Best Pi 3 case with 40mm+ fan mounting?

So far the best I've personally tried is the Solarbotics Pi S.A.F.E. V3 .

But it's still not perfect, with rather small gaps for fan airflow, and the included screws are a very tight fit. But I like it well enough, and it looks good sitting next to my TV and the like.

Still, I wonder if there's not a better choice that is both reasonably priced, and fully encloses the Pi.

Anyone?

BTW - I need active cooling because I run BOINC 25/7 when I'm not otherwise using my Pis. I hate to waste CPU time :-)

And I need to use a 40mm+ fan, because I've yet to find a quiet smaller fan. I wish I could find a quiet 30mm fan, as cases with 30mm fan mountings seem to be incredibly common, vs larger mounting sizes.

--
If you're not part of the solution, you're part of the precipitate.
Reply to
Jamie Kahn Genet
Loading thread data ...

I can't resist asking why a fan is needed on an RPI2 or RPI3. Mine don't seem to get more than lukewarm.

Inquiring minds and all that....

bob prohaska

Reply to
bob prohaska

Running the CPU of a Pi 3 maxed constantly, easily overheats it, forcing throttling. Which of course is to be avoided when your goal is to crunch BOINC work units during downtime. Or play the heftier games in RetroPie without slowdowns, as I often do :-)

Plus I like to prolong the life of my hardware, and I find running it cooler helps with that too.

--
If you're not part of the solution, you're part of the precipitate.
Reply to
Jamie Kahn Genet

On 2017 Jan 30 00:27:40, you wrote to Jamie Kahn Genet:

bp> What OS and clock speed are you using? I'm using Raspbian Jessie. bp> There is no visible option for overclocking. Under those conditions it bp> just barely gets warm

i don't think anyone said anything about overclocking... what was said was that running the CPU at full usage with no idle time brings the heat up to the point where a fan is needed...

so here's the deal, CPUs spend roughly 90+% of their time waiting on something to do... tools like BOINC and distributed.net are designed specifically to take advantage of that idle time and to use it to the CPUs full power... in other words, there is zero idle time because they are using all of it...

)\/(ark

Always Mount a Scratch Monkey Do you manage your own servers? If you are not running an IDS/IPS yer doin' it wrong... ... We are scientists, get outta our way!

Reply to
mark lewis

What OS and clock speed are you using? I'm using Raspbian Jessie. There is no visible option for overclocking. Under those conditions it just barely gets warm

Thanks for reading!

bob prohaska

Reply to
bob prohaska

Yes indeed.

It's unlikely anyone not running apps that push the CPU to high usage for extended periods, would need active cooling on any model Pi. But maxing CPU usage (and sometimes GPU too with certain console emulation) for extended periods is precisely what I am doing :-)

Try it yourself with any Pi 3. You'll find within five to ten minutes or so (depending on ambient temperature, passive cooling, and case airflow) of max CPU usage, the CPU will heat to the point where it throttles, in order to save itself from overheating damage.

In this situation only practical solutions, are to outfit the Pi 3 with an oversize heat sink of the sort one might salvage from a desktop computer, or (since I desire a small case) a fan.

--
If you're not part of the solution, you're part of the precipitate.
Reply to
Jamie Kahn Genet

well you could run the whole thing in a refrigerator. # or move to Lapland.

--
The theory of Communism may be summed up in one sentence: Abolish all  
private property. 
 Click to see the full signature
Reply to
The Natural Philosopher

Or immerse the Pi 3 in oil.

--
If you're not part of the solution, you're part of the precipitate.
Reply to
Jamie Kahn Genet

From the examples I have read you only need a modest heat sink on the CPU to improve the cooling enough to work at full speed. The throttling that happens only seems to be a 20% reduction in clock speed so it is not overheating terribly. Dig around on the Internet a bit. I found lots of good pages on the work others have done and it doesn't seem to be that much of a problem to deal with.

