Hi Earl,
As mentioned in my previous post I've been running multiple subs but without any electronic EQ other than the built in crossovers and level/polarity settings in my subs. I'm in the process of rearranging everything in my listening room so I'll need to go through setting up the subs again. From memory my room had a resonance at about 80 Hz I found objectionable. Large scale absorption built into the walls isn't going to be possible for me to implement so I was considering building one or two Helmholtz resonators tuned to this frequency to try reduce this resonance. I'm also looking at adding a DCX to the system to EQ the subs if necessary. From your statement, it sounds like a couple of helmholtz resonantors aren't going to be very effective compared to the EQ. Or am I misreading this?
In my case the room dimensions are roughtly 7.5'x12.5'x29' and I was looking at helmholtz resonators that were based around tubes that were less than 2' long with roughly 1' diameter.
thanks,
Dan
There is no way that a couple of resonators (unless they are almost as big as say half the room) are going to be as effective as EQ. Small resonators do very little. That's why EQ is so desirable. In my HEAVILY damped (at LFs) room, the lowest mode at about 30 Hz is hardly even touched by the damping - only EQ has any effect.
In a resonator the mouth area determines how effective it is. How much mouth area compared to the total room surface will give you an idea of effectiveness. To tune to 80 Hz, a 2' x 1' resonator would need a very small mouth, or a very long port length. One yields very little result and the other increase size substantially. There is no free lunch with resonators.
Damping IS the best choice, but when this option no longer works or is not available then EQ is the next best thing.
In a resonator the mouth area determines how effective it is. How much mouth area compared to the total room surface will give you an idea of effectiveness. To tune to 80 Hz, a 2' x 1' resonator would need a very small mouth, or a very long port length. One yields very little result and the other increase size substantially. There is no free lunch with resonators.
Damping IS the best choice, but when this option no longer works or is not available then EQ is the next best thing.
So I have a smallish music room and at the far end is a wide clothes closet* with sliding doors (which can be slid over half the opening... for those who are geometry challenged). Below we have results for a mic at my listening chair withe door(s) slid to the left, right, and middle. 5 dB/line
Even displayed this way, pretty dramatic differences.
Actually, my real point is that moving subs around, adjusting what can be adjusted, and measuring the results (and then EQing) is the way to go.
BTW, since the waves are bouncing around the room, adding stuffing in any odd place (like attached to the underside of a table or behind a couch) can be helpful.
Ben
Ummm, plus or minus 2.5 dB from 19 Hz to crossover.... not bad, eh (as long as I remember to keep the closet open on the left side)
*I know there are sim-lovers out there who want to know what's in the closet so they can simulate the parameters and build their own closet: mostly my winter motorcycling clothing.
Even displayed this way, pretty dramatic differences.
Actually, my real point is that moving subs around, adjusting what can be adjusted, and measuring the results (and then EQing) is the way to go.
BTW, since the waves are bouncing around the room, adding stuffing in any odd place (like attached to the underside of a table or behind a couch) can be helpful.
Ben
Ummm, plus or minus 2.5 dB from 19 Hz to crossover.... not bad, eh (as long as I remember to keep the closet open on the left side)
*I know there are sim-lovers out there who want to know what's in the closet so they can simulate the parameters and build their own closet: mostly my winter motorcycling clothing.
Attachments
Last edited:
It has too many degrees of freedom. This iterative process can be hard to do manually. Maybe this can help link
Yes, a closet sized resonator is about as small as would be effective at 80 Hz. Doing 40 Hz would require twice that size.
Welcome. If you have any questions about it, just post here or preferably in the topic from the link. Happy to help with it if possible
Thanks Earl, you've saved me a bit of time wasted on experimenting with small resonators. I definitely don't have space for a closet sized resonator. I may be able to fit a membrane type low frequency absorber along one of the side walls but that would be a project for a much later time. It sounds like EQ is the best next step forward at this point.
Dan
+1 to Earl's remarks - nearly all of them!
My point was that you have various resources and placement options available and you need to measure to gauge results. There's another thread with 32,127 hits that shall remain anonymous that has been going on for months with a fantasy goal of 16 Hz in a church hall and many hundreds of dollars "invested" and there isn't a single measurement yet posted.
True confession: earlier in life, I thought there is nothing more important to bass than the radiation resistance you get in a corner (even for subs besides my old Klipschorn). But testing has convinced me that subs must be shifted around to find their sweet spots in respect of individual rooms. And you'd be wasting your time with otherwise swell McBean Hornresp to help you with this near-final-stage experimentation (or for that matter, other acoustic influences that happen after you carry your box in from testing in the local park).
