Tux,
This measurement was indeed nearfield 12 in. This was back before I had a proper mic and sound card so the PC ADC low passed at 6 kHz. I guess my question is even if I low passed the woofer at say 300 Hz, if I applied a step voltage to the system wouldn't the rise time still be 0.5ms? Or are you saying the fastest rise time is now dictated by the lpf?
This measurement was indeed nearfield 12 in. This was back before I had a proper mic and sound card so the PC ADC low passed at 6 kHz. I guess my question is even if I low passed the woofer at say 300 Hz, if I applied a step voltage to the system wouldn't the rise time still be 0.5ms? Or are you saying the fastest rise time is now dictated by the lpf?
The rise time (talking step response here) is not affected by the LPF afaik. But, the IR is affected by the LPF and also by the Vifa. So it's not appropriate to look at the IR and make a judgement about the driver performance. Because you'll be judging the Vifa.
If you were to do step response measurements, then you could compare the 15" woofer to your 6.5" woofer. But then, who is to say the 6.5" woofer would have a better step response. What if the 15" woofer has a massive motor?
Also consider that step response is not a very good indicator of performance. It's basically doing this: Connect a 9V battery to the woofer terminals. Pop! The cone pops out really fast. Do it to a different woofer. Pop! That cone pops out just a bit faster. But... Music (sine waves) don't do that. Yaaa, I'm over simplifying it. But the point is there are much much more important data sets to look at than step reponse. Or ven IR for that matter. But I would rather look at an IR than a step response. At least you can estimate CSD, polarity, reflections, etc.
If you were to do step response measurements, then you could compare the 15" woofer to your 6.5" woofer. But then, who is to say the 6.5" woofer would have a better step response. What if the 15" woofer has a massive motor?
Also consider that step response is not a very good indicator of performance. It's basically doing this: Connect a 9V battery to the woofer terminals. Pop! The cone pops out really fast. Do it to a different woofer. Pop! That cone pops out just a bit faster. But... Music (sine waves) don't do that. Yaaa, I'm over simplifying it. But the point is there are much much more important data sets to look at than step reponse. Or ven IR for that matter. But I would rather look at an IR than a step response. At least you can estimate CSD, polarity, reflections, etc.
I would argue that musically, percussive instruments like piano, kick drums, bongo drums, high hat, triangles, do exactly this when struck by a hammer or drum stick. This is why you want nice IR and rapid rise time. I think you are right that the Vifa will dominate the IR. if music were just steady sine waves then IR would not be a consideration.
Those instruments are comprised of sine waves. If you want them to sound snappy, then you want speakers that have extended frequency response, flat frequency response, as little compression as possible, well damped room, and low distortion. For this you do want a nice impulse response.
So if the Vifa dominates the IR, then you agree that choosing the driver to use should be based on which will provide the speaker with the above criteria, not what the IR looks like?
So if the Vifa dominates the IR, then you agree that choosing the driver to use should be based on which will provide the speaker with the above criteria, not what the IR looks like?
Those percussive instruments if you look at the mic trace in real time with an O-scope start out with a step function that decays at the oscillating summation of the sine waves and their harmonics. A kick drum has no real frequency - it is a puff of impulse pressure wave followed by very weak sine wave oscillations of the membrane frequency but no one cares if that is 50 or 55 Hz. I don't think the Vifa (at least not one) can recreate the visceral punch of a kick drum. Four of them together in unity horn however is a different story and you can feel the punch of a kick drum. Also, an array of 25 Vifa's like the Roger Russel speaker can also provide a substantial impulse kick that rivals a 15 inch woofer, and I would argue has a better impulse response to boot.
I don't think we disagree other than that a low mass cone can accelerate faster and produce a more rapid punch (rise time or steepness of the pressure wave) given the same motor strength. Whether or not this can be judged by the IR is where I think you can see it when the woofer is driven full range.
I don't think we disagree other than that a low mass cone can accelerate faster and produce a more rapid punch (rise time or steepness of the pressure wave) given the same motor strength. Whether or not this can be judged by the IR is where I think you can see it when the woofer is driven full range.
All right I think I'm bored now 
But one last kick at the can.
Can you show me this to be true? A source maybe. Cause I disagree.
I'm not sure why you're talking about using the Vifa to make kick drum sounds. I certainly wouldn't bother. You should use the 6.5" or 15" woofers. Not the silly 3" Vifa.
What's an impulse kick? And what is a better impulse response?
Could you please show me how to read an IR to see the cone acceleration / rise time? I'm already aware of how to see it on a step response, but I clearly can't read an IR for this. Thank you.
But one last kick at the can.Can you show me this to be true? A source maybe. Cause I disagree.
I'm not sure why you're talking about using the Vifa to make kick drum sounds. I certainly wouldn't bother. You should use the 6.5" or 15" woofers. Not the silly 3" Vifa.
What's an impulse kick? And what is a better impulse response?
Could you please show me how to read an IR to see the cone acceleration / rise time? I'm already aware of how to see it on a step response, but I clearly can't read an IR for this. Thank you.
I'm also bored of trying to explain that the IR as derived from the frequency response is a mathematical reconstruction (inverse Fourier transform) of the true time response to an impulse. The Fourier transform of the impulse response is the frequency response. They are representations of the same thing in different spaces. Time domain responses of microphones are not something people often measure because it takes a fast microphone, a high speed storage oscilloscope, and good lab practice to set up. I'm going to stop kicking the can now.
