Need help picking a power supply

  1. #11
    Digerati is offline Senior Quiquagenarian

    Re: Need help picking a power supply

    With two of these mounted in a given enclosure (both mounted in exhaust mode), and, the enclosure has adequate (intake) ventilation holes/slots/orifaces/etc punched in it to enable 2.4 CFM of air flow, then that's what you will get. 2.4 CFM of air moving through that enclosure.
    Ah, I see the confusion. No, sorry, but that is not how it works - not in a power supply. We are not talking about two 80mm case fans sitting on the same back "plane" of the computer case, each with unrestricted flow out. In that case, I agree 100%, two fans, "in parallel" (like engines on an airplane's wing), doubles the total CFM out the back of the computer case (assuming adequate intakes to support two fans).

    But here we are talking about power supplies - two fans mounted on each end of a tube or air duct, blowing out the back of the PSU case. Fan1 on the intake side is providing intake to Fan2 on the output side. What you are suggesting is that you can mount any number (say 4, for example) of identical fans inside a tube, all blowing in the same direction and speed and the total CFM on the output side will be 1.2CFM x 4 or 4.8CFM. Or 8 fans for 9.6CFM. It does not work that way. If you put back pressure on a propeller (which is aerodynamically shaped to move air in only one direction) it stalls, there is no "lift" achieved across the blades so there can be no "added" CFM - Fan 2 would have to spin twice as fast to double the CFM. In the tube with 4 fans "in series", if CFM increased in that manner, Fan 4 would see 3.6CFM of air pressure on its back pushing it to run so fast, the bearings and housing might fly apart. Okay, that may be dramatic with little 80mm - 120mm DC fans, but the point is, any gain in total CFM on the output side of the air duct with fans "in series" is little, if any.

    "Identical" fans in series "keep" air flowing at a specified rate - unless you add some sort of accelerant in between, like jet fuel inside the combustion chamber of a jet engine, or more turning force from pressurized hot exhaust in a turbocharger of a car engine.

    The only time you "need" two fans in any power supply is if space does not allow for a single fan "big" enough to handle the heat removal needs - with "big" not just the length of the blade, but faster RPM (~horsepower!) + steeper fan pitch + greater blade width = more CFM with the same fan blade length. Since the only dimensions standardized by the ATX Form Factor for PSUs are length and width, the depth of the PSU can vary, leaving room for a "big" fan with deep pitched, wide blades, and strong enough motor to put out some torque, and serious CFMs.

    Beyond that, today's fan technology can easily meet the demands of most computer power supplies with only one fan. The exception may be with "monster" supplies that need forced air from the second fan blowing directly on to the rectifier heat sinks, rather than air flowing across as with single fan PSUs. Other than that, there is no real advantage to having two fans in a PSU (in a properly designed PSU, I should quickly add). It sounds good, but it is just more added parts (fan and circuitry to support it) that consume power intended for the real load. And any good PSU will shut down to self-protect itself and its load if it gets too hot, regardless the number of fans.

    My bottom line here is that the number of fans in a properly designed power supply is not a measure of PSU quality or cooling capability - it is the quality of the fan and its specifications that determine overall cooling efficiency.

    The real bottom line is
    Quote Originally Posted by Dan Penny
    ...get the best power supply you can afford. As long as it's ISA (Industry Standard Architecture), it can be moved to any future ISA enclosure/system you may build.
    Though actually it needs to meet "ATX" (or BTX) Form Factor standards for the case it will be installed in. These establish all necessary safety, size, mounting, and electrical voltage and current characteristics and tolerances. And the ATX Form Factor ensures any ATX power supply will mount in any ATX case and support any ATX motherboard, today, or in the future. That said, any ISA, UL or other safety certifications are always a good thing.


  2. #12
    Dan Penny is offline Techie7 Staff
    What you are suggesting is that you can mount any number (say 4, for example) of identical fans inside a tube, all blowing in the same direction and speed and the total CFM on the output side will be 1.2CFM x 4 or 4.8CFM. Or 8 fans for 9.6CFM.
    No, that's not what I was suggesting.

    two of these mounted in a given enclosure (both mounted in exhaust mode), and, the enclosure has adequate (intake) ventilation holes/slots/orifaces/etc punched in it to enable 2.4 CFM of air flow
    Which you seem to have understood here;
    I agree 100%, two fans, "in parallel" (like engines on an airplane's wing), doubles the total CFM out the back of the computer case (assuming adequate intakes to support two fans).
    But here we are talking about power supplies - two fans mounted on each end of a tube or air duct, blowing out the back of the PSU case. Fan1 on the intake side is providing intake to Fan2 on the output side.
    No, any two_fan power supplies I've ever worked with had one fan exhausting air out the "rear", and one fan exhausting air out of the "bottom" of the enclosure. ie; (both mounted in exhaust mode) The air flow is not "one in - one out", or, directed "one towards the other".

    In my references to "ISA", = "ATX" as ATX is todays standard. So it seems we're both saying the same thing (now that the basic dynamics have been elaborated upon), just with different terminology.

  3. #13
    Digerati is offline Senior Quiquagenarian
    No, any two_fan power supplies I've ever worked with had one fan exhausting air out the "rear", and one fan exhausting air out of the "bottom"
    No, something was wrong then, Dan. That would pump hot PSU air onto the CPU. Not good. And it would not promote air flow through the PSU, or the computer. I just powered up the two PSUs here with two fans, an Antec and a Seasonic, and both draw in from the bottom and exhaust out the back.

    Note the Form Factor's TAC Guide, which applies ISO standards to the ATX From Factor states
    The power supply fan must exhaust the internal chassis air outside of the system to assist with internal cooling.

  4. #14
    Dan Penny is offline Techie7 Staff
    Well, our experiences differ. This has dragged on far too long off topic.

  5. #15
    Digerati is offline Senior Quiquagenarian
    A timely test report: Seven affordable PSUs take on The Beast - The Tech Report - Page 1

    To get a handle on each PSU's true capacity, we have to determine the maximum output wattage of each voltage line. The math is easy enough thanks to Ohm's Law, which allows us to calculate wattage given voltage and amperage. However, power supply units with multiple rails are typically limited by how much power can be spread across those multiple lines—a total that's usually less than the sum of each rail's output capacity.
    And I note the Corsair 450W with a single 12V rail, took top honors, noting superior efficiency and quietness.

+ Reply to Thread
Page 2 of 2 FirstFirst 1 2