A cheap passive am4 cooler?

[idkidk]

I was looking at the alpine passive am4 cooler

Alpine AM4 Passive | Silent CPU Cooler for AMD AM4 | ARCTIC

The ARCTIC Alpine AM4 Passive is a passive and thus completely silent cooling solution for AMD AM4 CPU.

www.arctic.de

But it's out of production and was stocked at -20€, fairly cheap, in the used market no luck at finding it

Anything else? I'm fine with used, but not fine at spending ≥50€ for a cooler

 

[Bill_Bright]

Not sure going with a budget "passive" cooler is wise. Passive, of course, implies no fan. As one who really hates fan noise, I can appreciate a CPU cooler that makes no noise at all. But CPUs generate a lot of heat. If there is no fan, there must be a lot of surface area for the fins of the heatsink. That is why passive coolers tend to be HUGE. And the better ones have high copper content for better conduction. More copper means higher costs.

You didn't say what CPU you have, but you need to check its specs and make sure any passive cooler can support it. That said, the heat radiating off the cooler needs to be moved out, so make sure your case is providing sufficient cooling - and sadly, that often requires case fans.

Check this out: Best Fanless CPU Coolers for Building Silent PC [Passive CPU Coolers] (graphicscardhub.com)

 

[AusWolf]

I have the Intel version of that cooler, and I can say that it only cools up to roughly 40-ish Watts. On a chiplet-based AMD CPU, this value is even less.

All in all, I wouldn't recommend going passive on AM4. If you absolutely want a passive system, I'd rather go with a T-series Intel CPU.

If you already have an AM4 system, and you're just looking for a cooler, a low-profile heatsink with a big fan (like the be quiet! Shadow Rock LP) should be quiet enough.

 

[Vya Domus]

If you are concerned about noise you are better off with an AIO and just limit the fan speed to an acceptable level.

 

[Bill_Bright]

If you are concerned about noise you are better off with an AIO and just limit the fan speed to an acceptable level.

Except that is just trading one noise for another. AIO coolers have pumps and fans often two or three fans - and the fans typically are mounted up against the case where their noise is not suppressed at all - as compared to a CPU fan deep inside the case.

If concerned about noise and keeping the CPU adequately cooled, buy a quality air cooler with a quality (thus quiet) fan. Then put them in a quality case that has quality case fans. Last, position the case off to the side and not on your desk and next to your ear.

just limit the fan speed to an acceptable level.

If someone does not like fan noise, there is no such thing as an acceptable level.

 

[BoggledBeagle]

A truly passive cooler must get pretty hot to start moving the air by just warming it up.

 

[evernessince]

A good passive cooler requires a lot of materials and thus is expensive. The NH-P1 or Le Grand Macho for example.

Even at the $100+ price point, performance is still going to trail that of a good $50 active CPU cooler. You are paying for the silence.

I'd say maybe look for used on the two above mentioned coolers but I would not buy anything smaller unless your use case is really low end, like under 45w. I took a quick glance at Aliexpress and it doesn't even appear that there are any cheaper alternatives.

 

[Bill_Bright]

A truly passive cooler must get pretty hot to start moving the air by just warming it up.

Ummm, no. It just needs to get warmer than the surrounding air.

 

[DaemonForce]

If you are concerned about noise you are better off with an AIO and just limit the fan speed to an acceptable level.

When I drafted up the update to obsolete my FX workstation and go AM4, this was my answer.
Picked up a 360cu high restriction Koolance radiator + shroud kit, cut out more of the top panel to fit, drilled some holes, installed everything and set the fan curve to an obscenely low dB that only goes up when I move the slider in Fan Control. AM4 chips, mainly ones like my 3600 in the 65-88W capacity do not need much. I really mean that.

You're not missing a thing.

 

[idkidk]

Not sure going with a budget "passive" cooler is wise. Passive, of course, implies no fan. As one who really hates fan noise, I can appreciate a CPU cooler that makes no noise at all. But CPUs generate a lot of heat. If there is no fan, there must be a lot of surface area for the fins of the heatsink. That is why passive coolers tend to be HUGE. And the better ones have high copper content for better conduction. More copper means higher costs.

You didn't say what CPU you have, but you need to check its specs and make sure any passive cooler can support it. That said, the heat radiating off the cooler needs to be moved out, so make sure your case is providing sufficient cooling - and sadly, that often requires case fans.

Check this out: Best Fanless CPU Coolers for Building Silent PC [Passive CPU Coolers] (graphicscardhub.com)

5600g 70w
I sorta miss a case also, no cheap solutions a cooler block that serves also as a case ? Currently I just packed everything in a wood stand to not have stuff all around, but I can't stand to the noise

I had the half idea to do something diy, but... I fucked up and I am tight on money, so I can't play too much

 

[Bill_Bright]

I'd definitely say that is an "airy" case!

