Florian,
You are correct that the tweeter needs to be set behind the woofer so that their voice coils are located at approximately the same location. The two need to start rising at about the same time. It’s just that the woofer moves slower. So my solution was to put the tweeter on the bottom, assuming the ear is at the woofer axis. This introduces just enough offset similar to a stepped baffle.
You are correct that the tweeter needs to be set behind the woofer so that their voice coils are located at approximately the same location. The two need to start rising at about the same time. It’s just that the woofer moves slower. So my solution was to put the tweeter on the bottom, assuming the ear is at the woofer axis. This introduces just enough offset similar to a stepped baffle.
Hi xrk,
I do not agree with you. According to my post #2025 (see attached again), assuming that you listen at a distance of 1,5m = 5ft, the voice coils are not together when you listen on woofer axis. The woofers voice coils is 1,6in behind. So you are not listening with same distance for the two voice coils. Your solution is not enough offset, if you want both voice coils at the same distance. - But, that is actually not the point, I disagree with you, that the voice coils need to be at the same distance. In your case, your measurement or simulation, they DO start at the same time, but that means, that the top points of the step response are not together, as the woofer is slower, as you said. The point is, that the highest points of the slopes need to be together, and this means, that the woofer needs to start earlier! Then the measurement looks like my measurements what I did, with that bend in the rising slope. And then it sounds on the spot.
I do not agree with you. According to my post #2025 (see attached again), assuming that you listen at a distance of 1,5m = 5ft, the voice coils are not together when you listen on woofer axis. The woofers voice coils is 1,6in behind. So you are not listening with same distance for the two voice coils. Your solution is not enough offset, if you want both voice coils at the same distance. - But, that is actually not the point, I disagree with you, that the voice coils need to be at the same distance. In your case, your measurement or simulation, they DO start at the same time, but that means, that the top points of the step response are not together, as the woofer is slower, as you said. The point is, that the highest points of the slopes need to be together, and this means, that the woofer needs to start earlier! Then the measurement looks like my measurements what I did, with that bend in the rising slope. And then it sounds on the spot.
Attachments
Ahem....
So, looking at the graphs you provided earlier, I can see that your ears can distinguish a 0.5ms delay between the tweeter and the woofer.
I congratulate you on having golden ears that none of us will ever have.
In this regard, if you believe that X's design is flawed, by all means, please design a speaker that your ears will like and be done with it.
Start your own thread and people will either agree with you or not.
Have fun designing your perfect speaker! And remember to enjoy the music.
So, looking at the graphs you provided earlier, I can see that your ears can distinguish a 0.5ms delay between the tweeter and the woofer.
I congratulate you on having golden ears that none of us will ever have.
In this regard, if you believe that X's design is flawed, by all means, please design a speaker that your ears will like and be done with it.
Start your own thread and people will either agree with you or not.
Have fun designing your perfect speaker! And remember to enjoy the music.
Why are you unfriendly?
At the time I was writing this, I thought that my measurements are worse. I see it different today, one can differentiate even more, as I have explained in the last posts.
I not criticize xrk's design at all, I want to contribute to it. I took it more detailed, found something interesting and exciting, and want to tell you about it. You may try it yourself if you want. And I may tell you, that you will hear it, it is not subtle.
At the time I was writing this, I thought that my measurements are worse. I see it different today, one can differentiate even more, as I have explained in the last posts.
I not criticize xrk's design at all, I want to contribute to it. I took it more detailed, found something interesting and exciting, and want to tell you about it. You may try it yourself if you want. And I may tell you, that you will hear it, it is not subtle.
If you want to align the woofer and tweeter, they should play the same frequency. You can't compare a woofer (with crossover) IR to a driver that plays up higher. Make them play the same range and compare the timing of the peaks there.
Easy to do with some digital help. Just delay one of the drivers, make them play simultaneously and use a sweep in a range they can both play comfortably. Look at the distance between both peaks. Determining acoustic centers is best done acoustically. Not guessing based on voice coil positions.
Easy to do with some digital help. Just delay one of the drivers, make them play simultaneously and use a sweep in a range they can both play comfortably. Look at the distance between both peaks. Determining acoustic centers is best done acoustically. Not guessing based on voice coil positions.
Firstly, lets get this out of the way: you've heard something you like very much and no matter what any of us has to say about it, we cannot take that away from you. 
But... a Stereo setup isn't the way to judge acoustical offset between speaker drivers. Mono would have been better. With Stereo you 'suffer' from cross talk between both ears. Your left ear will hear the right speaker and vice versa. This creates area's of confusion, where that cross talk creates dips. What you may have done is find a position that clears up this cross talk happening at both ears. That would have the ability to actually sound better to one's ears.
You don't have to believe a word I say. Measure your stereo setup at ear position would remove all doubt that every (conventional) stereo setup suffers from this. No way around that. So while it may seem logical you've aligned the acoustical pulses of both drivers, something else may be going on. As said, quite often things are not as easy as they seem to be. Measurements can help with that. Personally I have a thread full of those as I do play with this phenomenon to get better sound. I have the measurements at ear positions to prove this to myself.
We're all free to believe what we want. You're free to shove my well meant opinion to the side. Enjoy what you've got! If you're curious enough, fire up a measurement suite and prove (to yourself) what it is that captivates you. That's what I do.
