Member
Joined 2003
Answered in post 3. VituixCAD.
Why? Single axis of design is not enough information for complex radiation patterns. Any speaker really, but especially so for weird omnidirectional stuff.
Last edited:
Member
Joined 2003
a speaker driver is axis symmetrical, whether pointed forward or at the ceiling. .
Measurement axis is relative to the driver axis, not listening axis. Is the driver round or rectangular? Then 2 axis is enough. Speaker is rotated vertically in the crossover design section post-measurement. As for weighting, I disagree. Weighting of CTA-2034 is appropriate for speakers in a room, regardless of radiation pattern.
In any case, if you disagree with CTA-2034, within the calculator tool in VituixCAD you can take any number of measurements and average them with the weighting of your choosing.
Last edited:
Thanks DcibeL, I’ll have a look at the VituixCAD method.
In the meantime, what do people think of this idea?
Position the speaker in the room, affix the microphone to the speaker with a 2m tether then move the microphone around the listening area on a 2m radius. I thought this might ensure that sound levels aren’t accentuated by the microphone being sometimes closer to the speaker.
In the meantime, what do people think of this idea?
Position the speaker in the room, affix the microphone to the speaker with a 2m tether then move the microphone around the listening area on a 2m radius. I thought this might ensure that sound levels aren’t accentuated by the microphone being sometimes closer to the speaker.
Member
Joined 2003
In the help file for VituixCAD you’ll find instructions for measurement for REW, ARTA, SoundEasy, or Clio. I would follow that process.
Did anybody agree on how to measure a speaker or what to measure it with and why? I agree that everybody disagrees
and this makes my head hurt.
I question a tweeter facing a woofer and people are talking about measuring? I can't think of a worse configuration.
Is there a wave guide on the bass driver? You just designed a way to distort the tweeter. Vibration is bad!
If the tweeter is convex it is better but concaved, planar, or ribbon you are asking for distortion and driver failure
issues. Any wave guide will collect bass frequencies. It is a lose lose design on it's own.
Flipping the tweeter over (pointing up) on the other hand and using a wave guide on the bass driver. You might have something. Just a thought. I know arrival times will be the issue. Everything is 90+ degrees off axis. The first point of reflection on the tweeter would be the ceiling not the woofer. The woofer would be the distance from the wave guide,
to the cone face. It will probably work.
Fruit for though. I would work out a different design that is subject to more conventional measurements then there is
not a question of the whacko reading your are going to get with that design.
Some designs have to work in your head. Whales don't fly and I'm not pretty.
Those are facts.
Regards
and this makes my head hurt.
I question a tweeter facing a woofer and people are talking about measuring? I can't think of a worse configuration.
Is there a wave guide on the bass driver? You just designed a way to distort the tweeter. Vibration is bad!
If the tweeter is convex it is better but concaved, planar, or ribbon you are asking for distortion and driver failure
issues. Any wave guide will collect bass frequencies. It is a lose lose design on it's own.
Flipping the tweeter over (pointing up) on the other hand and using a wave guide on the bass driver. You might have something. Just a thought. I know arrival times will be the issue. Everything is 90+ degrees off axis. The first point of reflection on the tweeter would be the ceiling not the woofer. The woofer would be the distance from the wave guide,
to the cone face. It will probably work.
Fruit for though. I would work out a different design that is subject to more conventional measurements then there is
not a question of the whacko reading your are going to get with that design.
Some designs have to work in your head. Whales don't fly and I'm not pretty.
Those are facts.
Regards
Member
Joined 2003
Ask the general public to agree on something? Good luck with that. Closest I've found is some smart industry professionals got together and created a standard for indoor loudspeaker measurement and data representation, but still here there appears to be disagreement...
IMO omnidirectional speakers are inherently problematic, but that isn't the topic at hand, it's how to capture the response. Have you any input on this topic?
Reading the OP, I don't understand the tweeter to face the woofer, it sounds more like Linkwitz Pluto sort of arrangement.
Last edited:
Hi, I like your string tether idea, but not sure how much keeping the exact distance away matters.
Trying to measure an omni has driven me crazy. I built a dodecahedron, and still haven't figured out the best practice.
I've tried moving mic and multiple mic measurements' averaged .....over listening window, all around the speaker, all kinds of ways.
And still end up tuning by ear lol.
Oh, by measure...i take it you mean transfer function, right? Not RTA...?
Transfer is what I've tried.
Although the more I think about it, the more omni a speaker truly is...well, I'm beginning to wonder if an RTA isn't actually the best measurement...
(i've been taught the best use of a RTA is a doorstop lol, so the idea of using an RTA for an omni comes hard...)
I know this will cause some controversy, but i just cant see any of the sim programs being worth a dang for an omni, above very low frequencies.
OP, That is what I am saying, there is no correct way to measure that design. There was no way to measure the first speaker designs either. Once they sounded close, then they started measuring? The chicken or the egg?