--

Rick C
Reply to
rickman

in general a modest heatsink in free air will about double power handling: shove a fan on as well and its up to ten times.

The worst case is a sealed box. No amount of heatsinking is going to help there.

--
?But what a weak barrier is truth when it stands in the way of an  
hypothesis!? 
 Click to see the full signature
Reply to
The Natural Philosopher

You jest, but I did that for a day while investigating the thermostat on the fridge. The wires to the waterproof DS18B20 temperature probes are quite thick, so would break the seal, but the Pi (original 256MB Model B) power cable is very thin, so I decided to put the whole thing in!

I was surprised that WiFi still worked from inside so I was able to remote monitor it from another Pi, as well as logging to SD card. Taking readings every minute showed the cycling of the temperature and the effect of the door being opened. Having the whole thing in, also enabled me to use a temperature and humidity sensor which produced some interesting readings as humid air was cooled.

But back to the subject; running the Pi at an ambient temperature of around 5C instead of 22C dropped it's idle temperature from 45C to 30C. In another experiment I left it in the car during a night where it got down to -5C but this only resulted in reducing the idle to 26C. Unfortunately I didn't try stressing it under these conditions, to see the affect on the maximum running temperature.

---druck

--
This email has been checked for viruses by Avast antivirus software. 
https://www.avast.com/antivirus
Reply to
druck

There would appear to be something wrong with your numbers. If you raise the ambient temperature 1 degree, the core of the CPU should rise

1 degree. If you lower the ambient temperature 1 degree, the core of the CPU should lower 1 degree.

Your first example is close enough it can be chalked up to measurement uncertainty (electronic thermometers are not very accurate). The second example is hard to explain. I suppose there could be some effect from the air being dryer at temps below freezing and so not being as effective at carrying away the heat.

--

Rick C
Reply to
rickman

anywhere near enough to prevent a Pi 3 from overheating and throttling, when run constantly at max usage.

--
If you're not part of the solution, you're part of the precipitate.
Reply to
Jamie Kahn Genet

You do know that Lappland gets 20-25C summertime ?

Reply to
Björn Lundin

depends how far north you go :-)

--
?It is hard to imagine a more stupid decision or more dangerous way of  
making decisions than by putting those decisions in the hands of people  
 Click to see the full signature
Reply to
The Natural Philosopher

I've been to Ivalo, 68 degrees North, and seen +32C and

-37 C. There was well below -40C when the Sovjets lost a missile from a ship on the Arctic Ocean into lake Inari, and the Finns dived its remains up.

--

-Tauno Voipio
Reply to
Tauno Voipio

On the two Pi's I've got heat sinks (a Pi3 in a large ventilated case and a Pi1 in an small enclosed case) you only see a 5C reduction at idle, but the hotter the chip gets the more effective they are. The peak temperature of my Pi3 used to get up to 80C now doesn't go above 70C. The Pi1 doesn't show as much benefit as there is very little airflow.

---druck

--
This email has been checked for viruses by Avast antivirus software. 
https://www.avast.com/antivirus
Reply to
druck

Only if there is a perfect heat transfer between the chip's core and the surrounding environment.

I suspect its the difference between the active cooling in the fridge, and the passive cooling inside the car. The Pi had no heatsink and was in a case, albeit one with quite a lot of ventilation, but probably enough to retain some heat. All the other temperature sensors were some distance away from the case.

What I'll do if I try it again, is to put one of the probes inside the case to see how the chip, ambient and case temperatures are related.

---druck

--
This email has been checked for viruses by Avast antivirus software. 
https://www.avast.com/antivirus
Reply to
druck

I'm not sure what *you* mean by "perfect heat transfer", but I can assure you, all heat created in the CPU is transferred to the environment.

At this point I'm not sure what you are doing. If you want to compare two things, you need to keep everything else constant.

--

Rick C
Reply to
rickman

Heatsinks are completely redundant on Pi 1 and 2.

Reply to
A. Dumas

ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.