the input and maybe 10uF+ on output would probably not be a bad choice.
We should have more of these little construction logs here on OCN. Nice work thumb.gif

how much did you already saw here something fammilliar to theese mini logs? biggrin.gif
post #23 of 40
Thread Starter 
i was thinking about this one simple controller, and this question igot, if i wanna double the channels ( add 4 more) then ill need to take another 12v rail or i can use the same one? because then i can control up to 16 fans. Or... just add two more channels to volt regulations and add 4 or 2 channels for pwm
post #24 of 40
Quote:
Originally Posted by mr one View Post

i was thinking about this one simple controller, and this question igot, if i wanna double the channels ( add 4 more) then ill need to take another 12v rail or i can use the same one? because then i can control up to 16 fans. Or... just add two more channels to volt regulations and add 4 or 2 channels for pwm

I wouldn't recommend running more than 2 fans per regulator, e.g. around 8-12W load.

You can boost the regulators with power transistors, but then you'll need even bigger heat sinks.

Other options are adding more regulators or giving PWM a go, but it's got it's problems as well, it's always a compromise. Only almost ideal solution I can think of would be an adjustable buck converter design, but it gets a bit complicated and more expensive. How many fans do you need to control?
post #25 of 40
Thread Starter 
Quote:
Originally Posted by jvjessen View Post

I wouldn't recommend running more than 2 fans per regulator, e.g. around 8-12W load.
You can boost the regulators with power transistors, but then you'll need even bigger heat sinks.
Other options are adding more regulators or giving PWM a go, but it's got it's problems as well, it's always a compromise. Only almost ideal solution I can think of would be an adjustable buck converter design, but it gets a bit complicated and more expensive. How many fans do you need to control?

im thinking around 12-14, what i was saying in that post is: if i wanna use two of theese controllers, then i need second controller to plug to another 12v rail or i can use the same where is plugged the first one? biggrin.gif yeah something like that smile.gif or just get some advices about your controller and how to make this one biggrin.gif but its needs to be programmed yeah?

Something like that
450

OR

450
Edited by mr one - 5/16/12 at 2:19pm
post #26 of 40
Quote:
Originally Posted by mr one View Post

im thinking around 12-14, what i was saying in that post is: if i wanna use two of theese controllers, then i need second controller to plug to another 12v rail or i can use the same where is plugged the first one? biggrin.gif yeah something like that smile.gif or just get some advices about your controller and how to make this one biggrin.gif but its needs to be programmed yeah?
Something like that

Got it, misunderstood you the first time. upsidedwnsmiley.gif Yeah, it should be ok to run 2 controllers on the same 12V connection. Worst case for 16 fans would be probably 500mA per fan giving a total of 8 A. That will be ok as long as the wire is not too thin or long. Just remember that this many fans will add an extra load of 70-100W on your PSU.

Yeah, I guess you could say my controller build got a bit out of control smile.gif Had a lot of ideas and decided to put it all together into one piece. If you are interested I could probably come up with a few simpler constructions after I'm done with the current one.
post #27 of 40
Thread Starter 
Quote:
Originally Posted by jvjessen View Post

Got it, misunderstood you the first time. upsidedwnsmiley.gif Yeah, it should be ok to run 2 controllers on the same 12V connection. Worst case for 16 fans would be probably 500mA per fan giving a total of 8 A. That will be ok as long as the wire is not too thin or long. Just remember that this many fans will add an extra load of 70-100W on your PSU.
Yeah, I guess you could say my controller build got a bit out of control smile.gif Had a lot of ideas and decided to put it all together into one piece. If you are interested I could probably come up with a few simpler constructions after I'm done with the current one.

oh man it would be very nice from you to get something simplier smile.gif
post #28 of 40
Quote:
Originally Posted by jvjessen View Post

I wouldn't recommend running more than 2 fans per regulator, e.g. around 8-12W load.
You can boost the regulators with power transistors, but then you'll need even bigger heat sinks.
Other options are adding more regulators or giving PWM a go, but it's got it's problems as well, it's always a compromise. Only almost ideal solution I can think of would be an adjustable buck converter design, but it gets a bit complicated and more expensive. How many fans do you need to control?
I gave PWM a shot and had trouble with it dumping transients on the 12V rail due to the kickback from the fans being pulsed (think what happens when you suddenly cut current to an inductor). A buck converter is the best one imho, or even buck-boost if you want over volt your fans a bit. I got a relatively simple design I've been thinking of implementing just for proof of concept.
 