I am lucky to have a closet. It is properly located to zing the principal eigentone, 27 inches deep (and 66 inches wide and 108 inches tall), filled with my winter and touring motorcycle clothing (suitable enough acoustically) and the guest bedding, and costs me nothing to open the door. Hard to say which parameter is doing the work, but clearly a major influence on my room.
Last edited:
I'm pretty sure everyone knows which "anonymous" thread you are talking about, Ben, and there was an even longer thread that preceded it with even more views and it lasted at least several months as well.
All the passive aggressive taunting in the world isn't going to change the fact the the OP of that thread is pretty new with measuring and seems to be having trouble with his measurement gear or procedure. He is trying, he has said that measurements are forthcoming and bullying isn't going to make that happen any faster.
Let's not forget that a big part of the reason those two threads are so long is because of your constant input. Suggesting things like using a sealed subwoofer and a Klipschhorn !?! for this prosound application, proving you don't know how sealed subs work, constantly attacking the OP for his only stated goal of faithfully reproducing 16 hz (which is more than possible, it's actually pretty easy given enough power, Sd and ported box volume).
That guy has spent over $32000 to date, over $30000 of that on getting the organ itself working and only a couple thousand on the speaker system and amps. He's trying very hard and putting in a lot of hours and a lot of his own personal money trying to get the thing fully functional. While I would agree that measurements are necessary and long overdue, he is trying and these constant passive aggressive pokes are not going to help.
16 hz is not a fantasy and your constant instance that it is only makes for more posts and more views in an already crowded thread. The fact that you recommended using a sealed sub to produce 16 hz in a large church but refuse to discuss the technical logistics of ANYTHING shows that you don't understand the subject matter being discussed. And now you've changed your mind and think 16 hz is impossible.
Constantly posting "The cowboy saddled his horse and rode off in all directions" dozens of times, over and over as you have done not only does not help the situation, it pads the post count you seem so disgusted by and it's actually a misquote.
Helping an organist build a PA system for an organ on a shoestring budget is hard enough when the organist doesn't know much at all about sound reproduction. It's almost a miracle that he has taken steps in the right direction, given the plethora of ridiculous suggestions he's received. And constantly harassing him for measurements won't magically speed the process up. He's working on it at his own pace. He's been told dozens of times what he needs to do, he just has to do it.
All the passive aggressive taunting in the world isn't going to change the fact the the OP of that thread is pretty new with measuring and seems to be having trouble with his measurement gear or procedure. He is trying, he has said that measurements are forthcoming and bullying isn't going to make that happen any faster.
Let's not forget that a big part of the reason those two threads are so long is because of your constant input. Suggesting things like using a sealed subwoofer and a Klipschhorn !?! for this prosound application, proving you don't know how sealed subs work, constantly attacking the OP for his only stated goal of faithfully reproducing 16 hz (which is more than possible, it's actually pretty easy given enough power, Sd and ported box volume).
That guy has spent over $32000 to date, over $30000 of that on getting the organ itself working and only a couple thousand on the speaker system and amps. He's trying very hard and putting in a lot of hours and a lot of his own personal money trying to get the thing fully functional. While I would agree that measurements are necessary and long overdue, he is trying and these constant passive aggressive pokes are not going to help.
16 hz is not a fantasy and your constant instance that it is only makes for more posts and more views in an already crowded thread. The fact that you recommended using a sealed sub to produce 16 hz in a large church but refuse to discuss the technical logistics of ANYTHING shows that you don't understand the subject matter being discussed. And now you've changed your mind and think 16 hz is impossible.
Constantly posting "The cowboy saddled his horse and rode off in all directions" dozens of times, over and over as you have done not only does not help the situation, it pads the post count you seem so disgusted by and it's actually a misquote.
Helping an organist build a PA system for an organ on a shoestring budget is hard enough when the organist doesn't know much at all about sound reproduction. It's almost a miracle that he has taken steps in the right direction, given the plethora of ridiculous suggestions he's received. And constantly harassing him for measurements won't magically speed the process up. He's working on it at his own pace. He's been told dozens of times what he needs to do, he just has to do it.
Last edited:
Don't know if anyone noticed this. Anyone know if 3 inputs can be processed this way with the DCX now? Is miniDSP the way to go for EQing multi subs now? The only one I could find that would process more than 2 analog inputs is the 10x10 HD at $599. A bit more expensive than the DCX. Are there any other good options to look at in 2016?
Dan
Hi Earl,
Would you mind explaining a bit more what you mean by damping? I thought that a resonator did offer damping, but at a very narrow bandwidth. I know that resonators can be designed such that the Q is modified and so you can have a big peak but very narrow bandwidth or a broader bandwidth but lower peak (sometimes such that the effect is totally gone). Having said all that, I thought you and I had chatted in the past about this, and it had been my understanding that damping at low frequencies is very challenging.
