THE IMPORTANCE OF TIME DOMAIN RESPONSE MUST READ
Speaker designers ignore or downplay the importance of TIME. Why?
A high end speaker should be as accurate as possible and that means it should not only be optimized with regard to frequency response but time response.
Back in the 70’s and around that time, speaker engineers thought that a perfect speaker would be one that had a flat response. This idea has waxed and waned in popularity over the years and even now there is no consensus.
What the speaker engineers forgot to consider is Time response.
The time reponse of a speaker is how fast it starts and stops. A perfect speaker would have a perfect time response of 0. Since this is not possible, we must get as close to it as possible. The problem is speakers engineers have neglected this aspect of the design and so speakers over the last 40 years have not improved in this respect.
Time is such an important aspect of the sound we hear. We not only hear tone but also time. The brain can detect time differences of only a few microseconds. Experiments have shown that the start of each note is what we use to determine what instrument is producing that sound.
We must ensure that our crossovers do not smear the time response because it will be heard by our ears. Time inaccuracy is why high end speakers do not sound like real instruments.
Diffraction from the cabinet can also cause time smear. We need spherical cabinets not square boxes. Tweeters need to be time aligned in order to ensure that when the woofer stops so does the tweeter. When the woofer starts, so must the tweeter. The woofer itself has to have a Qts of ZERO to prevent time smear. Ports must not be used or else you will get ringing.
We need to make it mandatory for speaker companies to publish the time response of all their speakers so that consumers can easily compare and decide exactly what they want. Some may actually prefer a speaker that has a poorer time response and that is fine. The problem is, we cant decide unless we know what we are buying can we?
Unfortunately, 90% of speakers on the market, even high end speakers have ports. And they are also made of cheap wood, even though there must be better materials by now. Some materials ring more than others.
So dont be deceived folks. If you want better speakers, you will probably have to make them yourself because speaker manufacturers dont care about sound quality. They spend millions of dollars on anechoic chambers all so that they can get a flat response but they spend zero effort on better time domain response. We are being duped.
Ole Kenjt needs to get a can of tomato soup and some super glue, head on over to his local Art Museum and do a proper protest. Spread soup on a Masterpiece and glue yourself to the wall. Bet the speaker manufacturers will listen right up. Well as much as we are anyway. Working well for the Environmentalists, not really |
@kenjit earlier you were talking about:
Do you have the square wave measurements to show for your speaker? Or is it a Gedanken design? |
Speaker starting and stopping. A large factor is also the Damping factor of the amplifier and the amount of back emf produced as the coil is being energized by power going through it and at the same time producing its own back current. Electrostatic won't have that issue. I believe that is a larger factor to be considered depending on the speaker matching the PA. |
Time domain is much more complex than just the QTS of the speaker. QTS is related to the magnetic intensity of the magnetic assembly of a speaker but also related to the mechanical Q. The lower the magnetic field is the higher the QTS of a speaker. Also if you increase the stiffness of the suspension the Volume displaced by the speaker is reduced and the resonant frequency is increased as the QTS of the speaker, so it's an electro mechanical relationship. Now time domain should relate to a wide band response. If for rxemple you have a two ways, the two point of emissions should arrive at the ears at the same time, so they are in phase and time correct. But it's important to consider the behavior of the cross-over (inductive and capacitive parts. 6db slopes allowing for a better time domain behavior (less phase rotations) than 12 or 18 db slopes. The position of the tweeter and woofer have to be such that cross-over frequency and behavior are taken in account. Introducing a square wave and measuring it through a microphone (able to reproduce it) can help to finalize the position of the tweeter. The ear is very sensible to time domain by the way |
@armagedon36 ^agree^
But… the OP at one point was talking more about cabinet ringing after the fact, and having a death grip on the driver with aa QTS=0 and did not seem to care about the “before” part. At least in chronological order, worrying about what happens later, is a bit It is probably best to start off right, and not have it go all down the sewer with cabinet ringing and resonances later. (IMO)
