1. This site uses cookies. By continuing to use this site, you are agreeing to our use of cookies. Learn More.

Crossover question

Discussion in 'Speakers' started by Olorin, Apr 10, 2013.

  1. Olorin

    Olorin Member

    Joined:
    Sep 8, 2010
    Messages:
    131
    Trophy Points:
    16
    Ratings:
    +0
    Hi,

    Just a question for those more knowledgeable. If I select a crossover in the receiver (active) would it not cascade with the speakers passive crossover elements?

    I heard someone say that to me earlier in the week which I don't understand fully. I told him that using the low-pass in the receiver and in the subwoofer would result in filter cascading and he said that passive crossovers would cause filter cascading with any active crossover selected.

    If you guys could please help me understand this I would appreciate it.
  2. PSM1

    PSM1 Well-Known Member

    Joined:
    Dec 20, 2008
    Messages:
    18,240
    Trophy Points:
    171
    Ratings:
    +2,906
    You may have to ask the person who said it to explain as it does not make a lot of sense to me.
    The passive crossover in the speakers just makes sure the right frequencies go to the right drivers in that speakers. Setting the crossover in the amp reduces the frequencies below this being sent to the speakers. Can not see how this affects the passive crossover in the speaker and would just mean the bass driver does less work.
    If you set the crossover in the amp and the sub then you could lose frequencies if you set the one on the sub below the one in the receiver (assuming the crossover on the sub affects the phono input). So this could cause an issue but as long as it is above the amp setting then again there will be no issue. Not sure what is meant by cascading though.
  3. BlueWizard

    BlueWizard Well-Known Member

    Joined:
    Jun 19, 2007
    Messages:
    14,930
    Trophy Points:
    166
    Ratings:
    +2,680
    Likely the Amp is not a brick wall. That is, assuming the crossover is set to, say, 80hz, the sound does not go down to 80hz and stop. Rather it goes to 80hz then below that it starts to fade very quickly. Likely at a rate of about -12dB per Octave or perhaps as much as -24dB per Octave.

    Out of the Subwoofer and out of the Amp relative, we have Low-Pass rather than the High-Pass that is being sent to the other speakers from the amp. In the Sub Low-Pass section there is a high roll-off. That roll-off is likely either -6db (which is slow) or -12dB/octave (more typical).

    Now the crossovers in the Subwoofer itself are very likely Active Crossovers as well, assuming we are talking about an Active Subwoofer (Sub with built-in amps). So, it likely has roll off slopes of either -6db/octave or -12dB/Octave.

    These crossovers and their accompanying Roll-Off don't really care what signal is coming into them, they are simply going to attenuate by the standard Roll-Off fixed into them. Let's say -12dB, and let's assume similar for the Amp Roll-Off.

    So, when we are past the crossover frequency and into the Roll-Off area, the amp is going to fade at a rate of -12dB. If the Subwoofer is set to the same crossover frequency, it is going to roll-off (fade) the signal at the same -12dB rate regardless of the level of the incoming signal.

    So -12dB from the Amp PLUS -12dB from the Sub EQUALS -24dB rate of roll-off or fading of the signal.

    (-12dB) + (-12dB) = -24db

    If this is what you mean by cascading then YES, the filters will cascade.

    Generally you set the Crossover of the Sub as high as it will go, so it really has no effect. Then you allow the Amp to control the crossover and accompanying Roll-Off exclusively.

    But you choose one or the other, not both, and by far the most common is to get the Sub controls out of the circuit and let the Amp handle everything, assume this is a standard AV Reciever.

    In Stereo Amp, there are no internal crossover controls, in this case, the Subwoofer control exclusively control what is happening to the bass in the Sub.

    Does that help?

    Steve/bluewizard
    Last edited: Apr 10, 2013
  4. BlueWizard

    BlueWizard Well-Known Member

    Joined:
    Jun 19, 2007
    Messages:
    14,930
    Trophy Points:
    166
    Ratings:
    +2,680
    I think I have a graphic on file to give you an idea of what I mean by Roll-Off Slopes -

    [​IMG]

    The section of the graph that is at an angle, about a 45° angle is the area where the frequencies are 'Rolling Off' or Fading.

