H115i RGB Platinum - Something isn't right here... (3900X)

[r.linder]

The Castle 360 V2 is supposed to perform slightly behind of the H115i RGB Platinum across the board, but it performs at least 7 degrees better under load than an H115i with Kyronaut evenly spread along with 4 fans in push pull. The Castle 360 V2 is only using its stock paste and fans (pull config).

The H115i is not even close to a year old and it's clean, and now cooling an 1800X with 2 fans and NT-H2 (1 degree behind Kyronaut with4 fan push-pull config), and it idles in the 40s with the pump consistently set to run at 100% and regardless of fan speed.

I suspect that my H115i RGB Platinum may be at least somewhat defective, because every test I've done and review I've read through has shown that it shouldn't be performing worse than the Castle 360 V2, let alone to this degree. Thoughts?

[Malygos]

What's the load temp I have the original h110 and I've noticed a few times after repasting or switching out for test my pump makes a weird ass noise and the temps are up until I wiggle the tubes a bit the noise stops and the temps even out. My idle temp on my 3700x is slightly higher but I'm overclocked and at idk like 40% fan speed or something so it deff sounds like something is up

[r.linder]

Originally posted by jefedemuchanina:

What's the load temp I have the original h110 and I've noticed a few times after repasting or switching out for test my pump makes a weird ass noise and the temps are up until I wiggle the tubes a bit the noise stops and the temps even out. My idle temp on my 3700x is slightly higher but I'm overclocked and at idk like 40% fan speed or something so it deff sounds like something is up

An example of the difference was when I ran Cinebench R20.

H115i maxed out at around 75C, while the Castle 360 V2 maxed out at 68 C, which is bad because every test I've seen indicates that the results for both coolers should be within a margin of error, without actual meaningful gaps in temperatures.

In both instances, my 3900X was running with Ryzen Master's Game Mode enabled, so it was running only one chiplet (6 core 12 thread) in both instances.

While idle, my H115i was keeping the 3900X in the 40s, while it's in the 30s with the Castle.

Black Desert Online is a game I frequently run, and the temperatures are several degrees lower with the Castle than it is with the H115i.

[iceman1980]

Basic physics the Castle 360 V2 has a larger radiator block therefore more surface area for heat to be exchanged. The Castle also has more fans across that larger surface area.

Castle 360 V2 (402×120×27 mm)
Corsair H115i (140mm x 312mm x 26mm)

[r.linder]

Originally posted by Adaptive:

Basic physics the Castle 360 V2 has a larger radiator block therefore more surface area for heat to be exchanged.

Castle 360 V2 (402×120×27 mm)
Corsair H115i (140mm x 312mm x 26mm)

Look at reviews and tests, the H115i RGB Platinum often performs 1 degree better under load and the same temperature while idle, but there's an arguably large difference in this specific case.

Reason why they're basically equal is because the H115i RGB Platinum has a superior pump from Coolit, while I'm pretty sure the Castle 360 V2 uses the "standard" Asetek.

[iceman1980]

Originally posted by Escorve:

Originally posted by Adaptive:

Basic physics the Castle 360 V2 has a larger radiator block therefore more surface area for heat to be exchanged.

Castle 360 V2 (402×120×27 mm)
Corsair H115i (140mm x 312mm x 26mm)

Look at reviews and tests, the H115i RGB Platinum often performs 1 degree better under load and the same temperature while idle, but there's an arguably large difference in this specific case.

Reason why they're basically equal is because the H115i RGB Platinum has a superior pump.

You've obviously never dealt with measurement error have you?
Or done physics the faster you move the water around will not affect how fast heat is dissipated if anything it make harm the efficiency. Because all thermal exchange is being done through convection initially, the water travels through small channels where this heat becomes infra-emission which is then again exchanged with convection then pushed out the radiator.

Larger the actual surface area you have the faster you can exchange heat.

[r.linder]

Originally posted by Adaptive:

Originally posted by Escorve:

Look at reviews and tests, the H115i RGB Platinum often performs 1 degree better under load and the same temperature while idle, but there's an arguably large difference in this specific case.

Reason why they're basically equal is because the H115i RGB Platinum has a superior pump.

You've obviously never dealt with measurement error have you?
Or done physics the faster you move the water around will not affect how fast heat is dissipated if anything it make harm the efficiency. Because all thermal exchange is being done through convection initially, the water travels through small channels where this heat becomes infra-emission which is then again exchanged with convection then pushed out the radiator.

Larger the actual surface area you have the faster you can exchange heat.

You're not understanding. They get the same results in every review and test I can find.

[iceman1980]

Originally posted by Escorve:

Originally posted by Adaptive:

You've obviously never dealt with measurement error have you?
Or done physics the faster you move the water around will not affect how fast heat is dissipated if anything it make harm the efficiency. Because all thermal exchange is being done through convection initially, the water travels through small channels where this heat becomes infra-emission which is then again exchanged with convection then pushed out the radiator.

