Since dipole peak frequency Fp = C/2d, you calculate
dipole peak = 344 * 1000/(2x88) = 1954 Hz
I have a 3" Tangband Full range, and actually that will be my next project when I finalise my last xo. Yes, someting even more skinnier than my current P13+tweeters. I'm looking at 14cm baffle, just because that is the stock pine cutting locally here. I dislike MDF dust!
I'll have outside masurements then and we can compare notes.
The low xo point may need to be about 300-500Hz to obtain enough spl after 6db/oct. boost. But the good thing is the system will be quite simple, just 2 amps and 2 way. My H-frame is able to do about 400Hz without requiring h-frame notch (quite shallow h-frame, d=24cm).
We should call it ob needle or something
Hi Andrew
I could have been more specific about what to look for in the measurements - in particular, for a dipole to be 'well behaved' off-axis, IMO, it should follow the dipole generality of being -1dB at 30deg, -3dB at 45deg and -6dB at 60deg off axis. If you look at the unbaffled version, you can see that off axis FR's look very similar to the on axis FR - just attenuated in level, and very close to what one would expect for an ideal dipole. Only place where the off axis behavior become 'irregular' is between 7-9k.
Looking at the baffled measurements, you can see that the off axis measurements are very much different then the on axis - they are no longer parallel, and have different FR's, and in fact, much of the off axis response is at a higher level than the on axis (from about 1.5k to 10k). I would call this fairly 'irregular', and you can expect this sort of behavior whenever an OB speaker is operating above its dipole peak. I assume that when people mount a 5" driver on a 600cm baffle, the peak is much lower, and the off axis behavior is much more irregular.
To me, the difference between baffled and unbaffled for the Neo3 tweeter was easy to hear - particularly, vocals sounded clearer. Compared to being baffled, unbaffled sounded like it had lost the sort of baffle step issues one hears with boxed speakers. It was a little surprising.
I've got a couple of regular sealed tweeters I've been planning on strapping back to back, and then measuring, to see what happens - they will definitely have a greater dipole separation distance compared to the Neo3, so the peak will be lower, and there will be more response irregularity. But you never know how much until you measure.
WRT edge treatment, like baffle roundover with box speakers, my impression has been that it may help a little, but won't take care of the majority of the problem. One should keep in mind that wings and very thick baffles (>25mm) introduce problems of their own WRT off axis behavior. There has been one or two interesting studies of baffle response variations for slightly different baffles - Gainphile did one recently. They are interesting, showing even small differences in baffle dimensions can have a substantial impact on the off-axis response. I think they showed edge roundover to be ineffective.
Since dipole peak frequency Fp = C/2d, you calculate
dipole peak = 344 * 1000/(2x88) = 1954 Hz
I might add that the baffle geometry can shift the predicted peak somewhat - particularly round versus rectangular, with round being a bit higher.
Hope those tweeters are of the small Neodymium type, like the Dayton ND16FA 6 and ND20FA 6. Otherwise it would be a big step back from what you have achieved already. I have yet to see dipole tweeter measurements that equal those of the Neo3. Those B&Gs clearly look like the best choice today.
That's what I tried to explain in length some posts above.
I'm going to try both neo and regular size tweeters, just to see what happens. Looking at specs alone, one could say that the Neo3 doesn't have enough power handling for very loud listening, but I've driven the Neo3's hard, and warmed them up some, but haven't burned them yet (while I did burn a Visaton B200 listening loud to recordings of drums I had made ) The nice thing is they aren't too expensive to replace, unlike some drivers.
I've become very interested in the Neo8 recently, for a couple of reasons - I'm guessing they have similar vertical dispersion to a Beyma TPL-150. I'd love to try the Beyma, but they are way to expensive to justify, and I'm concerned with how their vertical dispersion will effect the highs, so trying a Neo8 seems like a good way to find out. The other really interesting thing about the Neo8 (and possibly the Neo3 to some extent), is how low they can be crossed - being able to cross somewhere between 500-1000Hz seriously changes what drivers I'd choose below that, in terms of ideal driver bandwidths, dipole peak placement and system power response. It might pare a 4 way design down to a 3 way.
