Does anybody know if the plastic phase shield on the DQ25SC tweeter is actually useful or not? Is it just a physical protection device? I'd like to know if I can remove the entire face plate without affecting functionality.
I popped off the heatsink on the back and I plan to turn down the faceplate on a lathe to reduce the dipole offset when mounted back-to-back.
It looks like this tweeter, used in back-to-back pairing, could be a valid (or superior) alternative to the Neo3 for a dipole tweeter.
If you want to try this yourself (these tweeters are not that expensive), do this:
1) Unscrew the Philips screw located in the center of the heatsink
2) Make sure the Philips screw is fully loose - note that it is not likely to fall out on its own, nor does it need to
3) Using needle-nose locking pliers (Vice-Grips), grab a fin and twist it like a clock hand, keeping the face plate stationary. The fin will bend but the heatsink should snap off.
For those concerned about heat dissipation because the heatsink has been removed - a thin piece of aluminum bar can be inserted between the motors of the tweeters to pull heat away much more effectively than the heatsinks that come stock.
I popped off the heatsink on the back and I plan to turn down the faceplate on a lathe to reduce the dipole offset when mounted back-to-back.
It looks like this tweeter, used in back-to-back pairing, could be a valid (or superior) alternative to the Neo3 for a dipole tweeter.
An externally hosted image should be here but it was not working when we last tested it.
If you want to try this yourself (these tweeters are not that expensive), do this:
1) Unscrew the Philips screw located in the center of the heatsink
2) Make sure the Philips screw is fully loose - note that it is not likely to fall out on its own, nor does it need to
3) Using needle-nose locking pliers (Vice-Grips), grab a fin and twist it like a clock hand, keeping the face plate stationary. The fin will bend but the heatsink should snap off.
For those concerned about heat dissipation because the heatsink has been removed - a thin piece of aluminum bar can be inserted between the motors of the tweeters to pull heat away much more effectively than the heatsinks that come stock.
Last edited:
Some progress here:
https://picasaweb.google.com/104431754399072515246/August162011?authuser=0&feat=directlink
The overall depth of the tweeter is ~16 mm. The outer diameter of the tweeter is 43 mm. The distance from the center of the tweeter to the face edge is 21.5 mm. The distance from the OD of the modified tweeter to the voice coil (outside of the dome) is 8 mm. The dipole spacing should then be 16+21.5 = 37.5 mm not including half the thickness of a aluminum bar for mounting and heat sinking (say, another 2 mm). This leads to a dipole peak of approximately 4400 Hz.
Compare this to the Neo3, which is 61 mm x 89 mm (applicable raw dimensions), averaged to 75 mm, gives a dipole peak of approximately 2300 Hz. So, we're looking at true dipole behaviour for almost an octave higher, with better power handling, higher maximum SPL (probably), and higher sensitivity, for half as much for the pair. Also, you can mount this unit slightly closer to your mid, to reduce vertical lobing, than you can the Neo3. Winner?
I am concerned with diffraction off the proximal circular edge, which is why I added the relatively generous 45 degree chamfer, but most likely the best way to go is to wrap the unit with some open-cell foam or felt.
Honestly, I don't think you can find a smaller good tweeter on the market right now... this tweeter, for $13.25 a pop at Madisound, is a serious beast.
https://picasaweb.google.com/104431754399072515246/August162011?authuser=0&feat=directlink
The overall depth of the tweeter is ~16 mm. The outer diameter of the tweeter is 43 mm. The distance from the center of the tweeter to the face edge is 21.5 mm. The distance from the OD of the modified tweeter to the voice coil (outside of the dome) is 8 mm. The dipole spacing should then be 16+21.5 = 37.5 mm not including half the thickness of a aluminum bar for mounting and heat sinking (say, another 2 mm). This leads to a dipole peak of approximately 4400 Hz.
Compare this to the Neo3, which is 61 mm x 89 mm (applicable raw dimensions), averaged to 75 mm, gives a dipole peak of approximately 2300 Hz. So, we're looking at true dipole behaviour for almost an octave higher, with better power handling, higher maximum SPL (probably), and higher sensitivity, for half as much for the pair. Also, you can mount this unit slightly closer to your mid, to reduce vertical lobing, than you can the Neo3. Winner?
I am concerned with diffraction off the proximal circular edge, which is why I added the relatively generous 45 degree chamfer, but most likely the best way to go is to wrap the unit with some open-cell foam or felt.
Honestly, I don't think you can find a smaller good tweeter on the market right now... this tweeter, for $13.25 a pop at Madisound, is a serious beast.
Last edited:
With rigid dome tweeter, cancellation at about 15khz-20khz would occur due to the phase induced distance between center and edge (metal cones are pistonic). The phase shield would delay the center radiation much like phase plug in waveguide.
At the cost of stored energy, of course.
Some investigation proved that WITHOUT phase shield is better
http://www.diyaudio.com/forums/multi-way/190417-vifa-dq25-phase-shield-removal-surgery.html
Interesting. I don't know where you draw the conclusion that the off-axis response is superior without the shield than with; the responses look very similar... unless you are referring to the elimination of the peak in the top octave in the off-axis response.