 

[maxfly]

What restrictions are you looking at in regards to height, width etc? There are several Thermalright heatsinks well below $50 that work nicely as passive units. Ask our guy @freeagent for some help. He's our local Thermalright expert and will be able to nail down a few solid units for you.

 

[Count von Schwalbe]

A decent tower cooler like the PA120SE can be run nearly or fully passive. You can get even cheaper ones, but that one runs ~$35 over here in the states. The 5600G is a pretty cool-running chip, so you shouldn't have many issues with heat running the fan at nearly nothing.

I think @freeagent has tried running a PA120 passive on a 5600G or 5700G, with pretty good results. Most Thermalright heatsinks are very good value. In an open air "test bench" like you have there, passive heat dissipation is pretty good, but not quite as good as a case with a good airflow setup - although it is much quieter.

I would suggest running with your Wraith (or whatever you are using rn, it looks like a wraith) for a bit and playing with fan curves while monitoring temperatures in HWiNFO (its free) and seeing how quiet you can get it while keeping temperatures where you are comfortable. This will give you an idea of what you can do with a fan, and switching to a slightly better cooler would only help with that.

I would also suggest that swapping the stock fan on a cooler can do wonders for noise. I cannot remember how bad the Wraith fans are, but you can get something like an Arctic PWM fan and make it much quieter, while saving some money.

 

[Veseleil]

So this one won't be sufficient for a 5600G, unless you restrain it by undervolting or w/e is at hand.

 

[freeagent]

I have tried running semi passive on 5600X, 5800X3D and 5900X using PS120SE, and FC140. PS120 EVO will be able to do it too. I do not run my CPUs at stock, at the very edge I was able to run my 5900X with no fans installed at 180w PPT, but temps were in the 90s so 160w should be fine. I was using Linpack Xtreme to test..

I have not tried running PA120 semi passive, as I feel that is not their best cooler, though I have one on a 5600X right now.

I tested in a stock Fractal Torrent Compact setup, using just the two front fans that it came with.

 

[BoggledBeagle]

Ummm, no. It just needs to get warmer than the surrounding air.

You fail to understand, that tha passive heatsink needs to create sufficient air flow to carry away the heat. To get the air moving quickly enough, it needs to heat up to a pretty high temperature.

Normal air coolers for forced air movement are not good for passive use at all, because they have too dense fins - you can get air to squeeze through these narrow gaps only by actively pushing it by a fan.

 

[freeagent]

I have a Le Grand Macho RT that is rated for 90w passive, it can take it. With a good fan on it the rating changes to 320w. But it isnt the best for AM4, or really anything under 32nm.

 

[Bill_Bright]

You fail to understand, that tha passive heatsink needs to create sufficient air flow to carry away the heat. To get the air moving quickly enough, it needs to heat up to a pretty high temperature.

No, I did not fail to understand. You do not understand thermodynamics - nor what happens inside a computer case, plus you are making some invalid assumptions.

Cold is the absence of heat. Heat is energy. Heat moves into cold. So, if the surrounding air is cooler than the heatsink, the heat WILL move into the cooler air - even IF that difference is just a couple degrees.

Heated air is also lighter than cooler air. So heated air naturally rises. As long as there is room above for the warmer air to move away unrestricted (not get trapped and/or become stagnant), AND as long as there is a path for an adequate supply of cooler air to come in past the fins to replace that warm air, the heat will move away, keeping the fins from getting "hot".

The problem with your statement is, you are assuming the CPU is already creating "HOT" temps. You are also incorrectly assuming there is not enough surface area of the fins. A proper passive cooler - that is, one properly designed and one THE USER properly selected for that specific application, will work just fine when just "warm" when used in a proper case that provides good ventilation.

This also assumes the heatsink was properly mounted with a proper application of TIM.

"IF" the heatsink is getting "HOT", then (1) it is too small for the job and/or (2) there is not enough space around the heatsink to allow the heated air to freely move away and/or (3), there is not an adequate supply of cool air coming in.

 

[freeagent]

(4) The actual mating surface is so bad that no TIM will make it good. I have seen it on some of my own stuff. So bad in fact that I am sure a lap would not fix it heheh.

From what I have seen old school heat sinks, including some that have stayed fashionable by name do better with older CPUs.. looks like that may still apply, but to a broader range of products including AIOs and water blocks.

There are so many variables at play. My own home environment changes constantly with the furnace coming on, or the ac running.

 

[Bill_Bright]

(4) The actual mating surface is so bad that no TIM will make it good. I have seen it on some of my own stuff. So bad in fact that I am sure a lap would not fix it heheh.

LOL - This is very true. I too have seen some shoddy heatsinks like that. Sadly, 20+ years ago, both AMD and Intel had some OEM coolers like that - creating a reputation they are still trying to get away from.

And for sure, if a lap didn't fix it, time to toss it into the aluminum recycle bin and hope you get a couple cents of scrap metal value out of it.

