Nope, the doublet cancellations are due to path length difference. - same path length, same dips. The 3" driver is less directive at high frequencies. I saw this on subwoofers so I sort of assumed low frequency or non-directive other than teh doublet radiation pattern.
There will be differences because the 12" driver is radiating over its entire area and the 3" has a baffle, and because the 12" will break up sooner than the 3", but this seems to me rather academic vs. practical. It seems as if Hornresp is a good tool to run some virtual experiments to give you a better feel for this.
Of course, in your fantasy land everything is incorrect and even your own incorrect opinions are often contradictions of each other.
Again, you are very confused about what simulators do. WE ALL KNOW ROOMS CHANGE THINGS, PLACEMENT WITHIN THE ROOM CHANGES THINGS. This doesn't even need to be said. But a full room sim is NOT what simulators do.
"I don't think"? Is that the best you can do? Why don't you do some testing instead of this boring jaw-jaw? In this case, the guy that thinks the theory and the sims are wrong has the burden of proving his case.
In this case the sources are NOT different shapes, they are both circular (only the size is different) and the interference regime is NOT very different as the physical boundary is also the same shape (and the same size).
If you want to see what the diffraction profile looks like for these two scenarios in terms of frequency response, here it is, and I've used a square baffle to make you happy as you don't seem to like the circular model.
An externally hosted image should be here but it was not working when we last tested it.
That's a 3 inch round sound source on a 12 inch square baffle compared to a 12 inch round sound source on a 12 inch square baffle. In each case, sum the diffraction profile to the speaker's native response and that's your speaker's real world 2 pi response.
And don't give me this "I wish people would do some testing" garbage, this program is used on a daily basis to design speakers, and the sims match the measurements. Maybe not a perfect overlay but not too far off. I've seen hundreds of sims that match the measurements, the sims are not wrong.
Again, sims don't do full room models. But dipoles do sim very well IF you design them properly, including the first three primary reflection points.
Check MJK's OB design papers for example. The sims look pretty good because they are properly designed.
Last edited:
In case you don't know what you are looking at in the picture above, the big peak at about 1 khz in both cases is the dipole hump, corresponding somewhat to the baffle step frequency. Below the baffle step frequency you have SEVERE cancellation from the out of phase front and rear radiation resulting in severe roll off below the baffle step frequency. Above the baffle step frequency you have some rippled response.
ALL of this is due to diffraction effects, highlighting the fact that below the baffle step frequency there is cancellation because the waves are large enough to wrap around the baffle and at and above the baffle step frequency there is rippled response but still constant as the waves are too small to wrap around the baffle so they are not cancelled out.
This is what a diffraction profile looks like and there are a few programs that can do this sim. These sims are known to be accurate due to hundreds of users using these programs on a regular basis and providing sims vs measurements that match very well.
ALL of this is due to diffraction effects, highlighting the fact that below the baffle step frequency there is cancellation because the waves are large enough to wrap around the baffle and at and above the baffle step frequency there is rippled response but still constant as the waves are too small to wrap around the baffle so they are not cancelled out.
This is what a diffraction profile looks like and there are a few programs that can do this sim. These sims are known to be accurate due to hundreds of users using these programs on a regular basis and providing sims vs measurements that match very well.
OK, so we agree that a 12-inch radiating surface is different than a small source on a 12-inch baffle, even if the model doesn't seem to know the difference.
But you seem in reverse about what is academic and what is practical. It is academic (in the bad sense) to make assumptions (as has been done to this point) that there is no difference between the two conditions or that all interference follows the textbook picture. It is practical and useful to recognize that the academic does not satisfactorily describe the real-world interference when a dipole speaker is sitting on the floor in a room.
Glad we are finally getting things cleared up because phase interference in the real world just isn't the terror some worry about. Again I suggest, more people should try removing the back of their speaker boxes and seeing if they don't like the sound better.
Ben
Last edited:
The model is quite aware of the difference, it's just that the situation isn't that different. The boundary for both is a 12 inch circle and that's what the diffraction profile is based on (in addition to the size and location of the sound source in relation to the boundary).
You seem very lost.
The speaker does what it does long before (at least a few milliseconds) before the room gets involved. The speaker puts out waves. These waves get to the boundary of the baffle long before they get to the room walls. The real world interference (phase cancellation) of the speaker is the same in room or in an open park. AFTER the wave passes the baffle (and after the phase cancellation) the wave moves on to the room walls and that's when things change. And the sim can even predict the first primary reflection effects, but that's where it stops, it can't predict the full 6 boundary room condition.
There is a difference between a 12 inch driver with no baffle and a 3 inch driver on a circular 12 inch baffle. But the diffraction profile will look similar (peaks and dips in roughly the same places but different amplitude of peaks and dips) because in both cases the CAUSE of the diffraction (the physical baffle) is the same.
Your earlier comments show that you are confused about how waves work too. "In reality, the stuff the coming out the back heads off in all directions and bounces back in ways that are not feasible to predict..."
That's not how waves or primary reflections work. If you don't understand waves and primary reflections this whole topic is going to go right over your head. It's all very well understood and extremely predictable up to the point of the first primary reflections from the closest three boundaries (walls). The wave itself is quite uniform, it doesn't head off in all directions randomly and it doesn't bounce back randomly in ways that are not feasible to predict.
It appears nothing at all is getting cleared up because you don't understand the basics. Why have you not removed the back of your AR1 if you think it's so much better? I realize that for some reason you are not able to directly address me, but maybe you can tell someone else the reason that you don't do that.
Last edited:
Hi shank1207,
To add a reverse polarity point source without deleting an existing source:
1. Shift+Alt+Click the mouse at the desired location to turn the selected element pink.
2. Set the Reverse Polarity check box.
3. Shift+Alt+Click the mouse on the pink element.
As indicated in the Help file, to specify a point source it is necessary to "click the mouse" twice at the same location.
Kind regards,
David
- Status
- Not open for further replies.