Containing a 4790k overclock with air...?

[GuruMeditationError]

14 minutes ago, Demonic Donut said:

I wouldn't spend the money unless you want to go air for some reason. You should be fine to use an air cooler. A D15 is about equal to a 240mm AIO. The H115 might perform slightly better due to it being a 280mm. You should try taming the fan speed first. IMO anything under 70C is fine for temps. As the water temp goes up, the efficiency of heat rejection increases. That's why there are major diminishing returns when fan speeds increase and water loop temps drop.

 

 

Prime95 is also not a realistic load, what are your temps while using the computer the way you do everyday? And 53C is quite cool.

Thanks for responding, it's most appreciated.

I want to use air to reduce noise. I previously fitted my cooler out with 4 case fans in push-pull and the temps were okay during everyday usage but it was so quiet that the noise of the pump motor was no longer being masked by the fans. It needs the louder fans to drown out the pump noise (what can I say, a Liquid Metal re-lid on a 4790k kicks ass)...

...my current goal is to try and remove the pump from my system and move to a dual-tower heatsink that I can apply three of the quiet fans to. Just wondering if I've got enough temperature overhead to make it work.
 

13 minutes ago, Bombastinator said:

I’ve seen tests  that show it both going both ways.  I suspect the difference is small enough that make/model of cooler and fan start to come into play. 

Thanks, yes, I'm wondering how things might change if I replace the fans. For instance I find it really hard to believe that an NH-D15 could beat out an Akasa Venom Medusa if not for the fans they come with as stock (the Medusa has eight heat pipes and looks like it has more fin surface area).
 

12 minutes ago, For Science! said:

Get a program that can log data, even something as simple as MSI afterburner can do it. You see below for a 240 mm AIO (H100i v2) that I used to use, starting from the far left it starts at 31 degrees, spikes up and then gradually reaches a flat line. That is when you know you are at steady state. I think 15-20 minutes should be plenty

 

If you want to reduce your noise, just dial your fan curve back.

Thanks, I'll take another look at it. I ran it for about 10 minutes before but at about 10 minutes the temps and the fan speed dropped considerably. Do you think that might be the cooler settling or does Prime 95 vary its routines?

Maybe I should avoid Prime 95 entirely and go with something like Valley?

[For Science!]

4 minutes ago, GuruMeditationError said:

Thanks for responding, it's most appreciated.

I want to use air to reduce noise. I previously fitted my cooler out with 4 case fans in push-pull and the temps were okay during everyday usage but it was so quiet that the noise of the pump motor was no longer being masked by the fans. It needs the louder fans to drown out the pump noise...

...my current goal is to try and remove the pump from my system and move to a dual-tower heatsink that I can apply three of the quiet fans to. Just wondering if I've got enough temperature overhead to make it work.
 

Thanks, yes, I'm wondering how things might change if I replace the fans. For instance I find it really hard to believe that an NH-D15 could beat out an Akasa Venom Medusa if not for the fans they come with as stock (the Medusa has eight heat pipes and looks like it has more fin surface area).
 

Thanks, I'll take another look at it. I ran it for about 10 minutes before but at about 10 minutes the temps and the fan speed dropped considerably. Do you think that might be the cooler settling or does Prime 95 vary its routines?

Maybe I should avoid Prime 95 entirely and go with something like Valley?

I don't really remember whether P95 comes in waves (I somehow think it does). Valley will not stress your CPU in a "stressful" way as its more of a GPU thing. If you want an alternative "real-world" stress test my recommendation would be RealBench, but do note that it will also stress your GPU and RAM at the same time, so it can be even more stressful for your PC than P95 or any Unigine software.

 

Also if the temps and fan speed dropped considerably during a P95 run, make sure it didn't crash one of its workers (indicating an unstable overclock, needing more voltage).

 

If you are fixed on removing the pump, I think the NH-D15 class coolers will perform slightly worse than your current setup so you probably should still be okay anyway.

[HM-2]

4 minutes ago, GuruMeditationError said:

I want to use air to reduce noise. 

I suspect with proper tuning of your fan curves you'll get exactly the same thermal performance and noise out of your existing AIO. 