    In standard internal speaker crossovers the slopes or rate of Roll-Off is controlled by how many crossover components you use.

    If you use a single crossover component, the rate of roll off is -6dB per Octave.

    Add a second component and the rate become -12dB, that is (-6) + (-6) = -12dB.

    Add a third crossover component, and that Slope or rate of Roll-Off become-18dB (-6) + (-6) + (-6) = -18dB.

    Add a forth crossover component. and the Slope is 4 x (-6) = -24dB.

    In the case of the Sub and the Amp working together, the electricity doesn't care where the components are, if they are in the same box or in different boxes, the Crossover Slopes add together.

    Though 'cascade' would imply that they multiply, not so, the merely add.

    ...in my opinion.

    Steve/bluewizard
  5. Olorin

    Olorin Member

    Joined:
    Sep 8, 2010
    Messages:
    131
    Trophy Points:
    16
    Ratings:
    +0
    Correct me if I'm wrong but the passive crossovers for the drivers in a speaker are well outside the operating bandwidth of the amplifier crossovers? That's really what my friend is getting at I think. He thinks that if you use a crossover in the processor that it would interact with the crossover in the speaker.

    To my knowledge this won't happen.
  6. PSM1

    PSM1 Well-Known Member

    Joined:
    Dec 20, 2008
    Messages:
    18,240
    Trophy Points:
    171
    Ratings:
    +2,906
    The crossover in the processor will have no impact on the crossover in the speakers so no need to worry there. All you will be doing is taking some of the heavy load of the bass drivers and hopefully clearing up the mids.
  7. BlueWizard

    BlueWizard Well-Known Member

    Joined:
    Jun 19, 2007
    Messages:
    14,930
    Trophy Points:
    166
    Ratings:
    +2,680
    They are all in the Audio range, so I'm not sure what you mean.

    The Slopes will shift, but the cross over points won't. By that I mean, no crossover shift. If you combine an 80hz crossover with an 80hz crossover, it is not shift up to 160hz or down to 40hz. It stays at 80hz, but the two combined crossovers causes a faster roll-off, or described differently, a steeper slope on the roll-off.

    I was simply illustrating how ganging more crossover components doesn't change the crossover point, but does effect the Slopes. The parallel to the current discussion is, you are ganging the crossover in the amp with the crossover in the Sub, causing the crossover point to stay the same but the slope to increase.

    I confess, I'm not really sure what you are asking or what you mean by "cascading". I did my best to explain it in the context that I understood it.

    As you can see from the graphic I provided, the Sub crossover is about 80hz (yes, I know 200hz in the diagram, but the diagram was created for another post). That 80hz (or 200hz) crossover has no effect on the crossovers at 800hz and 3000hz.

    This is true in the example you gave, because the Amp has internal active crossovers, they are in the Pre-Amp section completely isolated from the passive components in the speakers. So electrically, there is no interaction between the Amp crossover and the speaker crossovers. But, depending on which outputs you are talking about and which speakers you are talking about, it is possible for the roll-off slope of one to be added to the roll-off slope of the other.

    Again, the question isn't clear enough to really know how to answer, so the best I can do is illustrate the general nature of interacting crossovers.

    If the amp had passive crossovers and the Speakers or Sub had passive crossovers, there might be some situation in which those component could interact, but that is an extremely unlikely circumstance. If there is anything that isolates the Amp's passive components from the Speakers passive components then the interaction is all greatly diminished. But this is all hypothetical because no such system exists.

    Did the best I could with what I had.

    Steve/bluewizard
    Last edited: Apr 10, 2013
  8. Olorin

    Olorin Member

    Joined:
    Sep 8, 2010
    Messages:
    131
    Trophy Points:
    16
    Ratings:
    +0
    First, thank you for taking the time to type all of this out. I appreciate it. Can you imagine a scenario where a given speaker's passive crossover could interact with a receiver crossover in same way?

    Aren't woofers in a small 2-way crossed over to like 400 Hz or above? A small satellite speaker might be able to only handle bass down to 100 Hz, but the crossover in the speaker between mid/range and bass might be 250 Hz, so that 100 Hz crossover slope (4th order) in the receiver isn't really interacting as I can tell.