) The nice thing is they aren't too expensive to replace, unlike some drivers.I've become very interested in the Neo8 recently, for a couple of reasons - I'm guessing they have similar vertical dispersion to a Beyma TPL-150. I'd love to try the Beyma, but they are way to expensive to justify, and I'm concerned with how their vertical dispersion will effect the highs, so trying a Neo8 seems like a good way to find out. The other really interesting thing about the Neo8 (and possibly the Neo3 to some extent), is how low they can be crossed - being able to cross somewhere between 500-1000Hz seriously changes what drivers I'd choose below that, in terms of ideal driver bandwidths, dipole peak placement and system power response. It might pare a 4 way design down to a 3 way.
Rudolf
Thanks for pointing out my silly error.
. . 1954 Hz presents a bigger challenge . .
Now am I getting D right?
“I'm looking at a 14 cm baffle”
If the baffle width is 140, a 1st estimate:
ok D is the extra distance sound has to travel to the ear, from the rear compared to the front . .
Half the baffle width is 70 mm.
For a rectangular baffle, directly on either side of the centre of the driver, this would be the minimum extra distance, from the centre of the back, to the side - where the sound wave is on the same plane (distance to the listener) as the sound waves from the front of the driver.
As the rear waves radiate at 360 degrees from there, that extra increases. Lets say the extra distance on the rectangular baffle averages 100 mm.
And the baffle is 19 mm deep, so D = 100 + 19.
Have I passed D 101 yet
Thanks for pointing out my silly error.
. . 1954 Hz presents a bigger challenge . .
Now am I getting D right?
“I'm looking at a 14 cm baffle”
If the baffle width is 140, a 1st estimate:
ok D is the extra distance sound has to travel to the ear, from the rear compared to the front . .
Half the baffle width is 70 mm.
For a rectangular baffle, directly on either side of the centre of the driver, this would be the minimum extra distance, from the centre of the back, to the side - where the sound wave is on the same plane (distance to the listener) as the sound waves from the front of the driver.
As the rear waves radiate at 360 degrees from there, that extra increases. Lets say the extra distance on the rectangular baffle averages 100 mm.
And the baffle is 19 mm deep, so D = 100 + 19.
Have I passed D 101 yet
Yes, the distance between the centre of front to the centre of back would be 140 mm (or very closer to it)
But is it the *extra distance that sound travels to the listener?
I understand it as in the attached. Excuse my crappy attempt with Paint.
The sound path from the centre of rear to the "front plane" is the red line, from the centre of the front of the driver to the same front plane is the bluish line
But is it the *extra distance that sound travels to the listener?
I understand it as in the attached. Excuse my crappy attempt with Paint.
The sound path from the centre of rear to the "front plane" is the red line, from the centre of the front of the driver to the same front plane is the bluish line
Attachments
From listener's point of view, it would be far more complex than that, because you need to take into account reflection etc, in other word, impossible.
With practical dipole, what I measure outdoor at 1m, is the same at 2m more or less. Theoreticaly if you hear it from infinite distance, it does not matter and will become a point source anyway.
I generally raise the speaker 2-3m from the ground, then measure outside from 1.5m on-axis The peak is readily visible from measurement, better without gating.
What's interesting is while the frequency changes, the Q and Depth were somewhat identical. Q is always around 1.4 - 1.6 and depty is 4.8db to 6db.
Anyway, we'll see when I built my 3" dipole in few weeks time. It may work out great with excellent simplicity, or it may be a miserable failure
With practical dipole, what I measure outdoor at 1m, is the same at 2m more or less. Theoreticaly if you hear it from infinite distance, it does not matter and will become a point source anyway.