How would you recommend removing the shield? I couldn't figure out how to remove the face plate without cutting into the glue, and that I did not want to do.
How would you recommend removing the shield? I couldn't figure out how to remove the face plate without cutting into the glue, and that I did not want to do.
Last edited:
Yes, the nasty peak.
With great care
. I managed to dent one of the dome. They are fragile. The plastic is very hard so perhaps a hot knife is better proposition than what I used (pliers).So, keeping in line with the low cost of the tweeters, I think these would make a good match from 1800-2000Hz and below:
Aura NS6-255-8A 6" Paper Cone Neodymium Magnet Woofer
High sensitivity and a relatively small rear footprint. Unfortunately I cannot find any data on the distortion figures but the NRT motor is fairly reputable. Also, the other drawbacks, like the stamped frame, can easily be fixed by 1) magnet mounting, and 2) damping.
If only the paper gasket on the flange weren't so damn ugly...
Aura NS6-255-8A 6" Paper Cone Neodymium Magnet Woofer
High sensitivity and a relatively small rear footprint. Unfortunately I cannot find any data on the distortion figures but the NRT motor is fairly reputable. Also, the other drawbacks, like the stamped frame, can easily be fixed by 1) magnet mounting, and 2) damping.
If only the paper gasket on the flange weren't so damn ugly...
Last edited:
Huh, pliers.Yes, the nasty peak.
With great care
I was thinking of using a Dremel (rotary cutter) to do mine.
Unfortunately the dipole spacing has to include the distance from the front dome plane to the back dome plane, making it 37.5 mm plus another 16 mm.
Gainphile and me are expecting another entry into the klingon-battleship-sonogram-gallery
, but it should be well worth the effort.Rudolf
Are you sure? What if you were to look at it from the perspective of the rear tweeter? Is its cancellation distance now only 21.5mm? Or also 37.5+16, so that both sound waves overlap for some distance before affecting each other?Unfortunately the dipole spacing has to include the distance from the front dome plane to the back dome plane, making it 37.5 mm plus another 16 mm.
Gainphile and me are expecting another entry into the klingon-battleship-sonogram-gallery
Rudolf
It looks like in 'd', 'e', and 'f', "D" should actually be "D/2", since in model 'd' the distance is from center to center whereas in subsequent models, it goes to half that distance.
Last edited:
I suppose what I'm trying to say is that if you pick the boundary plane to be at a position not equidistant to the front and back sources, the sound waves will not be 180 degrees out of phase upon reaching the boundary plane.
Those illustrations can be misleading
The effective dipole distance is not a driver property in the first place, but the difference of two distances: (1) from the front source to the ear and (2) from the rear source to the ear. If you take the front plate of the front tweeter as the start of the first distance, you have to go from the rear dome center to the edge of the rear front plate and then all the way to the front of the front plate of the front tweeter.
From there both distances to the ear are the same, whether you measure from the center of the front plate or the rim.
I consider the frame of the AURA as open enough. I see absolutely no problem up to 1 kHz and some deviations of up to 5 dB in the 1-2 kHz region. Would try not to exceed 2.5 kHz as Xover.
Rudolf
I agree that the dipole distance is not a driver property but a property inherent to the physical geometry of the transducer(s) and its environment.Those illustrations can be misleading
The effective dipole distance is not a driver property in the first place, but the difference of two distances: (1) from the front source to the ear and (2) from the rear source to the ear. If you take the front plate of the front tweeter as the start of the first distance, you have to go from the rear dome center to the edge of the rear front plate and then all the way to the front of the front plate of the front tweeter.
From there both distances to the ear are the same, whether you measure from the center of the front plate or the rim.
Rudolf
However, the differential distance between front and back waves at any given point (which you are describing, I think), is NOT the depth of the whole unit plus the radius of the unit.
For example, take a Neo3 where its half-width is 3 cm, zero depth, and the mic position is is 4 cm away, on-axis, on-center. According to your calculations, the dipole distance is 3 cm + 3 cm = 6 cm. However, it should be 3 cm (for the rear wave to wrap around), plus the difference between the front wave to mic and the rear wave (after coming around the edge) to mic, which is sqrt(4^2+3^2) - sqrt(5^2-3^2) = 1 cm. That means the difference in distance for the rear vs front wave is actually 4 cm, not 6 cm.
What were we talking about again? Ah, yes, the determination of the dipole peak. I still think it's based on 37.5 mm.
Last edited:
I see, somebody is trying to be even more nit-picky than me.
Where did I say that???
In all other respects your calculations seem to be right. But I wouldn't consider 4 cm in front of a Neo3 a useful listening distance.
Last edited:
It was just so that we could get the easy right-triangle calculation of squares. The principle still holds for any distance and width.I see, somebody is trying to be even more nit-picky than me.
Your calculations are right, of course. But I wouldn't consider 4 cm in front of a Neo3 a useful listening distance.
I didn't want to belabor the point anyway - it's just that I don't understand why I would need to consider more than the halfway mark when it comes to finding out the inherent dipole offset.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.