How might I model the effect on the bass response from placing a 10L 2-way with standard 5" midwoofer right on top of a desk?
I am planning to do that but this doesn't answer the question in any way.
Any boundary surfaces (like a rear wall) will have more of an impact on bass below the baffle step frequency.
jeff
jeff
Not sure I fully understand the question. Is the speaker isolated from the desk and you are asking about boundary reinforcement or is the question about the desk acting as a sounding board? Or something else?
Thanks, I think my question is rather - how does one simulate and visualize the effect on the frequency response of placing a speaker against a boundary; e.g. right on top of a desk/table vs short stands like 4" tall Isoacoustics?
The WLs are so large, best to assume free space (4pi); for example my desk is 30" x 66" = 1980"^2 = (1980*4)/pi = ~50.21" dia. = ~13543"_sec/50.21 = ~269.73 Hz WL 'bubble' (sound is round 😉).
Now this has an end correction based on its size/shape and how close to floor, wall boundaries it is, but unless in a concrete bunker or similar max gain construction room, it's probably not going to lower it to the 60-250 Hz bass BW on a typical size desk like mine does, much less the < 60 Hz sub-bass BW. Interactive BW chart
This also highlights the need for some form of baffle step compensation (BSC), so this is where you want to concentrate your efforts WRT 'enhancing' its (mid) bass, lower mids response, i.e. what speaker design is all about: trading efficiency for BW.
The most accurate approach for something like a desk which is not a complete room boundary is to use general acoustic software like a BEM program. I am currently working on a little project using BEM software to quantify the main issue of placing a speaker on/close to a wall along with quantifying the effectiveness of the solutions used to reduce or eliminate it.
On the other hand to get a feel for the issue look at the controls for desk compensation on speakers intended for use at mixing desks.
Erm, at the time I 'read' your interest as only being in its LF response, but any BW above the baffle step up to whatever frequency spans ANY driver's centerline to the desktop will be in 2pi space, so with a small speaker there will be some + 3 dB gain above the BS, which in turn will require even more than the normal BSC offset if not attenuated or otherwise accounted for in the XO.
My original query was yes, how does it affect the bass? It seemed intuitive that plonking a speaker on a desk increases the LF, but by how much?
I guess the simplest way to find out is to get some measurement equipment and compare vs raising them up on some auralex foam pads or little stands like Isoacoustics.
Does the BSC behave differently if the speaker is sealed vs ported?
I guess the simplest way to find out is to get some measurement equipment and compare vs raising them up on some auralex foam pads or little stands like Isoacoustics.
Does the BSC behave differently if the speaker is sealed vs ported?
In a nutshell - you can't. There are so many variables that you end up with data built on a Jenga tower. The fundamental question: where's the desk?
e.g. My desk is in a corner, facing the wall. The R/H speaker in a 3-way corner: desk, R/H wall, back wall. The L/H speaker is in a similar corner: desk, freezer, back wall.
As intimated by Adason, the point is moot. I too, have a subwoofer under the desk. In active system where desktop speakers are receiving very little signal below 100Hz, modelling bass response is a waste of time.
I don't see how active or 2.1 is relevant.
You make a good point about wall boundaries being significant. I would think it is still of interest to have some notion of how speaker placement on a desk near a wall (like how most desks are placed) affects the frequency response. Even if modeling such things is dubious or impossible.
You make a good point about wall boundaries being significant. I would think it is still of interest to have some notion of how speaker placement on a desk near a wall (like how most desks are placed) affects the frequency response. Even if modeling such things is dubious or impossible.
Modelling this type of thing is straightforward and accurate using BEM, FEM, FD or whatever acoustical software as mentioned earlier. You may not want to invest the time and effort to learn how to do it but it is neither dubious or impossible. It is in fact what the engineers at the larger speaker companies would use to answer questions like this in an accurate quantitative manner.
Imagine the desk as a mirror. You will see reflecton from your loudspeaker from the desk - both the real loudspeaker and that reflection contribute to the final frequency response. The simplest metod is to calculate interaction between two identical loudspeakers positioned vertically one above other, with C-to-C distance exatly double of height of the center of the driver above the desk. You will get comb-filter frequency response, with peaks and dips at your listening position. In reality, you will get about 5 dB increase in low frequency range, folowed by a huge dip on the frequency where waves from both drivers (one of them virtual) are in opposite phase. After that dip, there will be a peak with somewhat smaller magnitude than 5 dB, than another, smaller dip than the first one, and so on - it depends on the midwoofer dispersion.
Hmm, based on 'desk placement' meaning how close to a wall, then this is half space, so again using my large desk as an example, the BSC corner frequency would be ~0.707x ~269.73 Hz = ~190.73 Hz, so now in the bass BW due to a doubling of desk top area via mirror imaging and hard in a corner's 'mirror' drops it down to ~134.87 Hz, though of course there can't be any joint leaks to get this much reduction.
2.1 may be extremely relevant depending on your cross-over point there may be little or no bass to model. Typically, 5" woofer in a 10l cabinet will struggle to produce f0 below 100hz. A production model like the Mackie CR5-X may claim to go down to 50hz but at the lower frequency it is 10db down. Subsequently, in a green new world, you are flogging the desktop speakers with 50w to produce the same amount of bass as your sub can manage with with 5w. I have discussed the other end of this issue. In modelling bass response: a sub under a desk is effectively in a room with a ceiling height of the 30" or 10" if placed under a bed.
When I examined a sealed speaker at various close distances from a hard(no acoustic treatment) wall, the acoustic response was messy with large dips and peaks. My conclusion was that its would be hard to simulate.. There is Xsim 3D and it has boundary settings - I need to play with around with Xsim 3D
Last edited:
Yeah its quite simple. Only hard part is to come up with a speaker that is relatively problem free from the table position. In addition to the comb filter due to nearby reflection there is also some diffraction from the table edge causing some more, minor thing but still something.
Position speakers as high from table as possible, and as close to you as possiible so that first specular reflection does not happen on the table but on the floor, if thats something feasible. Make the table from some perforated sheet or something to reduce reflections.
Or just put any speakers on the table just like everyone else does 🙂
Position speakers as high from table as possible, and as close to you as possiible so that first specular reflection does not happen on the table but on the floor, if thats something feasible. Make the table from some perforated sheet or something to reduce reflections.
Or just put any speakers on the table just like everyone else does 🙂