Home recording studio acoustics questions

[Mark DuBerry]

I have just posted this on Ethan Winer's acoustics forum where I originally got a lot of help in treating my small home studio but I thought I would also put it here as it was suggested that I do so by someone who was recently helping me with REW.

I finally got to grips with REW and took some measurements of my room over the last week or so. I am posting the results of those measurements here because they have shown up some issues which I now need help in addressing.

Here is a link to the thread about the studio where I went through the whole room treatment process; there are also some photos of the room there. I hope it's okay to include that here.

http://forums.musicplayer.com/ubbthr...page/0/fpart/1

The following is the most useful frequency response graph with its respective LF waterfall. The measurement was taken at the 38% listening position (with the proper equilateral triangle etc. in place):

room measured with ecm.jpg

ecm waterfall.jpg

My monitors (Dynaudio BM5A's) have a stated fre. resp. of 50Hz to 20kHz and that is reflected in the graph with the drop-offs at both ends. As far as I can see, the concerns are the dips at 95 and to a lesser extent 350 and 700Hz. Also, I wonder about those at 1.3, 5 and 10kHz as well as the peaks at 50-60 and 140Hz.

I would appreciate any help in interpreting these results and most importantly, what I might do to improve the situation. With my very limited knowledge and what I could glean from the REW forum, I would say that the 95Hz dip is the main issue here. This reflects the fact that I started to question the room's response after noticing that my mixes were not translating well, especially suffering from a muddiness in the lower frequencies as well as a disappointing overall sound, much of which is no doubt to do with my limited mixing skills!

My main questions are:

1. How serious is the 95Hz dip and any other issues apparent in the graphs?

2. Overall, does this look like a reasonably healthy room, especially given its size (10x6.5x8ft.)? The main person advising me on the REW forum thought that it looked pretty healthy, apart from the specific issues already named, and that it reflected the effects of the treatment I have placed there.

3. What can I do in terms of further treatment etc. to address any problem frequencies?

Just to summarise the existing treatment: 4¨ rockwool panels used as corner bass traps in all 4 floor-to-ceiling corners as well as the 4 ceiling-to-wall corners and as broadband absorbers at first reflection points on the side walls and as a large 'cloud' over the mixing position to look after the reflection points on the ceiling.

Thanks,
Mark

[bpape]

The remaining issues are likely:

- SBIR issues due to speaker/wall proximity that is almost impossible to avoid in a small room like that. This can be tested by using a 2" or 4" panel directly behind each monitor.

- Nulls coming off the rear wall which can be cancelled with a 4" panel on the rear wall behind the mix position.

Bryan

[Mark DuBerry]

Thanks Bryan.

SBIR issues due to speaker/wall proximity

The areas directly behind and to the side of the speakers are all treated so although there are still likely to be some unavoidable issues there due as you say to the room size, at no point are the rear or sides of the speakers facing right into walls or corners. The corner bass traps are directly behind each speaker at a diagonal and the treatment continues along the front and side walls (to the end of the panels used for the side reflection points) with no gaps. I did it like this based on previous advice you gave me concerning SBIR. The monitors are rear-ported so those ports fire into the corner traps with a gap of only a few inches. I was previously advised that this would be okay and could in fact be beneficial.

Nulls coming off the rear wall which can be cancelled with a 4" panel on the rear wall behind the mix position.

I forgot to include in my treatment summary that I already have a 4¨ panel up on the rear wall to deal with the reflection points there. My mistake.

So unless I've done something wrong here, it looks like I've covered these two points as best as I can yet there still appears to be that dip / null at around 100Hz. I wonder if that could be caused by something else? Is it possible for example to have too much treatment or bass trapping; or would it be a good idea to fill the gaps in the corners behind the traps with loose rockwool? Would that be likely to deal with problems around a specific frequency?

I've also measured at the 62% position and the response was worse, especially at the 100Hz point.

Mark

[bpape]

OK. Next steps.

Try taking a measurement with the mic back 6" from where you are now.

Also, sometimes in a small space like that, the panels straddling the rear wall/ceiling intersection can cause problems. It's not that we don't want to hit it - but - there may have been something from that which was counteracting another issue in the room that now, isn't cancelled. After taking the measurement up front, if that doesn't work better, try removing the panel I just described.

[Mark DuBerry]

Right. I did do a variety of measurements, one 6¨ back as you suggest - it was no better. However, I left the speakers where they were, i.e. I didn't create a new equilateral triangle to that mic position. Should I do that? I will try a measurement like that and see what happens (but then of course I'll no longer be at 38%).

I'll also try removing the panel across the rear wall / ceiling corner and see how that measures.

Thanks again,
Mark

[Mark DuBerry]

Bryan, I did as suggested: measured 6¨ back from the 38% position (this time with a proper triangle with the speakers) and I also took down the panel straddling the rear wall/ceiling. I even took down any panels that were coming up to the rear ceiling and did a separate measurement like that.

Here is the 6¨ back graph:

6¨ from list. pos..jpg

There appears to be a small improvement in those 2 dips at 100 and 350 Hz but not much. Taking down the rear wall/ceiling panel(s) made no difference at all.

As an experiment, I put the mic in the middle of the room. Here is the graph:

middle of room.jpg

Again, not a huge difference to my eyes in that those dips are still around more or less.

Is there anything else I can try?

Thanks,
Mark

[bpape]

OK. Looks like we gained about 3db overall with the move. That's pretty good. You're now within +/-4 which is actually pretty good for real in-room response. My next thought would be to stay 6" back and shift 6" to one side and remeasure.