I generally raise the speaker 2-3m from the ground, then measure outside from 1.5m on-axis The peak is readily visible from measurement, better without gating.
What's interesting is while the frequency changes, the Q and Depth were somewhat identical. Q is always around 1.4 - 1.6 and depty is 4.8db to 6db.
Anyway, we'll see when I built my 3" dipole in few weeks time. It may work out great with excellent simplicity, or it may be a miserable failure
Sure, from an infinite distance, it probabaly does not matter.
But from a typical distance of say 3 or 4 m . .
And yes there are other factors, but I simply want to be clear on how to apply the theory to practise, correctly.
I like to cut the wood, knowing the specific potential implication . .
Is my drawing right?
But from a typical distance of say 3 or 4 m . .
And yes there are other factors, but I simply want to be clear on how to apply the theory to practise, correctly.
I like to cut the wood, knowing the specific potential implication . .
Is my drawing right?
Modelling recatangular baffles are not easy. If you like math, this is a good read. Basically they can be modelled by an equivalent circular baffle:
http://www.musicanddesign.com/Equivalent_Baffles.html
My path of least resistance is simply to build and measure, with some educated guess from past experience. Wood cost only $3 anyway. YMMV
Again, I can't recommend enough JohnK's modelling tool (ABCDipole). Great $14 investment I think.
Anyway sorry for being off-topic to the OP. I'll shut up now
http://www.musicanddesign.com/Equivalent_Baffles.html
My path of least resistance is simply to build and measure, with some educated guess from past experience. Wood cost only $3 anyway. YMMV
Again, I can't recommend enough JohnK's modelling tool (ABCDipole). Great $14 investment I think.
Anyway sorry for being off-topic to the OP. I'll shut up now
gainphile
Thank you for your suggestions
ABCDipole is probably a bargain, I must get to it soon.
However JK says rectangular baffles can be modelled by an equivalent circular baffle - at *low Hz, which is useful to know, but at present I was focussing on the upper mids to tweeter transition . .
The article doesn't clarify the measurement or estimate of D . .
Cheers
Thank you for your suggestions
ABCDipole is probably a bargain, I must get to it soon.
However JK says rectangular baffles can be modelled by an equivalent circular baffle - at *low Hz, which is useful to know, but at present I was focussing on the upper mids to tweeter transition . .
The article doesn't clarify the measurement or estimate of D . .
Cheers
I realize I largely "hadn’t seen the forest for the trees".
If I understand correctly, JohnK's article and the Edge can estimate the extra distance the sound travels - perpendicular to (at 90 degrees to) the listener–speaker axis. And that’s probably the larger part of the extra distance travelled.
My sketch was focusing on the extra *depth ie front-to-back that the sound from the rear of the cone travels, to get to the same ‘plane’ as the sound from the front of the cone.
So we can use JohnK's article or the Edge for the main part (around the bafffle's width): can someone clarify how to quantify the extra *depth travelled?
If I understand correctly, JohnK's article and the Edge can estimate the extra distance the sound travels - perpendicular to (at 90 degrees to) the listener–speaker axis. And that’s probably the larger part of the extra distance travelled.
My sketch was focusing on the extra *depth ie front-to-back that the sound from the rear of the cone travels, to get to the same ‘plane’ as the sound from the front of the cone.
So we can use JohnK's article or the Edge for the main part (around the bafffle's width): can someone clarify how to quantify the extra *depth travelled?
^ Grandinote - Mach 9
9 bass drivers with no xo, no eq and 16 tweeter in-line, height of 159cm
Avoiding comb filtering is nonsense, read more in other threads here
Line Array suggestions - 2 way
The making of: The Two Towers (a 25 driver Full Range line array)
9 bass drivers with no xo, no eq and 16 tweeter in-line, height of 159cm
Avoiding comb filtering is nonsense, read more in other threads here
Line Array suggestions - 2 way
The making of: The Two Towers (a 25 driver Full Range line array)
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.