But... a Stereo setup isn't the way to judge acoustical offset between speaker drivers. Mono would have been better. With Stereo you 'suffer' from cross talk between both ears. Your left ear will hear the right speaker and vice versa. This creates area's of confusion, where that cross talk creates dips. What you may have done is find a position that clears up this cross talk happening at both ears. That would have the ability to actually sound better to one's ears.
You don't have to believe a word I say. Measure your stereo setup at ear position would remove all doubt that every (conventional) stereo setup suffers from this. No way around that. So while it may seem logical you've aligned the acoustical pulses of both drivers, something else may be going on. As said, quite often things are not as easy as they seem to be. Measurements can help with that. Personally I have a thread full of those as I do play with this phenomenon to get better sound. I have the measurements at ear positions to prove this to myself.
We're all free to believe what we want. You're free to shove my well meant opinion to the side. Enjoy what you've got! If you're curious enough, fire up a measurement suite and prove (to yourself) what it is that captivates you. That's what I do.
I used Stereo, as with Stereo I am able to judge the 3D capability and transparency, the live-like stereo image. With the right offset it sounds on the spot, the resolution is perfect. With the wrong offset the sound is smeared, not that crystal clear.
I have done measurements and shown what has happened/ is happening with my setup.
I have done measurements and shown what has happened/ is happening with my setup.
Which kind of proves my point. With a Stereo signal you wouldn't be able to align woofer and full range driver by ear. But you would be able to align them so imaging seems to give you the best effect. You're basically creating a form of EQ with small alignment tweaks that 'could' be helpful to clear up the phantom center perceived balance.
(i've decided to leave the effects of a stepped baffle out of this equation, but that too will have it's effect on the total frequency outcome)
The IR/STEP from XRK shows he has quite an acceptable alignment of the acoustical pulses of woofer and full range driver. Something you put at doubt, but the measurements show otherwise.
Just accept that you like what you've got, maybe a few others will give it a try.
(i've decided to leave the effects of a stepped baffle out of this equation, but that too will have it's effect on the total frequency outcome)
The IR/STEP from XRK shows he has quite an acceptable alignment of the acoustical pulses of woofer and full range driver. Something you put at doubt, but the measurements show otherwise.
Just accept that you like what you've got, maybe a few others will give it a try.
Last edited:
Florian,
I think what you have discovered is that a 0.5ms delay between left and right can make the image sound “wider and more open”. This is a well known trick with headphone EQ to remove the “singer in between your ears” sensation and bring them outside your head. Try that sometime with headphones and a DSP delay using Audacity or jRiver.
A perfect right triangle SR may not give the best ethereal soundstage image, but it’s not adding something that probably wasn’t there in the original recording.
I am glad you found an alignment that sounds good to you. Note that with a Harsch XO - the delay is much greater. Yet it still provides the impression of a transient perfect pulse.
I think what you have discovered is that a 0.5ms delay between left and right can make the image sound “wider and more open”. This is a well known trick with headphone EQ to remove the “singer in between your ears” sensation and bring them outside your head. Try that sometime with headphones and a DSP delay using Audacity or jRiver.
A perfect right triangle SR may not give the best ethereal soundstage image, but it’s not adding something that probably wasn’t there in the original recording.
I am glad you found an alignment that sounds good to you. Note that with a Harsch XO - the delay is much greater. Yet it still provides the impression of a transient perfect pulse.
I have a possibly obvious (and honestly, off topic) question: the best qualities of this speaker design (and probably most speaker designs) seems really dependent on position, so that the wavefronts from each driver arrive at your ears with the appropriate timing. So its hard to have the ideal experience unless you're in the sweet spot (and you setup the driver positions properly in the baffle).
Is something like the XBush the only way around this problem?
Is something like the XBush the only way around this problem?
I am sad that you don't just take me as what I have written. It seems, that if nobody else tries what I have found out and have done, and will tell here in the forum, that what I am writing is the case, that it will forget and will only stay with me. Perhaps it is the matter, that you do not respect me yet, respect my expertise. You try to find excuses and reasons what all it could be, instead of trying to follow what I am saying, taking me seriously. Another reason I could imagine, why you do not take me seriously just as is, is perhaps that I do not engage defensing and arguing with you. I do not know the professions of the members contributing here, it seems many are proud about their knowledge, but are not open to new perspectives and ideas. I myself have studied Tonmeister/ audio engineering, and I have worked as an acoustician and Tonmeister ever since, also modifying gear, and I know what I am talking about. And I know what you are talking about, but that is not the case. It is exactly very simple what I have told you: With the right offset the resolution is on the spot, without the right offset it is blurred. The offset does not add anything. That's me, that I question things and have a detailed look at them, trying to understand those things I am going in for properly. I cannot help you, except advising to be open minded, to be open for new ideas. I would be sad, if that all stays with me, because that is not worth anything if knowledge does not get spread.
Kind regards,
Florian M.
Just start a thread of your own. Explain what it is you want to teach us and don't forget to back it up with some proper science. This means detailed measurements that might prove your theory. How else should we take it seriously? Don't assume you found the holy grail, show the proof and be prepared to have a thick skin.
Basically your invading someone elses space here making claims that certainly aren't clear without the theory to back it up.
In your own thread you can try to convince others. It isn't that hard not to step on anyone other's toes. Try it and you might like it. I did... but maybe I don't take myself as seriously. Even though I live an breathe this stuff and lost count on how many measurements and experiments I did. (all documented in my huge thread, where I don't step on anyone's toes)
But still that doesn't make me right. But everyone is free to make up their own minds...
You can't force us, but you might be able to convince us... just start that thread
.