What are your opinions on PSUs with zero-rpm fan features?

[Jokii]

Don't see the point of zero-rpm mode PSUs, when you can just use a fanless PSU.

PSUs don't emit that much heat and they can easily be cooled with some case airflow. It's only an issue with sustained very high loads, which affects a small minority of PC users.

 

[Deleted member 185158]

Don't see the point of zero-rpm mode PSUs, when you can just use a fanless PSU.

PSUs don't emit that much heat and they can easily be cooled with a some case airflow. It's only an issue with sustained very high loads, which affects a small minority of PC users.

I think all that gives people what they need, and that is a choice. Can pick fanless, 0 rpm low/idle or like mine upgraded and always running.
Simply Because, I have a specific use for the PSU.

 

[Bill_Bright]

I apologize if I made you feel I was arguing.

Not at all. We're good.

It's only an issue with sustained very high loads, which affects a small minority of PC users.

I think it important to point out "very high loads" is relative. If the load is 300W and you have an 80% efficient 350W PSU, that's a very high load and the PSU's fan may need to spin at full speed (and loudness) to keep the PSU's components adequately cooled. If you have that same 300W load and a quality 90% efficient 750W PSU, the PSU's fan will likely be able to spin much slower (and quieter) or maybe no spin at all.

As a general rule, PSUs with higher capacities tend to have larger heat sinks. With the significantly greater surface areas for the fins, the bigger PSU does not need as much air flow.

PSUs don't emit that much heat and they can easily be cooled with some case airflow.

Depending on the case, that is true. But some cases have an isolated and enclosed (except for the intake and exhaust vents) chamber for the PSU. Case cooling in those cases does not affect PSU cooling. And in turn, PSU heat does not affect or contribute to heat inside the computer case.

 

[cvaldes]

These zero rpm PSUs certainly have a market, like institutional/enterprise/government use. Those are situations that many people in PC discussion forums (not just here at TPU) repeatedly ignore or discount.

It's not like this zero rpm fan feature is designed for every PC 24x7.

No one can please everyone all the time.

However some people do care about this stuff, like the US Federal government that has energy efficiency mandates for computing equipment. If a PC power supply unit can be adequately cooled by natural airflow, why burn up extra electricity running fans unnecessarily.

While that might seem insignificant to Joe Consumer and his super-ultra-buffed out gaming computer and its 3 GHz, 600W graphics card, it makes more sense to that IT manager at the General Accounting Office supporting ten thousand desktop PCs.

 

[Shrek]

If a PC power supply unit can be adequately cooled by natural airflow

I would guess that they rely upon the case fans and would not do so well in still air.

 

[cvaldes]

I would guess that they rely upon the case fans and would not do so well in still air.

My assumption is that the PSU engineers who design these features test them in a variety of conditions.

I still don't understand why a number of people in these kind of forums seem to think that these are designed by pulling some random schematics out of a hat. The people who design PSUs for a living have advanced degrees in electrical engineering, mechanical engineering, physics, etc. from technical programs at some of the top universities in the world. As well as years (sometimes decades) or experience.

Remember that a typical ATX PSU is a box with metal walls on four sides, one of which is directed to the outside. Case fans aren't going to generate much air circulation inside the PSU.

It's also worth pointing out that notebook "wall-wart" style PSUs don't have fans.

Pick up your smartphone. Then tell us you don't believe that electronics can be cooled without fans.

 

[Shrek]

The fan comes on when the internal temperature gets high, so it will be fine in still air, just not zero rpm.

Case fans will produce an over or under pressure in the case and this will induce a flow through the power supply.

A laptop power supply is around one tenth the power of a PC power supply and has the distinct advantage of outputting a voltage higher than 12V which helps reduce heat production.

 

[cvaldes]

The fan comes on when the internal temperature gets high, so it will be fine in still air, just not zero rpm.

Yes, and who programs fan curves in a PSU? Again, PSU engineers pick when fans come on, what temperature, what speed, what curve to follow, etc.

A laptop power supply is around one tenth the power of a PC power supply and has the distinct advantage of outputting a voltage higher than 12V which helps reduce heat production.

Really? What voltage is my USB-C charger outputting? The one that charges my Acer notebook via the Thunderbolt 3 port? Or MacBooks?

I know Apple marketed 85W power adapters for their more powerful notebooks. I'm guessing that's not the highest capacity external AC adapter on the market.

 

[Shrek]

I have one that is 24V, but I can't tell you what voltage yours are outputting.

Since there are 850W PC power supplies, the 85W fits the factor of 10.

 

[cvaldes]

I have one that is 24V, but I can't tell you what voltage yours are outputting.

Ah, one USB-C charger says 15V / 2A and 20V / 1.5A amongst other things. The same charger works with my iPad mini and iPhone.

Not sure about the other USB-C chargers. I've collected many over the years.