[Demonic Donut]

6 minutes ago, GuruMeditationError said:

Thanks for responding, it's most appreciated.

I want to use air to reduce noise. I previously fitted my cooler out with 4 case fans in push-pull and the temps were okay during everyday usage but it was so quiet that the noise of the pump motor was no longer being masked by the fans. It needs the louder fans to drown out the pump noise...

...my current goal is to try and remove the pump from my system and move to a dual-tower heatsink that I can apply three of the quiet fans to. Just wondering if I've got enough temperature overhead to make it work.
 

Thanks, yes, I'm wondering how things might change if I replace the fans. For instance I find it really hard to believe that an NH-D15 could beat out an Akasa Venom Medusa if not for the fans they come with as stock (the Medusa has eight heat pipes and looks like it has more fin surface area).
 

Thanks, I'll take another look at it. I ran it for about 10 minutes before but at about 10 minutes the temps and the fan speed dropped considerably. Do you think that might be the cooler settling or does Prime 95 vary its routines?

Maybe I should avoid Prime 95 entirely and go with something like Valley?

I think you'd be fine with a D15. Or similar cooler. Also, two fans on the D15 is plenty. Especially when most cases have an exhaust fan right behind the cooler anyway.

 

I'm not familiar with the Akasa, but from pictures it looks like the cold plate is thicker and has more distance between the heat pipes and the CPU, which would make the D15 more efficient. Also size of heatpipes, and if the heatpipes are in the right location on the chip matters a lot. I wouldn't be surprised if it was within 1 or 2 degrees either direction of the D15.

 

I like kombuster's CPU burner for a more realistic stress test. I find it's on par with running Folding at Home. Also running long blender renders is a good one.

[Bombastinator]

15 minutes ago, GuruMeditationError said:

Thanks for responding, it's most appreciated.

I want to use air to reduce noise. I previously fitted my cooler out with 4 case fans in push-pull and the temps were okay during everyday usage but it was so quiet that the noise of the pump motor was no longer being masked by the fans. It needs the louder fans to drown out the pump noise (what can I say, a Liquid Metal re-lid on a 4790k kicks ass)...

...my current goal is to try and remove the pump from my system and move to a dual-tower heatsink that I can apply three of the quiet fans to. Just wondering if I've got enough temperature overhead to make it work.
 

Thanks, yes, I'm wondering how things might change if I replace the fans. For instance I find it really hard to believe that an NH-D15 could beat out an Akasa Venom Medusa if not for the fans they come with as stock (the Medusa has eight heat pipes and looks like it has more fin surface area).
 

Thanks, I'll take another look at it. I ran it for about 10 minutes before but at about 10 minutes the temps and the fan speed dropped considerably. Do you think that might be the cooler settling or does Prime 95 vary its routines?

Maybe I should avoid Prime 95 entirely and go with something like Valley?

There is more to coolers than number of pipes and fin surface area.  There is also the diameter of the pipe, the way the wicking is done, the flatness of the cold plate, and the connection between the pipe and the fin. Fans also play a role of course.  I’ve seen an example of a dual stack six pipe cooler that was so lousy it lost to 3 pipe coolers.  It had narrow aluminum pipes with groove style wicking and poorly pressed on fins. Prime95 doesn’t vary its routines all that notably but it’s not really benchmarking software.  It’s a power virus that makes things get as hot as they possibly can by banging unmercifully on the cpu.  Sometimes that is useful sometimes it is not.

[GuruMeditationError]

3 minutes ago, For Science! said:

I don't really remember whether P95 comes in waves (I somehow think it does). Valley will not stress your CPU in a "stressful" way as its more of a GPU thing. If you want an alternative "real-world" stress test my recommendation would be RealBench, but do note that it will also stress your GPU and RAM at the same time, so it can be even more stressful for your PC than P95 or any Unigine software.

 

Also if the temps and fan speed dropped considerably during a P95 run, make sure it didn't crash one of its workers (indicating an unstable overclock, needing more voltage).

 

If you are fixed on removing the pump, I think the NH-D15 class coolers will perform slightly worse than your current setup so you probably should still be okay anyway.

