This seems like the best place to post this, and I'm hoping to tap into some of the knowledge base I want to measure a variety of sound levels of fans. However, I'm struggling to find a way to easily do this without investing in very expensive equipment, and I can't really see a way around it. I'd like to be able to measure down to about 25dB minimum - lower would be better but seems even less realistic. Any ideas?
Step 1: Find some music Step 2: Turn it up to 11 Step 3: If you can still hear the fans, they're too loud.
Do you need to have an actual db reading, or is this just to compare different fans? If you just want to compare, you could just use a microphone and a recording program on your PC. Just remember to avoid getting wind into the mic, as that will create lots of noise.
I'd prefer an actual dB reading - a worst case scenario would be to have a base sound level (i.e. make this noise as loud as you speaking normally) which can be played through speakers and then record the fans so they can be compared against that, but the technical side of me wants actual figures for comparing
One of these goes down to 35dB, but to be totally honest you probably won't find a decent method for comparison without specialised (and, as you say, expensive) testing equipment. It's very difficult to get a reliable noise level for something like fans. What would be good in reviews would be to take a spectral analysis of the sound of the fan. Fourier transform that puppy and compare between fans to see which one has a nicer sound.
Thanks for the input Krikkit. I vaguely understand Fourier transforms, and there's definitely software out there which could transform inputs into the appropriate frequency graphs. I take it with those graphs I'd then have to find out what the 'best' and 'worst' frequencies for humans are and use those for analysis?
I have genuinely no idea what make our decibel meter is, I do know it was expensive professional equipment when I bought it, there may be something better now but IIRC it measures all the way down to 20dB. I'll look at it when I'm back in the workshop later and let you know what it is, although you said you were looking for an inexpensive solution which I don't think you'll find if I'm honest. Btw I use a large, thick walled wooden box that's lined with about 5 Quiet PC "Acousti Packs" worth of soundproofing foam as a quiet environment to test noise levels of PC's. It's not perfect but it works well enough for my purposes. To be honest with you if I ever do buy proper equipment for sound levels I will buy a meter that measures in Sone and a properly soundproofed test booth, but you're into big bucks for equipment like that and at present I don't need it. Most of the PC's we produce are quiet enough anyway thanks to the Mcubed controllers we use.
Thanks Unicorn That setup was my plan, although going by the price of the acousti packs I'm hoping to find some thick soundproofing material from somewhere else, maybe offcuts if I get lucky. I've been investigating Krikkit's suggestion, and it might actually work...I've worked out a way to get a decent breakdown of the spectrum, with fans seeming to be between 0Hz and 1kHz mainly? Going by Microsoft, the human ear can hear frequencies between 15Hz and 20,000 Hz, so the first results can be ignored... What I need to find now is which of the frequencies are 'annoying' sounds and which aren't. I'm looking around hoping to find a research document of some sort (the ideal situation would be a list of "good" frequency and "bad" frequency ranges, but that's something I'm not expecting in this world) - the only other option I can think of is to listen to bands of Hz individually and decide myself, but that's obviously going to be hugely biased even with other people doing the same test
It's subjective really - I was only really joking, but kudos to you if you manage it! Find a fan you like the sound of and analyse its frequencies. I think one thing which would be cool to see would be how a fan sounds when it's attached to something. I bought a couple of Zalman 120mm's the other day, and once they were in my system they sounded completely different to on the desk.
You may have been joking, but you've given me something to work with - I've got an idea on how it subjective to the user as opposed to me, which may work. I've got some old heatsinks and grills lying around; I'm already planning on testing air flow through a heatsink and a grill at the same time as compared to open, so sticking a mic next to it shouldn't be too much trouble.
I think there is at least a small amount of objectivity in knowing which frequencies from the fans will annoy you most. The real subjectivity of it will come from personal preference but the frequency response of the human ear isn't flat and we tend to hear certain frequencies louder than others. It varies from one person to another, but typically we hear sounds louder between 2.5 and 3kHz, while there are also other bands where our frequency response peaks (around 1kHz, for example). So if your fans emit sound within one of those frequency bands it's going to seem louder and presumably more annoying. Adjusting the speed of the fan will affect both the base frequency of the sound it generates and the frequency spectrum, including other harmonics/partials in the sound (particularly if the fan is vibrating against the case). This means that at certain fan speeds, there will be harmonics in the fan's sound that might fall inside those frequency bands which we hear louder, which I think is why it always takes quite a bit of experimentation to find a fan speed a given person can tolerate, rather than just picking an RPM value off a graph to suit a balance of performance and noise. It'll be difficult to make any really meaningful observations about the loudness of a fan without taking its spectrum and the ear's frequency response into account. I haven't actually made any study of the spectra of fans though, so it could be that their spectra don't vary that much, but I'm inclined to suspect they will. Just make sure that if you're testing with mics that you keep every fan at the same fixed distance and orientation from the mic. A small variation in distance or orientation will mean a proportionally bigger difference in results because of the Inverse Square Law. Also bear in mind that your choice of microphone can skew your results, as microphones don't usually have a flat frequency response either.
I'd probably simply test the fans at 12V and 7V, and see how the fans sound and how much air they push. Thank you very much for the information Zurechial - I've got a couple of questions if that's OK... The first would be - how much influence will my microphone have on results? I don't want to spend much on equipment if I'm going for a software solution, but would buying a better microphone mean the tests are so much more accurate its not worth doing with a cheaper microphone? Secondly, the information on the sound bands which we hear louder was really useful; where did you learn that? Any chance of a link so that I can get more details?