 

[Shrek]

Ah, it says 15V / 2A and 20V / 1.5A amongst other things.

So 30W supplies, not so hard to cool.

 

[Bill_Bright]

It's also worth pointing out that notebook "wall-wart" style PSUs don't have fans.

Or large (if any) heat sinks or vents.

Adding to your point about saving energy costs, if you've ever been in a large office area full of cubicles - especially those with shorter walls, even very quiet fans can make a lot of noise when there are dozens and dozens (counting case and CPU fans too) of fans spinning at once. And since the PSU is directly mounted to the back of the computer case, even a quality case does little to suppress a PSU's fan noise.

And while government bean counters have a duty to avoid wasting taxpayer's money, private sector bean counters worry about eating into profits and making money for their investors too.

A laptop power supply is a about one tenth the power of a PC power supply.

This is true but that does not mean it is not affected by heat. One must also remember a laptop power supply must be able to support running the laptop and charging the battery at the same time.

19V is typical

Nearly irrelevant because it is not really about the voltage - but the current/wattage along with efficiency. My 15" i5 MSI is rated for 90W. It would not be wise to toss something over it when running.

I don't see the point in discussing tiny USB chargers. We are getting way WAY OT now.

 

[cvaldes]

So 30W supplies, not so hard to cool.

Right.

So let's say a power supply uses 100W of power at moment A. It could be cooled without fans. That's what the zero fan rpm mode is for, right?

In the same way, if you can just open the window to cool your living room and keep the fan off, that's good enough, yes?

So if a 500W PSU has a 100W load and runs with the fans off, that's appropriate because the heat generated can dissipate without the need for a fan.

And ultimately, isn't this the same feature as a graphics card with a 0 dB fan mode? When the GPU fan(s) won't spin until the graphics chip temperature reaches a certain threshold?

 

[Shrek]

Nearly irrelevant because it is not really about the voltage - but the current/wattage along with efficiency.

Higher voltage means less current (for the same power) and so less losses.

So if a 500W PSU has a 100W load and runs with the fans off, that's appropriate because the heat generated can dissipate without the need for a fan.

I believe some zero rpm supplies can do 400W or more without their fan since they rely on the case fans.

 

[cvaldes]

I believe some zero rpm supplies can do 400W or more without their fan since they rely on the case fans.

The PSU fans kick on when they reach a temperature threshold, not the specific presence of case fans. And remember, the case fans themselves have their own performance, fan curves, whatever. Just having a fan running in the case is meaningless since it could be operated in a manner that doesn't significantly change airflow inside the PSU.

When running a Seasonic unit in hybrid on mode (with the fan grille directed up), the heat is actually rising upward from the grille opening. Ventilation in the case would move that heat away for the most part.

Again, PSU engineers try this in a variety of conditions but they mostly program the fan to come on based on an internal temperature threshold, not whether or not there's a case fan present.

 

[Bill_Bright]

Higher voltage means less current (for the same power) and so less losses.

No. In a power supply, efficiency determines loss. 90W is 90W. 80% efficiency is 80% efficiency.

90W x .8 = 72 with 18W lost in the form of heat. Note voltage is not in that equation. It does not matter if that 90W was achieved by 10V x 9A or 20V x 4.5A. Any differences in efficiency will be insignificant. Yes, there are exceptions but once again, exceptions don't make the rule.

Sadly, it is a common misconception that higher voltage results in better efficiency. That is just not true in the vast majority of circumstances. Twice the voltage will transfer twice the power, but the wattage consumed (and watts wasted) will be the same. And I note throughout the world, consumers pay for the watts (or kilowatts) consumed.

I believe some zero rpm supplies can do 400W or more without their fan since they rely on the case fans.

Sure! Why not? I mean Seasonic makes a 600W fanless PSU. And clearly, as you noted, it relies on air flow from the case cooling.

 

[Shrek]

Sadly, it is a common misconception that higher voltage results in better efficiency.

Then how do you explain that the same power supply is more efficient on 240V than 120V?

I'm just so tired of arguing the whole time.

 

[Bill_Bright]

I'm just so tired of arguing the whole time.

Me too.

I already said any difference will be insignificant.

But since you believe the whopping 1% difference from 90.5% to 91.5% at 50% load illustrated in your totally generic chart validates your claim and nullifies everything else, then fine! You win. Wire your home with 230V.

 

[Shrek]

10% less heat generated... and this thread is all about getting rid of the heat; the real issue is at high load when 20% less heat is generated (for the attached graph).

The reason that Tesla won the war of the currents is he could transmit at high voltage (using AC) while Edison could not (using DC), and the resulting losses were too high.

 

[Bill_Bright]

(for the attached graph).

How silly of me not to notice all that important information about heat in your graph. I already said you win. Funny how you have to keep arguing about it after saying you are tired of arguing.

