Hi there "AudioTom", and welcome to the Forum!
Most importantly I want to be able to record acoustic guitars and vocals quite dry and without capturing too much of weird room responses especially in the (in particular for acoustic guitars) critical range 80 - 500 Hz.
Great! That's a fairly common problem, and not too had to solve, fortunately.
The acoustic guitar recordings I used to make in a(nother) very small room gave me lots of trouble in this frequency range. It was almost impossible to tame popping out frequencies but at the same not to castrate the guitar completely in spite of using dynamic eq.
Right. EQ is, at best, only a partial solution to a recording with poor acoustics. It seldom gives good results. I've seen this likened to making soup: if you put too much salt in the soup, there's not much you can do later to fix that. Room sound is like the salt: once it's in the "soup", you can't get it out again. You can water it down with EQ, or maybe disguise it a bit with other herbs and spices (effects), but it's still there, and affects the entire "taste". The only real solution is to throw out that batch of soup, and make it again with the right ingredients. That's why you could not get rid of the room sound in yours tracks: the room sound is so intricately entangled with the guitar sound, that there's no way to get rid of it.
Also, I want to improve the room response in order to better judge my mixes.
Tracking and mixing in the same room can present some other issues, since the environment needed for mixing is a bit different than that needed for tracking. Fortunately, acoustic guitar and vocals are about the only two things that can be tracked in control room acoustics environment: many other instruments don't track well in control rooms. But guitars do. So I'd suggest aiming for control room acoustics in your studio, which basically means "neutral". The goal for a CR is that the room should add nothing to the direct sound from the speakers, and take nothing away, so that you can hear the speakers, only the speakers, and nothing but the speakers. With some CR designs you can have a room that sounds different from different perspectives (eg, NER, LR, and RFZ). With those design concepts the speakers "see" a different acoustic space from what you "see" as you sit at the mix position, or elsewhere in the room, so it is possible to have a room that is neutral as far as the speakers are concerned, but has a bit of life from the other point of view, for tracking vocals and guitars.
L: 417 cm, W: 302 cm, H: 250 cm
So nearly 13m2 and 31 m3. Small, but not terrible. The specs for a control room call for 20 m2 and about 45 m3. That does not mean that all smaller rooms are terrible and all larger rooms are amazing: it just means that smaller rooms will be more difficult to treat, and even then won't be quite as good as a well-treated larger room. 13m2 is on the smallish side, but can still be quite decent.
SPL Meter for level calibration: Digital Sound 8922
Measurement mic: Sonarworks XREF 20
Speakers: Yamaha HS8
Audiocard + preamp: HDSP9632 + Focusrite Clarett OctoPre
Isover Acoustic mineral wool 60x45x125 cm with 5000 Pa.s/m^2
That should be fine, if you have enough of them in the right places. And the GFR at 5k rayls is good too.
Here is the mdat-file with all measurements (also L and R separated):
You can post your MDAT files directly to the forum here: unlike some other forums, there's plenty of space to do that, and the forum does allow the MDAT file type. If you try to upload here and you get a message about file size, or file type, please let me know and I'll fix it. I want this place to be as "builder friendly" as I can make it, so that members can keep all the information about their project in one place, right here, without needing to host anything off-site. The problem with putting things in other places with links, is that the "other place" might disappear, or move, or change name... then your thread becomes useless, since it is missing key parts! So please upload all your project-related files direct to the forum.
Anyway: there's some pretty decent improvement evident in the REW data.
The good ones can be seen when having a look at the RT60 and waterfall graphs regarding the decay times
Right. There's some good improvement there, but you are getting a bit too low in the region around 200 Hz. I would take care to not go any lower in that region. You are still OK right now, but don't let it go any lower than that. You also need something more in the high end, which shouldn't be too hard to do: thin high-density mineral wool on the rear side walls should help with that. with something thicker on the front side walls and ceiling.
Since 36 Hz is a very low frequency I think it would not cause too much trouble.
It's in the range of the bass guitar. five-string and six-string bass guitars go down to about 31 Hz (B0). That might not be relevant for tracking your guitars and vocals, but it probably is for mixing, if you have bass guitars in your songs. The strange thing is that the peak (36.9Hz) does not seem to match any of your predicted modal issues. Your lowest mode (first axial for the 4.17m length), should be at 41.3Hz, but 39.6 is 6% away from that. I'm wondering if there is an alcove at the rear of the room, or maybe something else that is responsible for the frequency being lower that it should be? It might also have been the 0,2,0 axial at 114 Hz that was damped by your treatment.
For the treatment: Have you tried using that same insulation in your rear corners, instead of your front corners? I'd give that a go, and see what results you get. The bass issue should improve like that. Treating the front wall is important, but the rear wall is far more important, and you'll get much more "bang for your buck" in the low frequency region from doing the rear wall.
The bad effect is a nasty break-in in the SPL graph at about 90-120 Hz which has gotten worse after the treatment and creates a variation of about 20 dB in the room`s frequency response
This might sound confusing, but the insulation did not cause that dip. Rather, it cleaned up other stuff that was "masking" that dip. The dip was there before, but you had some modal issues that also appeared at the same frequencies. Now that you have treated those frequencies, the dip has become visible. It's at 113 Hz, which is almost exactly 3 x the 36.9 Hz peak, so my guess would be that your treatment killed the third harmonic of that peak, which is why you can now see the dip that was there all along.
Regarding the cause of that dip: There's a couple of possibilities... it might be SBIR, it might be the floor bounce, or it might be a modal null (less likely). It seems too high to be front wall SBIR, so my guess would be floor bounce, or maybe SBIR from the side walls or the ceiling. You could check that by measuring the exact distances for each of those, then seeing of one of them matches the frequency.
I am no expert at all and I do not know yet which graphs are more important regarding the measured room.
They are all important, but the time-domain graphs are the most important. Waterfall, spectrogram, and RT60, as well as impulse response (ETC). Look at those in combination with the frequency response graphs, and also look at the phase graphs. You can see how your phase situation has improved significantly with the treatment:
The top one is for the baseline test, the bottom with the treatment. The improvement in the phase is very visible.
Focus on the 90 - 120 Hz gap and try to eliminate or at least to soften it within the scope of the available possibilities given the room size and existing furniture.
First identify what is causing it, then treat it, if possible. If it is floor-bounce, then it might not need much treating.
I would also suggest that you try to damp that 37Hz modal issue: it's pretty big, and in fact has more of an effect on the rest of the spectrum than you would suspect. Fixing that will fix other things too. It's the 1.0.0 axial mode, so rear wall treatment would be the best solution. 36 Hz is very low, though, and hard to treat with insulation alone. Maybe some type of Helmholtz trap, or even a membrane trap, would do a good job of taming that.
2. step: Once the bass is treated well then I would focus on the mid and high ranges which have not been treated at all yet
Right, but as I mentioned above, do be careful about over-treating the 200 Hz region. That is very often a big problem in hoe studios: adding the bass trapping usually also goes too far with the lower mid range, so you need to be careful there.
Actually the microphone was pointing in a 70° angle from the ground to ceiling.
I was going to mention that! Glad you brought it up.
So, overall I would suggest trying those same bass traps in the rear corner, to see how they do there, then identify the source o the 113 Hz dip, and attack the 37 Hz issue, before moving on to the rest.
- Stuart -
Translate This Page