Some people say they can hear the effects of diffraction, some say they can't, similar with comb filtering. The changes to frequency response on axis can be relatively easily measured, polar response too, though harder. The smaller effects like the edge acting as a separate source not so easy to measure but can be seen in some simulations.
Hi Rob,
Your results look very close to me.
It's been good to see folks saying what kind of simulations they use.
I can see how xover or compensation sims have a good chance of being very close.
The sims I've been referring to are cabinet plan sims, like Hornresp, etc.
I'd jump up and down for joy, if I could ever get a sim within 5dB of measurements across the board.
But in fairness to the sims, a lot of the variation is just plain raw driver response variations.
I guess should probably be judging cabinet simulation closeness using some heavy response smoothing.
Good points and kind words, fluid
I apologize if i sometimes let my zeal for DIY, and conversing about DIY, make me forget the blessings i enjoy with my testing and building setups.
It's not fair for me to strongly tout methods that are less than widely available.......
Well unless you are measuring outside and depending of how the program is set-up as far as loading the cabinet I can see it being difficult to nail down predicted vs measured. I just set-up for a curve of my choice in a box program and go from there. I don't worry about how the in-room compares to the program. To many variables with placement and the like.
Heavy smoothing can help I guess but you just loose so much information.
Rob🙂
Last summer I honed in on my box simulations and measurements and subjectively my ears are confirming the accuracy of the two.
About one month of daily work with multiple woofers and TS param, test boxes and port alignments etc etc. The matching is very very close. Impedance simulation closely matches measured with a little interpretation/guessing of Ql,Qa and Qp. The nearfield response closely matches sims taking into account the cancellation effect as the frequency rises but as the mic is gradually moved away from the speaker, it starts look more like the sim until the room takes over(I still plan to repeat measurements outdoors)
Somewhere between well damped and max flatness is what I'm going for. With the ported speaker I'm working on, I'm hearing front ported bass that sounds more accurate than I could ever imagined or have experienced in the past. Some top designers on this site have mentioned how Ported can sound nearly the same as sealed. I trusted their experience🙂
With sealed boxes I had the same experience as above except Qa was problematic. Earl helped me out on this one as shown below. My plan now for all sealed boxes is to make them the correct size for a given alignment and not use stuffing for correction of box being to small.
Me - "I see something different when I measure and simulate. For a particular woofer, stuffing lowers the Q but does not fool the woofer into thinking its in a larger box in terms of low frequency extension, only the Q matches the the larger box. The low frequency extension in the larger box is lost in the smaller box..
Hmmm, ??? what am I missing?"
Earl - "It's a little more complicated than that. Stuffing a box absolutely makes the box look larger, the resonance will fall (although this is not a huge effect.) Although over-stuffing can have the reverse effect because a dense damping material takes up more internal space than it creates. So packing density matters.
Also you will see a reduction in Q, but how much depends on the woofer. If Qe is dominate then damping will not have as much effect than a woofer with a higher Qe.
But "low frequency extension" is different than these other factors because it depends on your definition of "extension." Lower Q reduces the output over a range of frequencies around resonance, but it raise them below this region. So where you define "extension" matters.
There is no singular answer."