The reason that Tesla won the war of the currents is he could transmit at high voltage (using AC) while Edison could not (using DC), and the resulting losses were too high.

That's not why. Tesla (Westinghouse actually) won because generating AC is much easier, less complicated and cheaper to produce and can much more easily used to transport energy over long distances.

 

[Shrek]

Exactly... 'much more easily used to transport energy over long distances'; and why might that be so?

Hint, power loss: P = I^2 R

One wants to get the current down (for a given power), and one does that by using high voltage; how does one get high voltage, by using a transformer, and that requires AC.


Yes, I am tired of arguing; just as I get tired at work, but that does not stop me from getting the job done.

I am sure you have much to teach me; I am here to learn, not argue, not to be right, not to win, but to learn.

Don't worry about missing the graph, we are here to figure things out together, not battle one ego against the other; I remain acutely aware that you will sometimes find error in what I say, and for that I thank you.

 

[Athlonite]

So all those reviewers, including this one, this one, and this one too are all wrong. Got it.

None of those reviews used a Zero rpm fan PSU Bill and that was my point you really need to have the fan going all the time otherwise you're looking at 45+ degrees celsius (depending on ambient temps) temps in your PSU at idle not something I want happening

 

[Bill_Bright]

None of those reviews used a Zero rpm fan PSU Bill and that was my point you really need to have the fan going all the time otherwise you're looking at 45+ degrees celsius

I realize that was your point. I was simply illustrating that, in most (as in non-extreme) scenarios, adequate cooling can easily be achieved without a fan - much or even all of the time. HTPCs used a PVRs commonly use totally passive cooling.

I don't know where you pulled 45°C at idle from. IMO and experience that is far from most realistic real-world scenarios. But yes, "IF" your ambient (room) temps are uncomfortably high (for us humans), then 45°C at idle might be realistic. Regardless, you seem to think 45°C (113°F) is something to worry about. It is not at all! Not in terms of safety. Especially for a power supply where mains voltages are present and doing a lot of work. Even heat sensitive electronic devices can easily handle 45°C and considerably higher without damaging or even long-term effects. Yes, output stability and regulation might start to temporarily degrade - though I note many of the better PSUs today are rated to operate up to 50°C. And we always recommend buying a quality PSU, right?

But if a user's ambient temps are routinely that high, get an AC! Or work (and especially play) in the cool of the night.

you really need to have the fan going all the time

And my point is, no you don't "NEED" to have the fan going all the time. At least not most users. This is another example of where users believe they know more than the highly educated and experienced electrical engineers and designers - folks who have decades and decades of established and widely published technical data, and computers to crunch the numbers, at their disposal.

You may personally "want" to have the fan going all the time. But that is totally different than "needing" it.

Do you really think any legitimate engineer, designer, manufacturer worth their salt is going to allow a PSU's temperature to reach critical levels before turning on the fan? Of course they aren't. Those fans are going to start spinning (at first, slowly and quietly) well before any damaging thermal thresholds are reached. If demands remain high causing the PSUs internal temps to rise further, the fan spins faster. Easy peasy.

Now "IF" the user is in such a high-temperature ambient environment where cooling fans "need" to spin full time, that is NOT a common scenario. And "IF" the demands on the PSU are such that, even in an air conditioned room, the fans "need" to spin full time, then perhaps the user/builder bought the wrong size PSU for the job, and should have gone with a more capable PSU to put the normal (most of the time) load/capacity ratio back down to where it belongs; closer to 50%. And that is on the user/builder - not the PSU, or the case.

not something I want happening

You clearly bought a top quality case, @Athlonite. And I applaud you for doing your homework and making an excellent choice. But it is important not to confuse (or to impose on to others) personal wants and desires with actual technical facts and needs. That's all I'm saying on this.

 

[Mussels]

Don't see the point of zero-rpm mode PSUs, when you can just use a fanless PSU.

PSUs don't emit that much heat and they can easily be cooled with some case airflow. It's only an issue with sustained very high loads, which affects a small minority of PC users.

They can when you throw an Ampere GPU on them or a modern intel or AM5 CPU

 

[Zach_01]

Don't see the point of zero-rpm mode PSUs, when you can just use a fanless PSU.

PSUs don't emit that much heat and they can easily be cooled with some case airflow. It's only an issue with sustained very high loads, which affects a small minority of PC users.

PC operation today is very dynamic in terms of performance and power draw. You can easily go with a factor of 10 between idle and full workload power. But even the majority of PC are at about the x3~5.
So when a PC is idling or doing very light stuff will consume around 100W (±20%). So at around 100W with a gold (90~91% efficiency) PSU the heat from losses of conversion will be around 10W. Easy...
When gaming it can go from 300W up to 500+W so then the heat from conversion is 30~50+W. The 40~50W cant be cooled passively without damaging the PCU components in the long run of years.


Enter zero fan rpm function.