But we do not know what the OP was thinking as they went silent pretty quickly. |
Thanks for that! So happy to hear Fritz is a nice as I have imagined from everything I’ve read. I’ve considered reaching out to him, but never been comfortable asking technical advice from a manufacturer/anyone in the industry I don’t know. As far as caps in his circuit, I was going by the links and what he wrote about his crossovers. I’d be curious what purpose a cap would serve in a crossover beyond protecting the tweeter. Unless it’s being used in a Zobel network on the woofer? While it works like a series crossover, don’t know I would refer t9 it as simply a series crossover. If it were, the designer would not have been able to get a patent on it. I’m a big fan of the approach. Have a way to go with getting a better understanding of its implementation (which is why I have wanted to reach out to Fritz), but seeing as how there is a commercial product I’m working on at the end of this, didn’t feel it was appropriate. Too bad this conversation has to happen in between the kenidjits ridiculous self aggrandizing rhetoric…
|
Caps and coils are fundamental to creating high and low pass filters. Here’s a post I wrote a long time ago that may help you, though it covers the more common parallel crossovers:
I know nothing about Fritz, and any patents or licensing involved, but patent reviewers are not omniscient or even domain experts. It is not that hard to patent BS or to create a patent which is a duplicate of another patent just reworded. |
If I'm not mistakebn, there has been a few different patents variations of series crossovers. I haven't looked in a few years, but I believe Eric Alexander of Tekton filed the last patent variation.... Either way, yes, the Acoustic Reality crossover is one of the more simple versions. I have built many, many, many crossovers. I've taken a ton of measurements and can tell you that series crossovers are not necessarily the only way to go. But if you want to use less parts and are careful with inductance, they are typically very good and on the right drivers can be the best for some designs. |
Sorry, should have been more clear :) When using the ARSXO, I am struggling to see what purpose a cap would have on the woofer, and how it would be implemented. In a more traditional Cap/Res/Ind series or parallel crossover, sure, use of a cap is not a mystery. Just not in the ARSXO. Would love to see how that is done! Always wanting to learn new stuff… |
Im a minimalist for sure :) Building a speaker for a friend, that is specifically intended to go into a shelf in a Ikea kallax shelving unit. 335mm cube. Not fun… So, small sealed full range driver and the rest is for a band pass sub. The boxes are doing most of the work, and the drivers were chosen for their suitability in the boxes. Even so, there are six components on the FR driver just to manage it. There are no”crossover” parts per say as the drivers/enclosures are doing the high pass/low pass work. But I digress :) Guess that was to illustrate how I very much prefer fewer parts. |
@erik_squires - one of the links mentions 1st order is “as good as it gets without going active filters and bi amping.” Do you have opinions on:
|
Hey @Holmz - Not sure teh link- / context of what you are referring to. Personally for home use I am not religious about active vs. passive crossovers. Anyone who has used a subwoofers has already at least dabbled in active filtering. The biggest technical superiority of active speakers is power efficiency, something you need most in large venues. At home, simplicity matters a great deal to me. My main speakers are custom built, passive crossovers. I'm thinking of making a high end center, which would be 3-way, active. DSP vs. Analog - With any speaker you have to make a crossover which, when added to your speaker's response, sums to an ideal end-result. DSP based solutions really are the easiest, and add the ability to time align the drivers. I like my stereo. Mytek DAC, Luxman Integrated, 2-way stand mounts. For me to go fully active I'd give up a great deal and add complexity. Not willing to make that trip from here. That doesn't mean I'm averse to the idea overall. |
Just to get back to the original theme of this, this thread prompted me to double check my speakers were both evenly raked. Come to find that they were off by an 1/8th of an inch.
Got them both exactly the same and what I thought sounded great sounds even better. Never discount the small things that are easy to overlook or forget about. That tiny adjustment made the speakers disappear even more and cleaned up the Soundstage. |
They are doing the best they can. Why don’t you build a perfect pair of speakers and start selling them @kenjit ?? |
