Dr. Geddes,
if I understand you correctly - you are using some kind of compensation in the crossover based on measurements? Could you be please more specific? All I know are the "textbook" crossovers with some small tricks like using a different pole in a third order highpass for CD correction.
if I understand you correctly - you are using some kind of compensation in the crossover based on measurements? Could you be please more specific? All I know are the "textbook" crossovers with some small tricks like using a different pole in a third order highpass for CD correction.
gedlee said:
Svantes "The edge" is a popular software that fit measured results close to perfect. http://www.tolvan.com/products/ (down at the moment)
Some do and some don't. I guess that many feel that the end user is more interested about practicall performance than how the designer got there and what tools that were used to get there.
/Peter
Yes! I know that addictive sound you are talking about, how nice it would be to be able to listen to it at home, at will...radianceaudio said:
I didn't think Orions were even close to re-creating that sound...
Closer to adding electronically adding extra reverb to the recording with a Lexicon, or whatever, plus detail was lost, comb filtering...
But hearing a good five-way horn system reproduce this sound, and a sympony, and jazz, and drums, from the faint rustle of the turning of a page of sheet music to the loudest crescendo, without compression, and being able to distinguish all the voices, individual parts within massed chorals, individual instruments...
That's what sold me on good quality, multi-way horns.
Well designed, without coloration, excellent imaging, soundstage, and unreal dynamics.
All this with a 5 watt amp, not even turned up...
Not using inefficient 100 dB / w drivers, that require so much excursion to produce sound, and produce so much distortion to begin with...
pelanj said:
This is the problem with "textbook" crossovers - they never actually work right. I don't use "textbook" filters so there is no way to put what I do into a textbook context.
I use whatever slopes and damping etc. produces the curves that I need. There are a multitude of components so there are a lot of degrees of freedom. Basically, there is an additional boost to the upper end of the woofer from the enclosure, what you would call a "baffle step", but its more than just the baffle. This requires the electrical low pass on the woofer to be lower than the "textbook" frequency where the LP and HP are approximately equal. In a Summa the LP is at about 700 Hz electrically, but about 900 Hz electro-acoustically. The HP ends up being at about 1 kHz. Clearly not "textbook".
Rybaudio said:
john k... said:
markus76 said:
I had the chance to take some measurements of the NaO Mini, without rear tweeter. I put up a short web page on them. I also added an old measurement of the NaO II with and without the rear tweeter turned on to show the effect at 90 degrees off axis. Also included is a measurement of the reverberant field in my design room. The reverberant field data indicates that the directivity index (DI) is relatively constant up to about 5K despite the change in polar response. Note that DI is an indication of the power radiated by a source compared to an omni directional source with the same on axis SPL. System with constant polar response will have constant DI, but constant polar response is not a requirement. For example, a dipole and cardioid have very different polar response but identical DI.
I would anticipate that such plots for the Orion would appear very similar to the NaO Mini. I can not speak for the Orion + with rear tweeter because the rear tweeter was added as an after though and not an integral part of the design as was done for the NaO II.
I stumbled across this thread and was struck by this post. Sorry for my late reply, but I couldn't resist. I am very amazed that you cannot measure the influence from the backwave on the movement of the speaker cone, even when the stuffing behind the cone was removed. The backwave excites standing waves inside the box and they have a clear effect on the cone (and sound).gedlee said:
When building box systems and carrying out far field measurements to be able to do a correct crossover design (NOT textbook), I also include a close range measurement with the mic one inch in front of the cone. The close range measurement can be done at a high resolution, even in a reverberant listening room (SPL from room reflections is very low when compared to the sound directly in front of the box at one inch, especially when measured above 100Hz). Doing this measurement, you can exactly see at which frequencies the standing waves in the box are. You don't have to put your mic inside the box, just place the mic in front of the cone. This way you can put just enough damping material in the box to eliminate the standing waves up to the point that is your design goal (also by looking at the waterfall plot). If desired, I can post measurement data.
Ofcourse, it could be that I completely missed your point and that you mean something else entirely.
John,
thank you very much for the data. But as we're talking about a dipole we would need to know what the pattern looks like from 0° to 180°.
I would remeasure my Nathan's any time but my room is too small to generate valid data at lower frequencies.
In the end we would know what kind of absorption within in the room is necessary to create a linear indirect sound field. I assume that this is one of the most important goals of any loudspeaker/room system.
Best, Markus
thank you very much for the data. But as we're talking about a dipole we would need to know what the pattern looks like from 0° to 180°.
I would remeasure my Nathan's any time but my room is too small to generate valid data at lower frequencies.
In the end we would know what kind of absorption within in the room is necessary to create a linear indirect sound field. I assume that this is one of the most important goals of any loudspeaker/room system.
Best, Markus
jeroen_d said:
The effect of the box on the cone is the stiffness that it provides and this is measureable. But at frequencies where the box could have standing waves I have not seen significant effects that I could absolutely attribute to the boxs internal resonances.
A nearfield measurement that shows aberations CANNOT be concluded to be from the box. These could be the spider (most likely in my experience) or the basket, or any number of things. Its just not that simple to find what is the box and what is something else. Some actual posts would be useful.
john k... said:
For the better part of a decade I've found your website to offer a lot of though provoking data on loudspeakers.
This is great stuff!
If you look at the measurements in your article, it makes a solid argument for the use of a waveguide down to around 2khz on your design. That would improve the polar response. The polar response from 1khz down is quite impressive.
Perhaps Emerald Physics is on to something here?
I have a nice example, the Peerless HDS134 midwoofer measured in a closed box. The closed box is a sort of tapered line of 20", going straight backwards behind the cone and then curling up with a discontinuity. The discontinuity is at 6.8" behind the cone, giving a reflection and standing wave at 900Hz. The tapered line lenght of 20" gives measurable standing waves at 340Hz and 680Hz. The measurement is carried out with an MLS signal, 51kHz sampling rate and 4096 FFT length. The mic is directly in front of the woofer at 1".
The first set of graphs (response and waterfall) is taken with no damping material inside the box. As you can see the waterfall diagram is horrible.
In the second set of graphs, the end of the line is stuffed with a small piece of open cell foam. This piece of foam at the small tapered end of the line kills very effectively the 340Hz and 680Hz standing waves. The waterfall looks at lot better, but the 900Hz resonance still is a problem.
The third set of graphs gives the result when the rest of the box is filled with longhair wool.
The first set of graphs (response and waterfall) is taken with no damping material inside the box. As you can see the waterfall diagram is horrible.
In the second set of graphs, the end of the line is stuffed with a small piece of open cell foam. This piece of foam at the small tapered end of the line kills very effectively the 340Hz and 680Hz standing waves. The waterfall looks at lot better, but the 900Hz resonance still is a problem.
The third set of graphs gives the result when the rest of the box is filled with longhair wool.
Attachments
In this situation you could be right. In my systems I've never seen anything like this, but then I would never use a long box directly behind the woofer for just this reason. When the box spans out laterally behind the woofer it is much less prone to standing waves that would cause this kind of effect.
I guess that the bottom line here is that one can setup a situation where this "box effect" will occur, but one can also just as easily design a box where it is negligable. I prefer to do the later.
I guess that the bottom line here is that one can setup a situation where this "box effect" will occur, but one can also just as easily design a box where it is negligable. I prefer to do the later.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.