Boundary Gain/Room Gain Questions

[paul_m]

Hey Everyone,

Its been about 10 years since I built my last set of speakers and frequented the board here. But I'm in the midst of a renovation that's going to make a new set of speakers a bit more functional than my current set. My tentative plan is to build 3 floor standing 3-ways for mixed home theater/music use. I'm actually thinking of going counter to the current trend and to make speakers that are designed to sit against the wall, so I want to factor that into the design from the ground up. Most likely, I'll go with speakers that are wider rather than deeper. I think the max dimensions might end up being about 33x12x10 (HxWxD) - so a reasonable size, but not huge. Part of the goal of a 3-way over a 2-way is to get a bit stronger bass support so that I can run without a subwoofer most of the time.

So, anyway, I need to make sure I'm making something that works with both a wider baffle and against the wall, so I'm thinking about boundary gain. I've got a good understanding of the Physics involved here (being a Physicist by education,) but am trying to get an understanding of the reality versus theory. Using Jeff Baby's great diffraction/boundary gain/room gain calculator, I'm seeing some gigantic gains as I go below 100 Hz for a woofer placed near the bottom of the cabinet as I described. Ignoring the wiggles above about 70-80 Hz, its telling me that I'd be about 3 dB up at 60 Hz, up to about 16 dB up at 20 Hz. That seems really surprising to me and would have a very large impact on what f3 I'd want and what type of rolloff would make sense. I've been contemplating sealed vs ported and the tradeoffs there and, if I get anything close to the gains that the calculator is showing, my initial idea of a sealed enclosure with a mid-40s f3 would actually work much better in this configuration than a vented one with an f3 of closer to 30.

As for my questions: does anything I've described here sound crazy? What tips does everyone have for designing near-wall speakers? What pitfalls do I need to look out for? Honestly, I'm surprised there's not more out there on this - near wall speakers would certainly make for easier room placement and higher WAF in most cases.

Thanks!
Paul

[ReissM]

Re: Boundary Gain/Room Gain Questions

... so I'm thinking about boundary gain. I've got a good understanding of the Physics involved here (being a Physicist by education,) but am trying to get an understanding of the reality versus theory. Using Jeff Baby's great diffraction/boundary gain/room gain calculator, I'm seeing some gigantic gains as I go below 100 Hz for a woofer placed near the bottom of the cabinet as I described. Ignoring the wiggles above about 70-80 Hz, its telling me that I'd be about 3 dB up at 60 Hz, up to about 16 dB up at 20 Hz. That seems really surprising to me...

Paul

Roy Allison did a bunch of research years ago on this. The boundary augmentation can be simulated... I would expect that is exactly what Jeff's software does. I entered the equations into an Excel spreadsheet years ago and made my own version of Roy's calculations. However, I think Jeff's program takes it further than what Roy or I did. Mine only calculated the boundary augmentation, not diffraction or room gain. Typically, each boundary has the potential to contribute 3dB of augmentation depending on the speakers' proximity to each boundary. It's the interference issues (deep notches) that you need to worry about. I'll see if I can dig up my spreadsheet and post a few examples.

[Jeff B.]

Re: Boundary Gain/Room Gain Questions

To best understand how the room effect the response click the button to use the "power response" setting for the boundary chart, and assume at least 50% leakage in room gain. A lot of room have more leakage than that. I wouldn't expect more than 10-12 dB of total room gain under real conditions, and sometimes it's closer to 8 dB. However, you are correct, a sealed roll-off almost always matches up with the room response better than a fourth order response does.

[billfitzmaurice]

Re: Boundary Gain/Room Gain Questions

However, you are correct, a sealed roll-off almost always matches up with the room response better than a fourth order response does.

+1, because they both have the same second order slope. That's why auto sound subs, which really can get close to the predicted cabin gain, tend to be sealed cabs with a half space f3 a full octave higher than the intended in use f3.

[Deward Hastings]

Re: Boundary Gain/Room Gain Questions

a sealed roll-off almost always matches up with the room response better than a fourth order response does.

Jon Marsh made exactly that point in his presentation at the Northern California DIY event last week. He uses a boundary effect model (that he wrote) to determine room response (which presumes you know the intended room and placement of the finished speaker), and then matches woofer rolloff to that. And he then uses a sealed alignment with a Q of 0.5 (when size, budget and design intent permits). For the speaker he demonstrated this resulted (as I recall) in a 40Hz inflection point, so the whole issue is mooted if you're using a subwoofer . . . (or perhaps more accurately it becomes an issue in the subwoofer design instead).