Bryan

[Mark DuBerry]

I've just spent some time doing more measurements with pretty interesting results. Tried quite a few different variations. Here are the most relevant.

First, I tried what you suggested Bryan and stayed 6¨ back and moved the mic 6¨ to either side. I did this with the speakers left in the same place as for the central measurement and also moving the whole triangle. Here is the best graph with the whole triangle moved 6¨ to the right:

6¨ right.jpg


6¨ right waterfall.jpg

Not bad but this would create some placement and seating problems so may not be practical.

Then I started experimenting with losing the equilateral triangle and moving the speakers around. Here is the best result I came up with, and the most hopeful of all so far I think in that it would be a practical setup. It breaks the rules with the listening position 6¨ back from 38% but in the midpoint between the side walls, the speakers 114 cm apart and 84 cm between the speakers and the mic. So I have neither an equi. triangle nor the 38% position. However, look at the graph:

wide.jpg


wide waterfall.jpg


I may be mistaken but this looks pretty good to me. Both the main areas of concern seem a lot smoother now and the whole thing looks more even. I know there is a dip at 1 kHz but again, not too bad? What do you think? There is still symmetry here in that each speaker is the same distance away from both side and front walls and also neither are facing directly into the corners. There is also a bit more room between the rear ports and the corner bass traps, 11 cm.

I'll be interested to hear what you make of this.

Mark

[bpape]

Looks good to me. The 'rules of thumb' are there as general starting points to get you in the ballpark. The equilateral triangle is much less of a standard than some of the other things - it's just easy.

What you've likely done is found the tradeoff where one issue cancels another - which is where the best compromise usually lies.

Bryan

[Mark DuBerry]

That's good news Bryan. Just one thing in terms of the compromise: I did think that the equi. triangle was a less important standard than the 38% rule. You mentioned to me before, in an original thread concerning fitting out the acoustics, that I should definitely stay at 38% and that in a room this size "an inch is a ton". So I'd just like to confirm that moving my listening position 6" (6 tons?!) back from there is acceptable and is not likely to cause other problems, or am I safe enough to trust this response? If it is ok, is that simply about the 'one issue cancelling another' tradeoff or is there a technical reason this might work?

Many thanks,
Mark

[bpape]

The 38% rule works in perfectly rigid rooms (there aren't any) just like calculating your room modes won't be perfect. Also, the 38% rule looks to keep you out of the lowest axial modes (length dimension). It does not, however, take into account the interaction of tangential (4 surfaces) and oblique (6 surfaces) modes.

Sometimes, where it best purely in terms of axial causes strong interactions from other modes which are more destructive. Normally, axials are about double the intensity of tangentials, which are double the intensity of obliques. So, axials usually dominate. However, if you fall into a place where you have multiple non-axial modes piling up, it can be very intense and needs to be considered.

Reader's Digest version - stay where you are - it works in your room.

Bryan

[Mark DuBerry]

That's excellent Bryan and really makes sense. It is great to have that explanation and I appreciate you taking the trouble to write it. I like to understand, at least to some degree, what I'm doing and why. I also find this whole area fascinating and can see that it's very complex. So having the simple version as well really helps to clarify it.

Many thanks again Bryan for your help with this. It has enabled me to fine-tune the acoustics and has also confirmed that all the work I put in last year to treat the room has really paid off.

I hope that this thread has been of help to others too and might encourage more people to undertake the DIY approach to acoustic treatment. It is very cost-effective, fascinating in terms of what can be learned and as an added bonus, it really works!

Best regards,
Mark

[bpape]

Just for clarity and future reference, the exercise we just went through is normally best performed prior to doing any treatment. This allows us to get the best response and concentrate on what's left.

Also, sometimes, if the treatments are doing their job, they can mask issues that you'd see otherwise. This is OK, but it's nice to avoid as much as possible up front and make things much easier to address.

Bryan

[Mark DuBerry]

Yes I know. That makes total sense and I would absolutely measure etc. first before doing any treatment if I moved to another room / studio. When I took this job on it was all completely new to me so I found the prospect of doing room measurements a little too daunting, given the need to buy special equipment and the learning curve involved in getting to grips with a measuring programme. So I took a pretty generic approach to treating my room on the assumption that such a small space would need pretty radical treatment and if I covered all the bases as advised, I'd probably deal with most of the issues in the process. However, I totally accept that to treat and then measure is doing things backwards and made it much harder to identify clearly where the problems lay.

Incidentally, I have done some more experimenting with slight shifts in position and discovered that the main issue seems to have been having the speakers too close together. Once they are wider apart, I get a similar result at both 38% and 6¨ back from that. The latter is somewhat better so I'll stick with that but 38% would be okay too.

Mark

[Mark DuBerry]

Bryan, another issue has been brought to my attention and I wonder what you make of it. It's not one that was mentioned before. That is the modal ringing below 60 Hz as evident in the waterfall. Here is the waterfall again:

wide waterfall.jpg


And with the time extended to 600 ms:

wide waterfall 600 range.jpg

Is this a serious issue and if so, what can I do about it? The one thing I wonder about is the fact that as I said before, the monitors' frequency response is 50 Hz - 20 kHz. Could that explain what's going on in the waterfall, i.e. could it be more to do with the limitations in the speakers' response below 50-60 Hz rather than room mode issues?

Mark

[bpape]

The ringing is well under control. For the most part, you're down over 50db within 400ms down to the 30's. You're good to go.

Bryan

[Mark DuBerry]

Excellent. Great news. Thank you very much for the quick reply.

Regards,
Mark