Announcement

Collapse
No announcement yet.

Offsets, Asymmetrical Slopes, and Mysticism - revisited *PICS*

Collapse
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • Offsets, Asymmetrical Slopes, and Mysticism - revisited *PICS*



    Maybe this should have been a blog entry? :rolleyes:

    Jeff B.

  • xmax
    replied
    Of course it matters what filter order you're using to combine two (or more) drivers. That's why when you have a system using LR2 slopes (acoustic slopes, not electrical) you reverse the polarity of one of the drivers in order to get proper summation at the crossover point. Similarly, when you have LR4 slopes (acoustic slopes, not electrical) you connect both drivers in the same polarity due to the extra phase shift.

    But when there is phase shift due to time of flight (acoustic offset problems) you can use slightly different transfer functions in the hope of making up some of that extra phase shift over a narrow range of frequencies in order to get better summation in the design axis at the crossover region. It's easier to accomplish with higher order filters.
    ^This is exactly what I am saying and I'm saying it for people who would like to understand it more than we do. Of course you can flip the polarity on the
    tweeter and get the phase CLOSE to the same at the corner frequency (2nd compared to a 4th order) However impulse response will suffer (some agree more than others)

    The main point I am expressing is, regardless of acoustical roll off, the phase will be changed between 2 drivers as you change the order of the electrical
    filter at the crossover frequency (one or both slopes) and as the post originally states we can use this to our advantage when we are dealing with offsets.

    Leave a comment:


  • bcodemz
    replied
    Originally posted by donradick View Post


    Mr XMAX - these are basically the nicest people you will ever meet on the internet.
    +1.

    Originally posted by Kornbread View Post
    If I was wanting to learn about designing/building my own crossovers, what sites//books would be most helpful?
    This is a tough question. I don't feel that the commonly suggested route of reading the standard Speaker Builder 201 or Loudspeaker Design Cookbook is a good suggestion. Personally I found them to be excellent theoretical resources, but they're lacking in practical applications, and a very poor choice for beginners. For example, I don't think either book explained very basic concepts what acoustical vs electrical slopes are, or the fact that a 3 way speaker needs a midrange chamber.

    Unfortunately there is no singular source I've found that will teach the basics to the advanced concepts in a clear manner. I feel that the best source is to ask someone to be your mentor. There is a lot of snippets of information scattered online, and it is difficult and time consuming to piece them together and properly understand them. This is why I feel there is so much misinformation for audio. Everyone is trying to piece scattering information together to make sense of them, but a lot of people piece them together wrongly, and end up misunderstanding the concepts. If not corrected early on, it will linger in the mind and become very difficult to change. Case in point, there are a lot of "great" debates in speaker design that does not need a debate at all as they have a very clear answer if one truly understands the concepts. All the debate is with people who don't actually understand the theory, or did things improperly, and reached the wrong practical conclusion and then blamed it on the concept.

    Leave a comment:


  • isaeagle4031
    replied
    Originally posted by Kornbread View Post
    If I was wanting to learn about designing/building my own crossovers, what sites//books would be most helpful?
    Well, right here of course. For books, Speaker Builder 201 and Loudspeaker Design Cookbook vol7. For measurements, Mr Appolitos book,

    Leave a comment:


  • Kornbread
    replied
    If I was wanting to learn about designing/building my own crossovers, what sites//books would be most helpful?

    Leave a comment:


  • Jeff B.
    replied
    Of course it matters what filter order you're using to combine two (or more) drivers. That's why when you have a system using LR2 slopes (acoustic slopes, not electrical) you reverse the polarity of one of the drivers in order to get proper summation at the crossover point. Similarly, when you have LR4 slopes (acoustic slopes, not electrical) you connect both drivers in the same polarity due to the extra phase shift.

    But when there is phase shift due to time of flight (acoustic offset problems) you can use slightly different transfer functions in the hope of making up some of that extra phase shift over a narrow range of frequencies in order to get better summation in the design axis at the crossover region. It's easier to accomplish with higher order filters.
    I was just never convinced that he agreed that a second order filter, when used with a tweeter with a second order roll-off could produce a textbook 4th order Linkwitz-Riley response in both amplitude and phase. And this point is important to going any further with crossover design.