Cecilia 2.0
(15 items)
 
 
CPUMotherboardGraphicsRAM
FX-8120, and a bunch more AM3/AM2+ Crosshair V, Crosshair IV Lots of green Dominator GT 
RAMRAMHard DriveCooling
Dominator GT Perfect Storm M4 64Gb Dihydrogen monoxide, Carbon dioxide, Liquid Nit... 
CoolingCoolingOSPower
OCN Marksman pot k|ngp|n F1 Dark pot Stripped Xp Antec 1200W HCP 
Case
Dimastech Hard 
CPUMotherboardGraphicsRAM
2600K @ 4.8g's with lots of v's Maximus IV Extreme P67 580 SLI 1866mhz RipjawX 
Hard DriveCoolingCoolingOS
Intel ssd, some 1Tb F3s Di-hydrogen Monoxide I don't allow air cooling in my house Win7 
PowerCase
XFX 850 Phobya WaCoolT Test Bench 
  hide details  
Reply
 
Cecilia 2.0
(15 items)
 
 
CPUMotherboardGraphicsRAM
FX-8120, and a bunch more AM3/AM2+ Crosshair V, Crosshair IV Lots of green Dominator GT 
RAMRAMHard DriveCooling
Dominator GT Perfect Storm M4 64Gb Dihydrogen monoxide, Carbon dioxide, Liquid Nit... 
CoolingCoolingOSPower
OCN Marksman pot k|ngp|n F1 Dark pot Stripped Xp Antec 1200W HCP 
Case
Dimastech Hard 
CPUMotherboardGraphicsRAM
2600K @ 4.8g's with lots of v's Maximus IV Extreme P67 580 SLI 1866mhz RipjawX 
Hard DriveCoolingCoolingOS
Intel ssd, some 1Tb F3s Di-hydrogen Monoxide I don't allow air cooling in my house Win7 
PowerCase
XFX 850 Phobya WaCoolT Test Bench 
  hide details  
Reply
post #29 of 40
Quote:
Originally Posted by just_nuke_em View Post

I gave PWM a shot and had trouble with it dumping transients on the 12V rail due to the kickback from the fans being pulsed (think what happens when you suddenly cut current to an inductor). A buck converter is the best one imho, or even buck-boost if you want over volt your fans a bit. I got a relatively simple design I've been thinking of implementing just for proof of concept.

Yeah, I don't really like PWM in this use case, it's great for efficiency and simplicity, but you don't want to add noise on the 12V can't be good in this environment.

Good idea to do both buck and boost to allow for over volt on the fans. Would be fun to do some experiments and find a good solution. Wouldn't need to worry too much about ripple on the output etc, so it could be fairly simple. Maybe we should start a new thread, not polite to high-jack this one smile.gif
post #30 of 40
Thread Starter 
Quote:
Originally Posted by jvjessen View Post

Yeah, I don't really like PWM in this use case, it's great for efficiency and simplicity, but you don't want to add noise on the 12V can't be good in this environment.
Good idea to do both buck and boost to allow for over volt on the fans. Would be fun to do some experiments and find a good solution. Wouldn't need to worry too much about ripple on the output etc, so it could be fairly simple. Maybe we should start a new thread, not polite to high-jack this one smile.gif

no one is hi-jacking this thread biggrin.gif
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Other Hardware Mods
Overclock.net › Forums › Case Mods & Cases › Other Hardware Mods › DYI fan controller