I think there's a good chance P95 isn't a consistent load; also I ran a search on the subject and it seems it's patchy due to complications with cache and other things, so, not a consistent test.

As for whether one of its workers crashed, how would I tell? Can you see that from the basic interface, or does it require more knowhow?

I'm thinking I might just test it by installing some steam games as that seems like the heaviest use the CPU gets outside of the boot sequence and loading screens.
 

1 minute ago, HM-2 said:

I suspect with proper tuning of your fan curves you'll get exactly the same thermal performance and noise out of your existing AIO. 

Thanks so much for responding it's most appreciated.

For Science! mentioned that before also but it slipped my mind to respond to it...

...is there a good way of controlling fan curves? The BIOS manual fan controller seems a bit rudimentary, and I'm not sure how to do it.

[For Science!]

 

2 minutes ago, GuruMeditationError said:

I think there's a good chance P95 isn't a consistent load; also I ran a search on the subject and it seems it's patchy due to complications with cache and other things, so, not a consistent test.

As for whether one of its workers crashed, how would I tell? Can you see that from the basic interface, or does it require more knowhow?

I'm thinking I might just test it by installing some steam games as that seems like the heaviest use the CPU gets outside of the boot sequence and loading screens.
 

Thanks so much for responding it's most appreciated.

For Science! mentioned that before also but it slipped my mind to respond to it...

...is there a good way of controlling fan curves? The BIOS manual fan controller seems a bit rudimentary, and I'm not sure how to do it.

It would say in one of the thread worker windows that P95 spawns, you just have to read it.

 

Steam games are not a good stress test at all, I would heavily recommend you to stick to stress applications if you actually want to see if your OC is stable. Again, I recommend RealBench.

 

Controlling fan curves is usually done through the BIOS, but as you have a H115i it should be done through iCue, but honestly there are so many parameters here such as where the fans are plugged into, whether you plugged in the USB cable of the H115i into the MB, whether your fans are 4-pin PWM, etc etc. That it's a bit out of my willingness to diagnose. Hopefully somebody is willing to do it, but best if you can get somebody knowledgeable physically to your location as that would be by far the fastest way.

 

[GuruMeditationError]

4 minutes ago, Demonic Donut said:

I think you'd be fine with a D15. Or similar cooler. Also, two fans on the D15 is plenty. Especially when most cases have an exhaust fan right behind the cooler anyway.

 

I'm not familiar with the Akasa, but from pictures it looks like the cold plate is thicker and has more distance between the heat pipes and the CPU, which would make the D15 more efficient. Also size of heatpipes, and if the heatpipes are in the right location on the chip matters a lot. I wouldn't be surprised if it was within 1 or 2 degrees either direction of the D15.

 

I like kombuster's CPU burner for a more realistic stress test. I find it's on par with running Folding at Home. Also running long blender renders is a good one.

 

1 minute ago, Bombastinator said:

There is more to coolers than number of pipes and fin surface area.  There is also the diameter of the pipe, the way the wicking is done, the flatness of the cold plate, and the connection between the pipe and the fin. Fans also play a role of course.  I’ve seen an example of a dual stack six pipe cooler that was so lousy it lost to 3 pipe coolers.  It had narrow aluminum pipes with groove style wicking and poorly pressed on fins. 

Thanks, both of you; I really appreciate the input.

I want to go for the Black D15. It's a shame there's not a Black D15S.

But yeah, I see what you mean re. build quality. The Medusa's been tested at about a degree or two hotter than the D15, which isn't much but the D15 has cutouts for high profile RAM, which I have.

Also slipped my mind: The fan curve thing wouldn't work because it'll unmask the sound of the closed-loop's pump. I fitted quiet fans to my radiator once before and the noise coming from the pump was awful; it needs radiator fans to drown out the pump noise. 

[For Science!]

1 minute ago, GuruMeditationError said:

Also slipped my mind: The fan curve thing wouldn't work because it'll unmask the sound of the closed-loop's pump. I fitted quiet fans to my radiator once before and the noise coming from the pump was awful; it needs radiator fans to drown out the pump noise. 