Larger the actual surface area you have the faster you can exchange heat.

You're not understanding. They get the same results in every review and test I can find.

By 1 degree. Remind me why you can so much about 1 degree?

[r.linder]

Originally posted by Adaptive:

Originally posted by Escorve:

You're not understanding. They get the same results in every review and test I can find.

By 1 degree. Remind me why you can so much about 1 degree?

You didn't read the post correctly then, because the difference in this particular case is often closer to 10 degrees and not 1. If it was 1 degree, this thread wouldn't exist.
The purpose of this thread is gathering insight on why the H115i is performing worse than it actually should be based on the information available that indicates the difference is within a margin of error when both coolers are working as they should, out of the box.

[iceman1980]

Originally posted by Escorve:

Originally posted by Adaptive:

By 1 degree. Remind me why you can so much about 1 degree?

You didn't read the post correctly then, because the difference in this particular case is often closer to 10 degrees and not 1. If it was 1 degree, this thread wouldn't exist.
The purpose of this thread is gathering insight on why the H115i is performing worse than it actually should be.

Ambient air-temperature?
Internal case temperature?
Well you can easily eliminate a point of issue by using a non-contact thermometer. And checking how hot the radiator actually is.

Those benchmarks you saw how fast were they running the fans?

[r.linder]

Originally posted by Adaptive:

Originally posted by Escorve:

You didn't read the post correctly then, because the difference in this particular case is often closer to 10 degrees and not 1. If it was 1 degree, this thread wouldn't exist.
The purpose of this thread is gathering insight on why the H115i is performing worse than it actually should be.

Ambient air-temperature?
Internal case temperature?

I don't have a proper thermometer to check that. The furnace is set to 27 degrees celsius, but my PC is on the basement level where it's colder, and the air inside the case is definitely below 30, even under load.

I've tried different fan configurations, different pastes (all evenly spread), with no improvement, hence why I suspect something may be wrong with the pump.

[iceman1980]

Originally posted by Escorve:

Originally posted by Adaptive:

Ambient air-temperature?
Internal case temperature?

I don't have a proper thermometer to check that. The furnace is set to 27 degrees celsius, but my PC is on the basement level where it's colder, and the air inside the case is definitely below 30, even under load.

When touching the radiator is there any "hot spots" as in comparison to everything else is one part hotter than the other? Rather than more or less "uniform"

Five fail points for a radiator are

1. Radiator block
   
2. Pump
   
3. Fans
   
4. Hoses
   
5. Heat exchange block

[r.linder]

Originally posted by Adaptive:

Originally posted by Escorve:

I don't have a proper thermometer to check that. The furnace is set to 27 degrees celsius, but my PC is on the basement level where it's colder, and the air inside the case is definitely below 30, even under load.

When touching the radiator is there any "hot spots" as in comparison to everything else is one part hotter than the other? Rather than more or less "uniform"

Not particularly, no.

Though I can barely feel any airflow through the radiator with the stock fans, but when I try other fans, there's more noticeable airflow through the rad.

My friend is running BDO and the temperatures are in the 50s with a colder CPU (1800X), while I'm running it in the 40s with a hotter CPU. (He's using the H115i, I'm using the 360 V2)

[iceman1980]

Originally posted by Escorve:

Originally posted by Adaptive:

When touching the radiator is there any "hot spots" as in comparison to everything else is one part hotter than the other? Rather than more or less "uniform"

Not particularly, no.

Though I can barely feel any airflow through the radiator with the stock fans, but when I try other fans, there's more noticeable airflow through the rad.

When the benchmarks were done were they done with the same CPU as yours?
Also this isn't Kryonaut liquid metal is it?

[r.linder]

Originally posted by Adaptive:

Originally posted by Escorve:

Not particularly, no.

Though I can barely feel any airflow through the radiator with the stock fans, but when I try other fans, there's more noticeable airflow through the rad.

When the benchmarks were done were they done with the same CPU as yours?

Yes, I was using the H115i on my 3900X since I got it at launch in July. I only installed the 360 V2 today and noticed immediately that it was running quite a bit colder.

[iceman1980]

Originally posted by Escorve:

Originally posted by Adaptive:

When the benchmarks were done were they done with the same CPU as yours?

Yes, I was using the H115i on my 3900X since I got it at launch in July. I only installed the 360 V2 today and noticed immediately that it was running quite a bit colder.

Well there is definitely a larger surface area on the Castle 360 V2. Which will translate to better cooling furthermore the Castle has 3 fans not 2. 104CFM * 2 fans is obviously not as good as 3x fans at 69.34CFM