Thoughts on Riser Vents
Erskine has been very adamant that risers with vents are not bass traps, but are instead broadband absorbers that target modal frequencies. Since modal frequencies tend to be in the 20-300Hz range, though, that does essentially mean that the riser becomes a bass trap for all practical purposes.
There's quite a bit of evidence that the riser vents actually work (unlike the "sand in a stage" dogma which had little to no empirical testing until recently). What's still missing is any kind of range of tests to try and figure out what kind of results should be expected, and what are the sweet spots of vents vs performance.
published his results in this thread, which was linked from this thread earlier. They absolutely showed that the riser vents were very effective. The results might not be fully transferable, though, since mtbdudex also constructed his riser to be a series of Helmholtz absorbers rather than have one large cavity. The typical advice for building a riser is to have one large cavity with the vents, so its unclear how much of his results were due to the vents vs how much due to the multiple cavity sizes.
I'm going to try and add a few more points to the existing corpus of data with the hopes that we can learn some more from this.
My theater is now at the stage where it has a sand-filled stage and a fiberglass stuffed riser, but no soffits, columns, screens, or any acoustic treatments of any sort. In theory, I could be sure that any results I see are the results of just the changes I'm making now, since there is so little in the room. It looks like this:
The main riser frame is made of 2x10's with 2x6 stringers spaced 12" O.C. and then stuffed with R-30 pink fiberglass. The main riser cavity is 72 cu ft, with a foot print of 94 sq ft.
I tested using three speakers made of some SpeakerCraft CRSOne 6 ceiling speakers I stuffed in some enclosures, and then my M&K subwoofer. Measurements were done using a Dayton Audio UMM-6 microphone situated in the "listening position" just in front of the riser, 1 meter off of the ground and facing the speakers on the stage.
I did a full 0-20,000Hz measurement using REW, but only the 20-300Hz range is shown here since that's where the riser should be most effective. All REW waterfalls have settings for 50-100dB, 20-300Hz, and with a window of 500ms.
I ran some tests while I was building the riser just to see how the measurements would progress as I went. I don't have any measurement for the room with no riser at all, for reasons that I don't remember. Maybe there wasn't any reason and I just forgot.
The first measurement was done with the riser frame was completed, but with no fiberglass in it and no top at all. I would sort of expect this to work as a completely un-tuned diffuser. It looks like so:
Ringing galore! Practically speaking, it was a very echo-y room at that point.
I then stuffed the riser with fiberglass and re-ran the test (still with no top attached). I expected that the decay times would be far better since I essentially had a 94 ct ft trap on the floor. Indeed:
It was a night and day difference. But yeah, that's not practical. You can't actually sit on fiberglass and nobody wants that much fiberglass exposed to the air.
The Actual Tests
The riser top is made of two layers of 3/4" OSB, glued together with PL-375 construction adhesive and screwed regularly to the framing using 2-1/2" screws. I made my first measurement with no holes for the vents cut at all:
Some things to note. First, the frequency peaks are notably different with the top on. The peak at 52Hz was around 105dB with no top, but is around 98dB with it on. The frequencies between 100Hz and 300Hz are all a good 5dB higher, though. I can't say I was expecting that. It was reproducible, though. Second, it does very notably better than an empty riser with no top, implying that there is some benefit to a fiberglass stuffed riser even without any vent holes at all.
After that, I cut a 4" x 36" hole centered on the back wall, and offset 2-1/2" from the wall. Its measurement looks like this:
Next up is the vent hole on the right wall, which is also 4" x 36" but offset 2-1/2" from the back wall and 2-1/2" from the right wall. I tried to get it as close to the corner as possible.
And then I cut out the hole on the left wall, which is a mirror of the hole on the right wall:
There is a clear improvement between the beginning and ending measurements, especially around 52Hz, maybe 175Hz, and somewhere around 250Hz. In fact, you can see incremental improvement with each hole that was added, although it really seems to kick in with all three holes.
I will say that I was expecting a bit more of a change, though. This knocked off maybe 30ms from the ringing. Based on earlier tests I've seen, I was thinking it would be more dramatic than that.
My vent holes combine to total 3 cu ft of access to the underlying cavity. Maybe more importantly, they take up 23% of the back wall and 40% of each of the size of the riser on the right and left walls. All told, there are 27'-5" of riser cavity facing a wall and 9' (33%) of that is opened via the vent holes.
I wonder if we could say that if your riser has 33% open areas to the walls, that you could expect similar performance gains to what I saw.
Would I get notably better performance if I created a few more holes? What if I had two more 4" x 14" holes on the back wall and one more of the same on each of the side walls. That would increase my coverage to 50% of the open area (41% on the back wall and 58% on the side walls).
What I wonder is if placement matters more than amount. Like if I put the two back 4x14s as close to the corners as possible, would it have a bigger impact than if they were closer to the existing hole? Not sure how I could test that without jumping through a lot of "fixing it up later" hoops.
Another thing I want to test is what kind of vents are appropriate. Linear Bar grilles are the recommended type, but those can get decently expensive. Since I can only find Dayus grilles online (no idea where I could buy Nailor or others), it looks like a 4x36 grill would run me $55 not counting shipping. I can get decorative floor grills for maybe half that both locally and online. Would a decorative grill suffer in performance compared to a linear bar grill? I need to figure out some way to test that.
And if you guys have anything else you want me to test or any other details I've forgotten, let me know!