Anything over 100hz can often be handled with big thick velocity absorbers (meaning something using thick (like 6" to 12" thick) fluffy stuff, with higher density being more ideal. However, once you fall farther and farther below that point, the waves become pressure waves farther and farther from the wall, rather than velocity (If I recall, there is an inverse relationship between velocity and pressure of a wave at low frequencies, yes?), and this makes velocity absorbers ineffective unless they were unsuitably thick. Damping at say 40hz might require 3' depth of damping material, which would be untenable for most people. However the alternative is to use resonant devices that convert the pressure waves back to velocity waves (within their bandwidth) and then absorbing that with damping material. This is the premise of Peter D'Antonio's work with RPG. His products and the ones that generally made the most sense to me would use panel dampening approaches, but my understanding is that the "panel" is still a resonant panel and that its bandwidth is defined by the resonant frequency of the panel and its Q. Different materials can impact that. These systems are generally more effective when the panel is covering a cavity that has a good amount of absorbing material in it. However, it was also my understanding that even these devices, which I had thought were more efficient at such low frequencies, still needed to be quite large.
So in short, is this understanding not right? Is it not true that at low frequencies the waves become pressure waves farther out from the barrier and that this makes velocity absorbers less effective at such low frequencies? I've always had trouble figuring out what these thresholds would be. I've seen calculations, including in Peter's own acoustics book, but based on those calculations none of the velocity absorbers should be doing much of anything in most common listening rooms below say 100-150hz, yet I constantly see measurements claiming pretty significant LF absorption coefficients at such low frequencies.
I've wanted to play with panel traps for a while but had little ability to do so. I've experimented on small scales, but the designs were flawed. Either the boxes were not rigid enough and had multiple resonances (the side panels become resonant panels as well), which ruined their performance, or the overall design was too small. I built one using an old subwoofer box with the front removed and a piece of mass loaded vinyl over the front, sealed in place. It had two big problems, first was that its resonant frequency ended up being far too low, it wasn't "stretched" tight enough I don't think. Second was that the Q was too low and so it operated over a pretty broad range, but with little damping effect (based on a mic measurement and accelerometer measurement).
I am using a minidsp unit and trying to figure out a multi sub optimizer software. Minidsp is much more flexible than dcx and has excellent software and plugins.
PjPoes
All waves are pressure anti-nodes at a rigid boundary, the velocity into the surface is zero (node). There is then velocity anti-node at a 1/4 wave out from the surface. Soft materials will work best at the velocity maximum, which can be many feet out from the surface at LFs. Hence, soft materials placed on a wall do virtually nothing at LFs. Regardless of what kinds of data you may see, this is just a fact of life.
A panel means that the surface is not rigid and hence it becomes more of a velocity anti-node at its resonance. This makes the softer absorbent materials more effective at LFs as well as the internal damping of the construction. My walls are basically all panel absorbers as they are suspended CLD panels. This is the only way that I know of to get good LF absorption, which is why I don't tend to see modes in my room - only the very lowest one where the walls are actually part of the mode.
the subject of damping is far too extensive for a forum post and in fact there are many texts on just this subject - my favorite being Uno Ingard's (rare.) Ingard shows how the thickness of the absorber as well as its distance from the wall affects its absorption.
Helmholtz resonators do have Q's which varies with design. Broadband ones have very low effectiveness and effective ones will always be very sharply turned - neither being what one wants.
All waves are pressure anti-nodes at a rigid boundary, the velocity into the surface is zero (node). There is then velocity anti-node at a 1/4 wave out from the surface. Soft materials will work best at the velocity maximum, which can be many feet out from the surface at LFs. Hence, soft materials placed on a wall do virtually nothing at LFs. Regardless of what kinds of data you may see, this is just a fact of life.
A panel means that the surface is not rigid and hence it becomes more of a velocity anti-node at its resonance. This makes the softer absorbent materials more effective at LFs as well as the internal damping of the construction. My walls are basically all panel absorbers as they are suspended CLD panels. This is the only way that I know of to get good LF absorption, which is why I don't tend to see modes in my room - only the very lowest one where the walls are actually part of the mode.
the subject of damping is far too extensive for a forum post and in fact there are many texts on just this subject - my favorite being Uno Ingard's (rare.) Ingard shows how the thickness of the absorber as well as its distance from the wall affects its absorption.
Helmholtz resonators do have Q's which varies with design. Broadband ones have very low effectiveness and effective ones will always be very sharply turned - neither being what one wants.
I have always wondered what i could do with the chimney, at least for the 6 months of the year there isn't a fire in it....