Also honestly, if the pump of the H115i is so noticeable you should get it checked out by Corsair's RMA team, even at full speed it should not be so noticeable. If its really loud usually its a sign that something is wrong and its on its way out, or installed incorrectly.

[GuruMeditationError]

1 minute ago, For Science! said:

 

It would say in one of the thread worker windows that P95 spawns, you just have to read it.

 

Steam games are not a good stress test at all, I would heavily recommend you to stick to stress applications if you actually want to see if your OC is stable. Again, I recommend RealBench.

 

Controlling fan curves is usually done through the BIOS, but as you have a H115i it should be done through iCue, but honestly there are so many parameters here such as where the fans are plugged into, whether you plugged in the USB cable of the H115i into the MB, whether your fans are 4-pin PWM, etc etc. That it's a bit out of my willingness to diagnose. Hopefully somebody is willing to do it, but best if you can get somebody knowledgeable physically to your location as that would be by far the fastest way.

 

I'll give RealBench a try. I've been gaming a lot with the current overclock and it seems stable. I'm really just worried about the temps at this point, and whether I've got the overhead to move to a quiet air solution, but yes, as you say, a lot of variables, and stuff that's kind of unforeseeable e.g. pump noise being drowned out by radiator fans; who knew.

I think I'm just going to have to give it a try and see if it works out. It's a gamble but if it works out okay it'll have been worth it. The worse case scenario is I just put some beefier fans on the D15.   

[GuruMeditationError]

7 minutes ago, For Science! said:

Also honestly, if the pump of the H115i is so noticeable you should get it checked out by Corsair's RMA team, even at full speed it should not be so noticeable. If its really loud usually its a sign that something is wrong and its on its way out, or installed incorrectly.

I hadn't thought of that. I just assumed it was inherent to all AIO pumps and designed to operate with their proprietary fans. I'm not sure if I'd be able to question them on it as I'm not sure how they'd react if I'm not using the fans that shipped with the pump.

Also, the fans I'm talking about were super-quiet. Some of the quietest I'd been able to find and I'd searched a lot. 

But yeah, I think I'm just going to try it and see what happens. I was posting here really to see if the general consensus might be that it absolutely was not advisable given the information provided, but I'm not hearing that, and it seems it might be doable so...I'm going to give it a go.

[For Science!]

37 minutes ago, GuruMeditationError said:

I hadn't thought of that. I just assumed it was inherent to all AIO pumps and designed to operate with their proprietary fans. I'm not sure if I'd be able to question them on it as I'm not sure how they'd react if I'm not using the fans that shipped with the pump.

Also, the fans I'm talking about were super-quiet. Some of the quietest I'd been able to find and I'd searched a lot. 

But yeah, I think I'm just going to try it and see what happens. I was posting here really to see if the general consensus might be that it absolutely was not advisable given the information provided, but I'm not hearing that, and it seems it might be doable so...I'm going to give it a go.

Since I had some time, I recorded what NF-A15, NF-F12 and my h100iv2 pumps sound like. I tried really hard, but I couldn't actually get any solid sound to record on my shitty phone, but thats the reality too, the pumps are super quiet even at full rpm. SInce you can't really hear the pump turn on, go by the LED light that you can see that shows red initially then white when its on

 

 

[Bombastinator]

Pumps are moving parts and occasionally just die.  Another problem that can happen with AIOs is with enough time liquids and gasses can actually migrate through the plastic so you may get air bubbles where there were none before.  Air bubbles in AIOs are exceptionally loud and annoying.  They produce a sort of gurgling  sound quite unlike fan noise. 

[For Science!]

3 minutes ago, Bombastinator said:

Air bubbles in AIOs are exceptionally loud and annoying.  They produce a sort of gurgling  sound quite unlike fan noise. 

Yes, and that's why its important to install it in such a way that the air stays inside of the radiator and not the pump.

[GuruMeditationError]

1 hour ago, GuruMeditationError said:

I hadn't thought of that. I just assumed it was inherent to all AIO pumps and designed to operate with their proprietary fans. I'm not sure if I'd be able to question them on it as I'm not sure how they'd react if I'm not using the fans that shipped with the pump.