I will add one thing to my comments above. There are some systems where the CPU fan does not even spin up because the heatsink is already radiating away from the CPU enough heat while barely warm - again pointing out the heatsink does not need to get "hot" before being able to do its job.

 

[BoggledBeagle]

Heated air is also lighter than cooler air. So heated air naturally rises.

Sure. Lets get away from your general statements and calculate instead, what force is acting on a body of air say 20°C hotter than surrounding air trapped in a volume 40x120x1,5mm. I just measured these dimensions on a traditional cooler. You can then recalculate this force using area of 120x1,5mm to pressure and compare this figure to a pressure created by a fan.

I can tell you the result: the pressure is MUCH SMALLER than the pressure created by the fan.

You know that there are things like an aerodynamic drag and skin friction , right?

You put these things together and you come to the conclusion I posted above.

Look, how wide are the gaps in the passive air cooler by Noctua.



Now you understand why!

And even they are so wide, the cooler needs to heat up anyway to move enough air. Check reviews of Noctus NH-P1 cooler and see, how hot is it getting.

 

[LabRat 891]

At best to my knowledge, you'll be looking @ "Semi-Passive". Meaning case still needs active flow.
My FSP Windale 6 atop my 3600 and 5600, was sufficient cooling w/ no fan attached. (Top fans pulled air thru it).

IMO, get a 'big tower' cooler (in black) that you can afford, and leave the fans off it. (just make sure the case is actively moving air)



BTW, Noctua makes what OP is looking for. (It's just anything but 'cheap')

Noctua NH-P1​

Normal air coolers for forced air movement are not good for passive use at all, because they have too dense fins - you can get air to squeeze through these narrow gaps only by actively pushing it by a fan.

Correct. However, if used in a chassis w/ active airflow (through the HS), many 'densely finned' heatsinks can function reasonably well "semi-passive".

In minimal airflow, there's something to be said for black, too. I've seen 'black' outperform 'uncoated' in multiple anecdotal (but repeat) tests (w/ intercoolers and basic alu heatsinks)

 

[SL2]

Can we stop writing essays about what thermodynamics is, and help the OP instead?

 

[Bill_Bright]

Can we stop writing essays about what thermodynamics is, and help the OP instead?

Would love to. But posting incorrect information is not helping the OP.

Sure. Lets get away from your general statements and calculate instead, what force is acting on a body of air say 20°C hotter than surrounding air trapped in a volume 40x120x1,5mm. I just measured these dimensions on a traditional cooler. You can then recalculate this force using area of 120x1,5mm to pressure and compare this figure to a pressure created by a fan.

You are not making any sense. You cannot randomly pull arbitrary numbers out of thin air, pretend they represent the whole, then use that to make moot the real point. And who says what the dimensions of a "traditional" cooler are? Coolers come in all shapes and sizes. Even OEM coolers come in all shapes and sizes these days.

But also, it is not even about the dimensions as much as it is about the total surface area of the fins, not to mention the composition of the metal. Copper, for example, conducts heat better than aluminum. The design of the heatsink matters. The use of heatpipes, for example, to more quickly move the heat from the block to the fins can make a significant difference.

The air is not "trapped". So why you are trying to suggest it is, is simply, and deceptively misleading. There is no reason any air should be trapped inside a computer case. Period. If it is, then it is time clean out the dust, or remove the cover you put over the heatsink or the case's vents.

the cooler needs to heat up anyway to move enough air. Check reviews of Noctus NH-P1 cooler and see, how hot is it getting.

Come on! This is just total nonsense. The cooler does not "NEED" to heat up. That is NOT how it works. If the temperature of the cooler is just 1° warmer than the surrounding air, heat WILL start to move into that air. The Laws of Physics, dude! Cold is the absence of heat. Heat is energy. Heat moves into cold.

What you are suggesting is if I put a monstrous heat sink on a CPU, a heat sink 10 times bigger and more efficient than it needs to be, because it will not get hot, it will not provide enough cooling for the CPU. Does that make sense? Of course not.

You are suggesting that Noctus cooler is only doing any good when it gets "hot". Total nonsense. It gets hot because the heat from the CPU has moved into it.

Now "hot" is a relative term. What I should have said above (my apologies for not doing so), is

"IF" the heatsink is getting "TOO" "HOT", then (1) it is too small for the job and/or (2) there is not enough space around the heatsink to allow the heated air to freely move away and/or (3), there is not an adequate supply of cool air coming in.

Now, in light of SL2's reminder, unless the OP has further questions, I agree it is time to move on.

 

[SL2]

Currently I just packed everything in a wood stand to not have stuff all around, but I can't stand to the noise

Please tell me that box has an intake on the side where the CPU fan is? Otherwise it's one more thing that makes a mediocre heatsink perform worse.

Also, do you want to get a better heatsink before getting (new) case? People usually do it the other way around, as there's a chance you get a heatsink that doesn't fit the case.