Ok so my understanding was right on. How do you ensure that your walls act as a panel absorber rather than a rigid surface? I would have thought that all walls are technically not totally rigid and would offer some damping, but that the resonance would be too low and the q too low to be useful. I'm assuming that your building approach helps ensure that the panel is operating at an appropriate frequency, my guess is each panel is a little different in that regard, and with a higher q to be useful?
Would a cld composed of two layers of half inch gypsum board and urethane binder work well as long as the wall cavity was made sufficiently deep and batting was in the wall cavity, but with an air cavity between it and the panel itself? Is the separation from studs critical, or can it be rigidly mounted to the studs?
Ok I have another question about LF acoustics. I know we have chatted about this in the past, but it recently came up again and I was having trouble with the science.
LF sources that are all colocated act like a single larger LF source. The general rule that people seem to apply is that for every increase in identical drivers you get a 3db increase and for every doubling of power you get a 3db increase. If we pretend that a subwoofer is a fixed unit with a driver and amplifier such that adding more subwoofers always adds one more driver and more amplifier power, then going from 1 subwoofer to 2 subwoofers gains 6 dbs and going from 2-4 goes up 12dbs (because we added two more subs). However that idea only holds true if they are colocated, if not, they don't couple in that fashion correct? When we last chatted about this you had mentioned that you get a few more dbs output, but not his theoretical maximum that you would get had you colocated them.
Ok so assuming the above is true, mark seaton had implied on a post (don't quote me I might have misunderstood him) that if they were all time aligned with dsp that they would couple. I took that to mean that they would act like colocated sources and load the room just as if they were one bigger subwoofer and you would get the theoretical maximum increase in output. My understanding was that this is not correct. Which is it and why?
I had a thought when thinking about this issue. I could see reasons why this would not be true at frequencies below the Schroeder frequency transition and on down to a point at which the wavelengths in the room would be so large relative to the room dimensions that this isn't true anymore and in fact at these very low frequencies you would again see more coupling. Do you in fact see greater very low frequency output with 4 subwoofers as compared to 2 due to better coupling at those frequencies? So let's say that at 80hz you don't have a meaningful increase or at best it's a few dbs, but at very low frequencies there is a much greater increase?
In my own 3 sub setup, I noticed that once I had three impulse responses aligned with each other I had a problem below 30hz or so. It seemed like in the delay settings that aligned the impulse responses best, that gave the flattest response up to 150hz or so, caused a cancelation below 30hz. I had to flip the absolute phase of one of the subwoofers relative to the others to fix this, but it caused a slight bump at higher frequencies. Playing around with the crossovers and eq could fix this but nothing could fix the problem below 30hz. I was never very confident in my measurements that low, but this problem was very repeatable. I don't have great hearing that low so I couldn't actually hear a difference, this was all a measurements issue.
LF sources that are all colocated act like a single larger LF source. The general rule that people seem to apply is that for every increase in identical drivers you get a 3db increase and for every doubling of power you get a 3db increase. If we pretend that a subwoofer is a fixed unit with a driver and amplifier such that adding more subwoofers always adds one more driver and more amplifier power, then going from 1 subwoofer to 2 subwoofers gains 6 dbs and going from 2-4 goes up 12dbs (because we added two more subs). However that idea only holds true if they are colocated, if not, they don't couple in that fashion correct? When we last chatted about this you had mentioned that you get a few more dbs output, but not his theoretical maximum that you would get had you colocated them.
Ok so assuming the above is true, mark seaton had implied on a post (don't quote me I might have misunderstood him) that if they were all time aligned with dsp that they would couple. I took that to mean that they would act like colocated sources and load the room just as if they were one bigger subwoofer and you would get the theoretical maximum increase in output. My understanding was that this is not correct. Which is it and why?
I had a thought when thinking about this issue. I could see reasons why this would not be true at frequencies below the Schroeder frequency transition and on down to a point at which the wavelengths in the room would be so large relative to the room dimensions that this isn't true anymore and in fact at these very low frequencies you would again see more coupling. Do you in fact see greater very low frequency output with 4 subwoofers as compared to 2 due to better coupling at those frequencies? So let's say that at 80hz you don't have a meaningful increase or at best it's a few dbs, but at very low frequencies there is a much greater increase?
In my own 3 sub setup, I noticed that once I had three impulse responses aligned with each other I had a problem below 30hz or so. It seemed like in the delay settings that aligned the impulse responses best, that gave the flattest response up to 150hz or so, caused a cancelation below 30hz. I had to flip the absolute phase of one of the subwoofers relative to the others to fix this, but it caused a slight bump at higher frequencies. Playing around with the crossovers and eq could fix this but nothing could fix the problem below 30hz. I was never very confident in my measurements that low, but this problem was very repeatable. I don't have great hearing that low so I couldn't actually hear a difference, this was all a measurements issue.