[paul_m]

Re: Boundary Gain/Room Gain Questions

Thanks for all the advice. I think that I've gotten the basics down now. Jeff: I actually hadn't noticed that the power response box was a button!

Looking at it now, it looks like the ideal case would be a sealed enclosure with an f3 in the lower 30s. I think that's unlikely in the size speaker I'm considering. An f3 of the mid-40s in a sealed box or low 30s with vented seems a lot more doable. But if anyone has recommendations on a woofer that can get lower than 45 Hz or so in a smallish (max ~1.5-1.6 ft3) sealed enclosure and can be crossed at 300-400 Hz, I'm all ears!

I'll dig around a bit more and keep everyone posted about how this design progresses!

Thanks!
Paul

[Jay1]

Re: Boundary Gain/Room Gain Questions

Thanks for all the advice. I think that I've gotten the basics down now. Jeff: I actually hadn't noticed that the power response box was a button!

Looking at it now, it looks like the ideal case would be a sealed enclosure with an f3 in the lower 30s. I think that's unlikely in the size speaker I'm considering. An f3 of the mid-40s in a sealed box or low 30s with vented seems a lot more doable. But if anyone has recommendations on a woofer that can get lower than 45 Hz or so in a smallish (max ~1.5-1.6 ft3) sealed enclosure and can be crossed at 300-400 Hz, I'm all ears!

I'll dig around a bit more and keep everyone posted about how this design progresses!

Thanks!
Paul

Budget? The RSS265HF-8 will have an F3 of 37hz in 1.5 ft3 sealed, and should be plenty clean for 300-400hz.

[paul_m]

Re: Boundary Gain/Room Gain Questions

I just took another look at the RS subs. That looks potentially awesome, except for the low-ish sensitivity and the price. I may be able to swing it, though.

[Jay1]

Re: Boundary Gain/Room Gain Questions

If you put it near the floor in your design you will have around 86 db sensitivity at 8 ohm for the one I linked.

[ReissM]

Re: Boundary Gain/Room Gain Questions

... I entered the equations into an Excel spreadsheet years ago and made my own version of Roy's calculations. .... Typically, each boundary has the potential to contribute 3dB of augmentation depending on the speakers' proximity to each boundary. It's the interference issues (deep notches) that you need to worry about. I'll see if I can dig up my spreadsheet and post a few examples.

I found my quick boundary augmentation simulator. I can enter the distance that the speaker is from the floor, the back wall, and the side wall and it spits out one single graph that calculates the total augmentation based on the proximity to each of the three boundaries.

I should point out that I'm not an expert on this topic, but it seemed reasonably straightforward to simulate this. What follows is my attempt. If a subject matter expert notices any gross problems with this approach, please educate me on the errors of my ways. LOL.


Let’s start out simple… one boundary (the floor) with a distance to the speaker set at 35 inches. Here’s the theoretical augmentation you can expect:




Notice approximately 3 dB of gain at the low frequencies from this one boundary, followed by some constructive and destructive interference starting about 80 Hz.

Now we'll add a second boundary and recalculate for the augmentation effect of two boundaries (for example the floor and the back wall), both 35 inches from the speaker:




Now you’ll see approximately 6dB of gain at the low frequencies due to these two boundaries, followed by some constructive and destructive interference. Also notice that this augmentation curve is starting to look worse.

Ok, let’s get real. Most of us will have three boundaries to worry about. The floor, the back wall and the side wall. Here’s the augmentation when the speaker is 35 inches away from each of the three room boundaries:




Yikes! We’ve got 9dB of bass augmentation, but 11dB of nastiness (deep null) thrown in there. Fortunately for us, it’s simple to remedy this problem. All we have to do is make those three distances as dissimilar as possible. I've found that a ratio of about 1: 1.625 : 2.375 works well for most full-range speakers. Like this: (20” to floor, 32.5 inches to back wall, and 47.5 inches to side wall. Ahhhhhh yes… now that’s a significant improvement over the 35", 35", 35" in the previous example.






More to come. (Using boundary augmentation curves to optimize full-range speaker locations VS Subwoofer/ Satellite locations.)