    Leave a comment:


  • Pete Schumacher
    replied
    Of course it matters what filter order you're using to combine two (or more) drivers. That's why when you have a system using LR2 slopes (acoustic slopes, not electrical) you reverse the polarity of one of the drivers in order to get proper summation at the crossover point. Similarly, when you have LR4 slopes (acoustic slopes, not electrical) you connect both drivers in the same polarity due to the extra phase shift.

    But when there is phase shift due to time of flight (acoustic offset problems) you can use slightly different transfer functions in the hope of making up some of that extra phase shift over a narrow range of frequencies in order to get better summation in the design axis at the crossover region. It's easier to accomplish with higher order filters.

    Leave a comment:


  • dkalsi
    replied
    Originally posted by donradick View Post


    Mr XMAX - these are basically the nicest people you will ever meet on the internet.

    +1.

    I personally cannot thank Pete, Jeff, Dlr (and others) for sharing their knowledge with the DIY community. I'm always humbled when they take the time to answer even the most elementary questions.

    In my short experience here, I noticed the members that are the most knowledgeable about the subject keep their calm and are ALWAYS cordial/respectful when challenging other members' posted comments.

    I really respect them for that. I'm sure deep down they must be fuming when they read inaccurate information being posted :-), but they always seem to do a good jobs keeping calm and will often simply ask the opposing side to elaborate on their claim.




    Leave a comment:


  • donradick
    replied
    Originally posted by xmax View Post
    I'm only speaking of the phase (relationship between 2 or more drivers) and how it is affected by the
    electrical filter slope. I don't mean to insult any of you but please stop being A holes.

    Mr XMAX - these are basically the nicest people you will ever meet on the internet.

    Leave a comment:


  • donradick
    replied
    Originally posted by xmax View Post
    I'm only speaking of the phase (relationship between 2 or more drivers) and how it is affected by the
    electrical filter slope. I don't mean to insult any of you but please stop being A holes.

    Mr XMAX - these are basically the nicest people you will ever meet on the internet.

    Leave a comment:


  • augerpro
    replied
    Originally posted by xmax View Post
    2nd and 4th order filters will have dramatically different phase characteristics regardless of acoustical roll off.

    The difference between those two filters (180 degrees) will be directly applied to the driver's acoustic response, so I'm not sure what you mean by "regardless"? It almost sounds like you are saying "all things equal" which of course the final response can't possibly be equal going from 2nd to 4th order filters.

    Leave a comment:


  • augerpro
    replied
    Originally posted by xmax View Post
    I'm only speaking of the phase (relationship between 2 or more drivers) and how it is affected by the
    electrical filter slope. I don't mean to insult any of you but please stop being A holes.

    Focusing on only the electrical filter's contribution to the acoustic output of the driver is only half the story, but ok, let's look at that. What is your position exactly?

    Leave a comment:


  • dlr
    replied
    Originally posted by xmax View Post
    I'm only speaking of the phase (relationship between 2 or more drivers) and how it is affected by the
    electrical filter slope. I don't mean to insult any of you but please stop being A holes.
    Now there's some projection.

    dlr

    Leave a comment:


  • xmax
    replied
    I'm only speaking of the phase (relationship between 2 or more drivers) and how it is affected by the
    electrical filter slope. I don't mean to insult any of you but please stop being A holes.

    Leave a comment:


  • Mark65
    replied
    Originally posted by xmax View Post
    Hopefully that is simple enough, approximately 180 degrees is about as different as it gets, last I checked.
    I suppose that when discussing xo topologies at a given impedance, what you're saying makes sense. The problem (as Dave, Pete, and Jeff have pointed out) is that nearly every driver's inherent transfer function will muck up that theoretical response. And IME, relaxing the slope would be a situation where the overall acoustic slope would start at, say, LR2 at the knee, then resolve to the originally desired LR4.

    Leave a comment:

Working...
X