Also, the fans I'm talking about were super-quiet. Some of the quietest I'd been able to find and I'd searched a lot. 

But yeah, I think I'm just going to try it and see what happens. I was posting here really to see if the general consensus might be that it absolutely was not advisable given the information provided, but I'm not hearing that, and it seems it might be doable so...I'm going to give it a go.

 

23 minutes ago, Bombastinator said:

Pumps are moving parts and occasionally just die.  Another problem that can happen with AIOs is with enough time liquids and gasses can actually migrate through the plastic so you may get air bubbles where there were none before.  Air bubbles in AIOs are exceptionally loud and annoying.  They produce a sort of gurgling  sound quite unlike fan noise. 

 

19 minutes ago, For Science! said:

Yes, and that's why its important to install it in such a way that the air stays inside of the radiator and not the pump.

Thanks, guys, my loop does make weird noises at times. I just tend to assume an air bubble must have found its way down to the pump. I know people recommend rocking the PC when that happens but I tend to juts ignore it until it sorts itself out (my P.C. is recessed in some wooden IKEA shelving so it's tricky to get to).

 

Thanks for the recording For Science! You can really hear no sound from the pump at all?  That was the main problem I had running my H115i Pro with quiet fans; I actually documented the process in another thread here, detailing how I had to put my Akasa's back on the radiator. Maybe I should give it another try before I pull it. Not sure if I've still got the receipt for it if it is faulty, or if they'd honour it if I'm not using their stock fans? Who knows. I know the receipts here somewhere but I've no idea where.

I'm wondering if there's a way I can power the pump without powering on the P.C.  I think I'll google it and see if I can find a way. I'm still waiting on the fans to arrive. I've got three in my everyday use PC but can't remove them until I've dig out some replacements. It looks like I've got some work to do.

 

[For Science!]

3 minutes ago, GuruMeditationError said:

 

Yes, that recording is without any fans. I made a small pump torturing video, and in this state you can hear the pump scream. It takes a while for it to return to its normal state and takes quite a lot of vigourous convincing (shaking) once you've introduced air into the pump. Not using stock fans should not void the warranty, and especially if you just don't say it as pump performance is not related to fans blowing through the radiator

 

 

[GuruMeditationError]

1 minute ago, For Science! said:

Yes, that recording is without any fans. I made a small pump torturing video, and in this state you can hear the pump scream. It takes a while for it to return to its normal state and takes quite a lot of vigourous convincing (shaking) once you've introduced air into the pump. Not using stock fans should not void the warranty, and especially if you just don't say it as pump performance is not related to fans blowing through the radiator

 

 

 

Maybe there was air in the pump when I tried it and I attributed it to the pump. :0/

I'm going to drag it off the shelf and try and power the pump from the PSU in my everyday use rig. If it's silent I'm going to be seriously happy.

[GuruMeditationError]

Yeah, it's making noise. 
 

I made a video where you can just make it out, but the microphone on my phone isn't good so you have to turn the volume way up...so anyone listening to it, take care not to blow your eardrums if you get a Windows notification...
 

Radiator fans damp the sound, but with quiet fans I remember being able to hear it across my room.

 

Edit: I found a few more on youtube...

 

 

https://www.youtube.com/watch?v=UnslfeNEmHY

 

https://www.youtube.com/watch?v=p1ux0c1lZMk

 

 

 

[For Science!]

Its a little bit whiny, You can try to change the pump speed in iCue I think there should be two presets. I do not think it is air in your case.

[GuruMeditationError]

47 minutes ago, For Science! said:

Its a little bit whiny, You can try to change the pump speed in iCue I think there should be two presets. I do not think it is air in your case.

Thanks, I'll give that a try once the fans have arrived. I didn't know that was possible; do you know if it effects the cooling performance?

[For Science!]

2 minutes ago, GuruMeditationError said:

Thanks, I'll give that a try once the fans have arrived. I didn't know that was possible; do you know if it effects the cooling performance?

not significantly, but don't mess around with it beyond the presets. The flowrate of these pumps are not very high to begin with.