??? I've been saying that from the beginning. The FR is derived from the IR through the FFT. You don't have to explain that to me. But what this stuff about rise time is I dunno? And you can't seem to explain it either.
The microphone reads the IR and the software creates the FR from the IR. That's it. Both a 6.5" and a 15" driver are the same in this regard. One is not better than the other at it.
The microphone reads the IR and the software creates the FR from the IR. That's it. Both a 6.5" and a 15" driver are the same in this regard. One is not better than the other at it.
Post #20.
That's all there is to it. Pick the 6.5" for other reasons than impulse response.
That's all there is to it. Pick the 6.5" for other reasons than impulse response.
6 Driver Slot loaded OB
Here is a design I am thinking of: 12 in wide x 30 in tall x 10 in deep. I will probably add a rear H frame to this to extend the bass. The 3 drivers opposite the visible ones are mounted with basket facing in so that they can operate out of phase electrically but in phase acoustically to reduce distortion caused by hysteresis in cone movement.
Here is a design I am thinking of: 12 in wide x 30 in tall x 10 in deep. I will probably add a rear H frame to this to extend the bass. The 3 drivers opposite the visible ones are mounted with basket facing in so that they can operate out of phase electrically but in phase acoustically to reduce distortion caused by hysteresis in cone movement.
Attachments
Last edited:
The most ideal width of the baffle Rudolf came up with in his measurements were about half of that 12".
(not saying your goals should be the same or even similar)
Did you make it this wide to extend the bottom end of the Vifa?
Here's a nice design OB with very good response:
http://www.diyaudio.com/forums/multi-way/190789-fin-full-range-phase-linear-open-baffle-speaker-measurements.html
(The good response is done with lots of DSP but I like the transition to the small woofer and tweeter)
(not saying your goals should be the same or even similar)
Did you make it this wide to extend the bottom end of the Vifa?
Here's a nice design OB with very good response:
http://www.diyaudio.com/forums/multi-way/190789-fin-full-range-phase-linear-open-baffle-speaker-measurements.html
(The good response is done with lots of DSP but I like the transition to the small woofer and tweeter)
I am still playing with the design to get the woofers to go up to 500 Hz as the small 6 in baffle makes it hard for the Vifa to go much lower. The woofers actually can go up higher very well with the slot loading but it is not flat and will require DSP to flatten. Thanks for the link to the OB project with flat phase. That is what I want to work on next.
Narrow baffle for tweeter
Here is the latest evolution of this OB design, now with a 5 in wide baffle for the Vifa to improve polar dispersion and imaging. The slot woofer chamber is now 7 in deep x 3 in wide x 21 in tall with a 1.75 in wide x 21 in tall slot. The lower woofer baffle has a 7 inch deep wing. It is interesting that adding a 0.5 in lip extension surrounding the exit slot port really helps to smooth the mid-bass. I am now running the woofer with a low pass at 210 Hz (-12dB/oct) and the tweeter with a high pass at 500 Hz (-12 dB/oct). The Vifa has to be attenuated by -3dB to match the output of the 6 woofers.
Here is the response at 16 volts rms corresponding to xmax.
Here is the latest evolution of this OB design, now with a 5 in wide baffle for the Vifa to improve polar dispersion and imaging. The slot woofer chamber is now 7 in deep x 3 in wide x 21 in tall with a 1.75 in wide x 21 in tall slot. The lower woofer baffle has a 7 inch deep wing. It is interesting that adding a 0.5 in lip extension surrounding the exit slot port really helps to smooth the mid-bass. I am now running the woofer with a low pass at 210 Hz (-12dB/oct) and the tweeter with a high pass at 500 Hz (-12 dB/oct). The Vifa has to be attenuated by -3dB to match the output of the 6 woofers.
Here is the response at 16 volts rms corresponding to xmax.
Attachments
Sorry forgot to mention that the above is a simulation (thought everyone knew I was still in design stage) - says Akabak in top right corner of plot.
I have not yet built the slot baffle OB - drivers are out of stock until April!Regarding baffle width: the imaging on the Nataloss monitors is superb and the baffle is 5 in wide. I have not measured the polar with the 12 in wide baffle but would imagine it to not be as good. I am just adapting the design based on Rudolf's measurement of the Vifa in a narrow baffle.
One caveat: It is not the narrower baffle per se, which leads to better imaging. It is the compact off-axis sound pattern of the narrower baffle with less SPL radiating sideways, which leads to less unwanted reflections. If you sit 20 feet away from your speakers, the narrow baffle will make no difference, because you are already drowned in reflections. If you sit 5 feet away, the difference should be literally "re-veiling". For anything in between, it depends on your rooms acoustic quality. If your room is "dead", it won't make much difference. But if your room is "live", you will appreciate the improvement.
Rudolf
The wide baffle has always been an issue for me as it won't get much playtime if the WAF is low. I think a 12 in wide lower baffle and a 5 in wide upper baffle are very livable and that is sort of the goal in am aiming for now with average listening SPL's. I am still surprised by how low the efficiency of an OB is vs a sealed box driven by the same power. I guess it makes sense because the cone is spending a lot of power mixing the air on both sides of the baffle rather than projecting the wave. Now if only Parts Express will get those 6.5 in woofers back in stock.