7 Volt Mod and PWM at same time?

Zarathustra[H]

Zarathustra[H]

Extremely [H]
Joined
Oct 29, 2000
Messages
38,862
Hey all,

Does anyone know if this works (and doesnt ruin anything? :p )

I successfully done a 5 volt mod and PWM at the same time in the past, such that the 5V speed is lower, and you still maintain PWM control of that lower speed.

I've done this by making my own adapter, where +5v and ground come from a molex plug, and the motherboard gets the rpm and pwm connections.

Question is, is this possible with the 7 volt mod?

The reason I am concerned is that a 7 volt mod uses +5v as ground.

The PWM signal only drives up to 5v, and if it expects to have a ground to sink to, it could do anything from just not work, to actually destroying the motherboard.

Anyone have any thoughts on this?
 
Its a waste of time. Just get pwm fans that are geared at a lower rpm if you want the rpm to go that low.
 
ZodaEX said:
Its a waste of time. Just get pwm fans that are geared at a lower rpm if you want the rpm to go that low.
Click to expand...

I have good fans (140mm Noctua iPPC-2000 PWM). They are rated from 500-2000rpm via PWM control which is great. In practice they actually go all the way down to 0 RPM at 0 duty cycle and will spin below the 500 rpm at lower duty cycles as well.

Very impressed by these fans.

Not so impressed by my motherboard BIOS though.

I was a little annoyed by the CPU fan control which requires a minimum duty cycle of 20%, but I can live with that, as they are very quiet at 20%.

The Case fan control - however - has a minimum PWM setting of 60% (!?)

I presume the 60% was chosen as the minimum for voltage control, to make sure that the fans would actually still spin at the lowest setting, but unfortunately even when you switch it over to PWM it won't let you set anything below a 60% duty cycle.

I don't understand why these things need limits at all. It's my build, if I want to set it low, I should be able to.

Anyway, so according to my calculations, if I drop to 5v, a 60% PWM signal should wind up being ~500rpm, and a 100% PWM signal should wind up being ~833rpm.

This might wind up being enough for just case fans, as these fans move a lot of air at most RPM's, but I am trying to figure out if doing a 7v mod is actually an option.
 
Zarathustra[H];1042004216 said:
I have good fans (140mm Noctua iPPC-2000 PWM). They are rated from 500-2000rpm via PWM control which is great. In practice they actually go all the way down to 0 RPM at 0 duty cycle and will spin below the 500 rpm at lower duty cycles as well.

Very impressed by these fans.

Not so impressed by my motherboard BIOS though.

I was a little annoyed by the CPU fan control which requires a minimum duty cycle of 20%, but I can live with that, as they are very quiet at 20%.

The Case fan control - however - has a minimum PWM setting of 60% (!?)

I presume the 60% was chosen as the minimum for voltage control, to make sure that the fans would actually still spin at the lowest setting, but unfortunately even when you switch it over to PWM it won't let you set anything below a 60% duty cycle.

I don't understand why these things need limits at all. It's my build, if I want to set it low, I should be able to.

Anyway, so according to my calculations, if I drop to 5v, a 60% PWM signal should wind up being ~500rpm, and a 100% PWM signal should wind up being ~833rpm.

This might wind up being enough for just case fans, as these fans move a lot of air at most RPM's, but I am trying to figure out if doing a 7v mod is actually an option.
Click to expand...

Do you run Windows? If so i would invest in an aquacomputer fan controller. Dont be at the mercy of your particular motherboard's fan controller. :) this controller can handle everything you are wanting to do and more. Since ive purchased and set one up i can no longer ever hear my computer anymore after logging into windows. Even during heavy gaming.
 
ZodaEX said:
Do you run Windows? If so i would invest in an aquacomputer fan controller. Dont be at the mercy of your particular motherboard's fan controller. :) this controller can handle everything you are wanting to do and more. Since ive purchased and set one up i can no longer ever hear my computer anymore after logging into windows. Even during heavy gaming.
Click to expand...

I've looked at those every couple of years, been very impressed, but in the end not pulled the trigger, as it seems like an awful lot of money for a fan controller.

German engineering I guess :p

I didn't realize they were Windows only though. I dual boot to Linux, so that probably would not be an option for me then.
 
Zarathustra[H];1042005017 said:
I've looked at those every couple of years, been very impressed, but in the end not pulled the trigger, as it seems like an awful lot of money for a fan controller.

German engineering I guess :p

I didn't realize they were Windows only though. I dual boot to Linux, so that probably would not be an option for me then.
Click to expand...

Oh right technically they are not Windows only if you are willing to stick physical hardware temperature probes on all of your components you want monitored, and use the fancy deluxe model with a screen and remote than can be controlled/programmed directly threw the fan controller it's self. What I do is buy the super basic lowest end model with no screen, and use their windows utility to pass along temperature information from "open hardware monitor" (a windows program) threw the motherboard's internal usb header, to the fan controller it's self. Then at that point you can use their windows utility to program the fan controller threw a software GUI. The thing is, this controller is sooo endlessly programmable. I have highly efficient heatsinks and a vented top on my pc's case so amazingly, at idle my entire machine produces so little heat that all my fans are programed to turn themselves off whenever all components are under 45C. No hard drives either so my computer has no moving parts when idle. My CPU fan turns on at 200 RPM once the temp hits 45C and speeds up linearly with temperature increase. Same for my GPU except it turn on at 50C and will spin at max RPM at the 80C point but this had never even come close to happening. One other really neat side effect of this system, is since the fans turn themselves off at idle, my computer doesn't collects dust anymore since I don't have fans constantly puling dust into the system.





 
based on http://www.formfactors.org/developer\specs\4_Wire_PWM_Spec.pdf there are two things to focus on in page9.