By design you have a bass driver facing a tweeter. I can't get that out of my head. I have never seen that design without a wave guide over the top of the bass driver. Linkwitz Pluto is one of a dozen designs. ABS speakers?
The stuff I've seen in the name of cabinet designs and just being weird. I'm all for weird, ugly to for that matter, where would we be without ugly and weird? I like big and heavy too. So, Big, ugly, weird, heavy speakers. I have several pairs. I also have one next door. He's bald too.
Regards
By design you have a bass driver facing a tweeter. I can't get that out of my head. I have never seen that design without a wave guide over the top of the bass driver. Linkwitz Pluto is one of a dozen designs. ABS speakers?
The stuff I've seen in the name of cabinet designs and just being weird. I'm all for weird, ugly to for that matter, where would we be without ugly and weird? I like big and heavy too. So, Big, ugly, weird, heavy speakers. I have several pairs. I also have one next door. He's bald too.
Regards
How is that so difficult to visualize?
https://www.linkwitzlab.com/LXmini/Introduction.htm
For off-axis measurements, the best way is to set the speaker on a stand that sits on a turntable made of two plates bolted together. Plates must be so big that the driver (radiating plane) is at centerpoint. Cut the upper plate hemispheric and make markings of degrees. Then the mic just stands still, and reflections from floor and room are standard. No need to make it automatic. But be careful to not to make a lip just below the speaker (it makes awful diffractions)
Ground planemeasurement for bass reflex tuning, high stand and gated measurement for mid-tweeter xo. RTA or MMM for final voicing (that's what Carlson was doing at home, but he lacked modern understanding of radiation patterns)
https://www.audiosciencereview.com/...pinoramas-with-rew-and-vituixcad.21860/page-9
Last edited:
To answer a few questions:
I’m building a passive crossover.
I’m building a passive crossover.
I was thinking that for X/O simulation it would be important for the woofer and tweeter levels to have the correct relative sound level difference.
Member
Joined 2003
Depends on the specific design, balancing radiation patter that's all over the map from varying speakers pointed in varying directions, too much reflected energy makes for a lot of ambience with complete loss of imaging. I don't personally get the benefit of pointing a woofer at the ceiling, but a lot of people like the Linkwitz pluto so who am I to judge, everyone likes different speakers for different reasons, maybe omni is your preference, don't let my opinions get in your way 🙂.
In any case, keeping constant mic distance for measuring speaker for design is crucial. Most people will keep the mic stationary and rotate the speaker, depending on the speaker I understand that could present challenges of its own. Moving mic within the room will be difficult to impossible to address the room reflections as a moving target. I'd suggest reading through measurement guides for VituixCAD, found in the help here, there's a lot of good information in there:
https://kimmosaunisto.net/Software/VituixCAD/VituixCAD_help_20.html#How_to_start_with_VituixCAD
Maybe the method needed is different, and it's difficult to work out. We need to think geometrically.
This method is geared toward ranking conventional existing speakers. When designing speakers we can talk about the limitations.
Have you tried this? What if you want to work into something other than free space. Something where the standard Vituixcad measurement method doesn't work?
Member
Joined 2003
The standard is for providing the performance of a speaker in a room in a common format. "conventional" or not does not matter.
I used it for different reasons than this topic. Use the calculator tool, load whatever set of response data you like. Select "average of A responses". To apply weighting, use the scale dB for each response. The complexity in this process is that the sum of scale dB values should be equal to the number of loaded responses in order to get the correct SPL scaling of the result.
For example, for 4 responses with scaling dB values of +6, 0, -6, -6, weighting is 2,1,0.5,0.5 = 4.
Clearly, it can be used. This was not in doubt.
This is not the same as developing a scheme for a non-standard speaker in a non-standard radiation space, and seeing that Vituixcad is capable of accepting the scheme. I have had this discussion previously and am no further towards a solution.
This is not the same as developing a scheme for a non-standard speaker in a non-standard radiation space, and seeing that Vituixcad is capable of accepting the scheme. I have had this discussion previously and am no further towards a solution.
Yes, drivers must be given same voltage, to reveal difference in sensitivity. 2-channel measurement system is needed to give exact acoustic delay (time of arrival of sound) difference.
Please remember that the woofer will have 6db SPL loss in bass (Baffle step/loss) that must be compensated, which means that in most cases you must attenuate tweeter. Upfiring woofer is giving extra challenge for design, because it's response at 90deg is very different from typical on-axis used for simulations and design.
LXMini measurements by Erin #bikinpunk https://www.erinsaudiocorner.com/loudspeakers/linkwitz_lx_mini/
This speaker uses a 4" fullrange as "tweeter" to get crossover below 1kHz
Ok I got it through my thick head they are not facing each other. I've seen this a few times. ABS speakers.
Fun for sure.
I went back and read the OP. I don't know why I was thinking the drivers were facing each other. It's a sign.
Regards
Fun for sure.
I went back and read the OP. I don't know why I was thinking the drivers were facing each other. It's a sign.
Regards