Spectral analysis of noise
Today’s article serves as an introduction of a new player on our market, besides the usual review. This is Valkyrie, a company that offers mainly liquid AIO coolers, although you can also find one air cooler in their offer. The focus of today’s article, however, is the Dragonfang 360, which is a liquid AIO cooler that promises solid efficiency at a reasonable price, which of course I intend to rigorously test on my test rig.
Spectral analysis of noise
Also new is a spectral analysis of the noise of the cooler. This is a simple analysis at noise levels of 39 dBA, 45 dBA and at maximum speed of the cooler. For the analysis I use the UMIK-1 microphone from miniDPS in conjunction with True Audio’s TrueRTA application. The resolution of the analysis is 1/24 of an octave and the shape of the spectrogram is an average of 100 samples acquired over 20–30 seconds.
- Contents
- Key Features
- Measurement methodology
- Results – 36 dBA
- Results – 39 dBA
- Results – 42 dBA
- Results – 45; 48 and 51 dBA
- Results – maximum speed
- Spectral analysis of noise
- Conclusion and evaluation
Great analysis as usual guys
Question, do you take the pump noise levels into account as well?
Any way to test pump noise reliability (vibration or rattling during idle, like the typical zzzzzzz slightly audible if close to the case).
Im thinking about the LT720 DeepCool, but my current LE720 DeepCool produces a slight buzz noise that I want gone.
Thanks!
Hello and thanks for your comment.
Yes, I always take pump noise into consideration as it’s part of all those noise normalized testing. And it’s always an issue with AIO – they can never be whisper silent as air coolers due to pump noise. I have fully fledged custom liquid cooling loop in my PC and even it is not completely silent, my DDC at 20% (lowest possible setting) is still slightly audible.
LT720 is a great cooler, I used it as an interim cooler while waiting on some cooling parts after periodal upgrade of hardware. But you will still hear the pump, no matter what. You want complete silence? Go passive/semi-passive, no other way around it (which sucks a little for me as well, I’d love dead silent system, but I also need computing power which means higher power output and thus more cooling capacity which comes along with some noise as well).
An isolated evaluation of the pump would be nice to see, even if it’s only frequency analysis.
The question is how relevant in terms of accuracy it would be. I suppose that the technical level of deaeration, which is significantly reflected in the noise level of the pump, can be different within the same model. Especially for cheaper liquid coolers. I don’t think that this is something that manufacturers are concerned with in terms of quality standards. Of course, then it is possible to take more samples for analysis, but as you can see, we have different time frames for different tests. We can’t devote as much space to CPU cooler tests as we can to fan tests, for example.