1) the tach output signal is drained/collected to ground by the fan (the fan's ground). The motherboard is responsible for pulling up this signal to 12v. The mobo presumably uses some logic high/low threshold voltage near ground to count the pulses for the tach reading.

2) The PWM control input signal is drained/collected to ground by the mobo. The fan pulls this line up to 5V MAX. The fan uses a logic high/low of 0.8V on this line to determine if the fan should be on or off for the current PWM signal.


If you give the fan a "ground" value of +5V and feed it +12V on the power line I can forsee the following problems.

1) Tach problems: The fan will drain/collect the line to +5V instead of near 0V. This likely means the tach measurement will be screwed up (0 rpm I think, not sure). I don't see electrical damage being a problem as the mobo powering the line expects between 12v and 0v and 5v is still within this range.

2) PWM control problems: Since the fan is being fed +5V as ground and +12v as power, instead of bringin the PWM control line to a maximum of 5V the fan will probably bring this somewhere between 12V and 5V. This is outside the spec and could cause electrical damage after prolonged use.

Furthermore, when the mobo drains/collects this line to 0V to create the PWM signal, this will be -5V with respect to the fan's "ground". This is not stricly out of spec, but I'm assuming most modern fans have some basic controller IC's on them, and I've always been taught that digital inputs to an IC should generally never be outside the voltage range of the ground and power. Not sure how dangerous that one is.
 
You could get a PWM splitter and run all of your fans off of the CPU PWM signal. That would be a cheap and easy fix perhaps.
 
Erasmus354 said:
You could get a PWM splitter and run all of your fans off of the CPU PWM signal. That would be a cheap and easy fix perhaps.
Click to expand...

That's a pretty sloppy and imprecise way to do it. To each his own I suppose.
 
serpretetsky said:
There are some cheap PWM fan controllers on ebay
http://www.ebay.com/itm/DC-12V-PWM-...450656?hash=item2ee81909e0:g:vhQAAOSw14xWPHRV

never used them, so i can't comment on how they work.
Click to expand...

Nice, I've never come across those before.

I've always found the drive bay fan controllers sold at computer retailers lacking, but these seem pretty cool, and are dirt cheap too.

Not sure if I'll wind up using one, but I may just buy a couple at this price just to have handy.
 
As an eBay Associate, HardForum may earn from qualifying purchases.
Zarathustra[H];1042004216 said:
~
~
I was a little annoyed by the CPU fan control which requires a minimum duty cycle of 20%, but I can live with that, as they are very quiet at 20%.

The Case fan control - however - has a minimum PWM setting of 60% (!?)

I presume the 60% was chosen as the minimum for voltage control, to make sure that the fans would actually still spin at the lowest setting, but unfortunately even when you switch it over to PWM it won't let you set anything below a 60% duty cycle.

I don't understand why these things need limits at all. It's my build, if I want to set it low, I should be able to.

~
~
Click to expand...

"I don't understand why these things need limits at all."

Agreed, but there have been improvements in the way they've gone about trying to save us from ourselves :p. Example, my Skylake board allows a minimum 20% duty cycle for both CPU and case fan headers in either PWM or DC Mode except when CPU headers are in DC Mode. Minimum is still limited to 60% in that case.

I suspect AMI is calling the shots in Asus' case. Example, AMIBCP allows me to change the lower PWM default threshold below 20% for CPU headers in my Sandy Bridge board's BIOS but won't let me go below 60% for case headers.
 
Zarathustra[H];1042008965 said:
Nice, I've never come across those before.

I've always found the drive bay fan controllers sold at computer retailers lacking, but these seem pretty cool, and are dirt cheap too.

Not sure if I'll wind up using one, but I may just buy a couple at this price just to have handy.
Click to expand...

Local retailers, if you are in the U.S. absolutely I agree. That's why you've gotta import the good stuff from other countries!
 
ZodaEX said:
That's a pretty sloppy and imprecise way to do it. To each his own I suppose.
Click to expand...

I was referring to the problem with the CPU PWM signal on the motherboard allowing 20% duty cycle and the Chassis PWM allowing only 60%. You can get a PWM splitter that allows you to simply and easily run all your fans from the CPU PWM signal. Not a sloppy or imprecise way to do it at all. Certainly a lot more practical to use a $9.99 splitter than a $199.99 Aquaero.
 
Erasmus354 said:
I was referring to the problem with the CPU PWM signal on the motherboard allowing 20% duty cycle and the Chassis PWM allowing only 60%. You can get a PWM splitter that allows you to simply and easily run all your fans from the CPU PWM signal. Not a sloppy or imprecise way to do it at all. Certainly a lot more practical to use a $9.99 splitter than a $199.99 Aquaero.
Click to expand...

Practical? Yes. Imprecise? Extremely imprecise compared to an Aquaero.
 
You must log in or register to reply here.
Back
Top