    But I'm not the guru. Please tell me what you think. I know the question wasn't very clear but I imagine he was referring to the crossover in the processor affecting the speaker crossover. That's about it, but he worded it poorly.

    Thanks again for your help.
    Last edited: Apr 10, 2013
  9. BlueWizard

    BlueWizard Well-Known Member

    Joined:
    Jun 19, 2007
    Messages:
    14,930
    Trophy Points:
    166
    Ratings:
    +2,680
    Well, I confused the discussion because I assume we were talking about the Amp internal Active Crossovers interacting with the SUB's internal Active crossovers.

    On second reading, I see you were talking about "speakers passive crossover elements".

    So, I tried to fudge and blend the two points together.

    But because the amps Crossover components are Active and are Part of the Pre-Amp section of the amp, they are completely isolated from any connected passive speaker. As a result of this, there can be no interaction between them.

    But it does depend on what amp you have, and which speaker we are talking about. But in general, because the Amp's crossover controls are Active and in the Pre-Amp section, there can be no interaction between them and any passive speaker relative to the crossovers.

    Is that a more concise an on-point answer?

    I don't rescind anything I said previously, I just hope this new post in more to the point.

    Steve/bluewizard
  10. Olorin

    Olorin Member

    Joined:
    Sep 8, 2010
    Messages:
    131
    Trophy Points:
    16
    Ratings:
    +0
    Thanks BlueWizard, but I need your help once more. :) I spoke to my friend about this but he basically told me (and I quote):

    "the passive crossovers in the main speakers do not magically disappear the typical passive (having designed and built some myself even) will have a lowpass and a highpass right there on the pcb.. many home theatre receiver have highpass of 12db/oct and lowpass of 24db/oct because the assumption is that most hifi passives are 12db/oct, thus creating a sum total of 24db/oct, thus keeping the phase in-line .."

    What is he saying? I'm not a designer, but I don't have the foggiest clue.
  11. BlueWizard

    BlueWizard Well-Known Member

    Joined:
    Jun 19, 2007
    Messages:
    14,930
    Trophy Points:
    166
    Ratings:
    +2,680
    If you look at my diagram above, you can see a 3-way speaker with crossovers at 800hz and 3,000hz. These are both High Pass and Low Pass together. The Woofer is a Low Pass blocking (rolling off) everything above 800hz. The Midrange is a Band Pass, which is comprised of a High Pass allowing everything above 800zh PLUS a Low Pass blocking everything above 3000hz. Between the two they make a Band Pass covering everything between 800hz and 3000hz. Finally, the tweeter section is a High Pass allowing everything above 3,000hz.

    So -

    Woofer = Low Pass (everything below 800hz)

    Midrange = High Pass (everything above 800hz) AND a Low Pass (everything below 3000hz)

    Tweeter = High Pass (everything above 3000hz)

    Now, the only crossover in an AV Receiver is the Subwoofer Crossover, and that is typically 80hz. How does that ELECTRONIC Crossover at 80hz in any way effect the other speaker passive crossovers.

    AV Receivers can be set to one of two settings - LARGE and SMALL.

    Large = Full Range Front Speakers (in summary)

    Small = Limited range Front Speakers, all the bass below the Subwoofer crossover is diverted to the Sub, no low bass to the front speaker.

    In the case of SMALL, you have two cross overs, a LOW PASS for the Low Bass, which is diverted to the Sub. And a High Pass, which sends everything above the Subwoofer Crossover (80hz) to the Front Speakers. So, if we assume an 80hz Sub Crossover, everything above 80hz goes to the main Front Speakers.

    Now, again look at the Diagram I provided. These crossovers are not brick walls, they are slopes. Speaker response doesn't end, it fades out.

    But again both these AV Receiver Crossovers occur inside the amp. The speaker has no awareness that they are there. The Amp merely sends 80hz to 20khz to the front speakers, and the speaker reproduce it.

    Again, because the AV Receiver Crossovers are Active Electronic Crossovers, and because they occur in the Pre-Amp section of the AV Receiver, there can be no interaction between these Active Crossovers, and the internal Passive Crossovers of the Speakers. Each has no awareness that the other is there.

    For the most part, to the best I can interpret it, what your friend said is gibberish intermixed with a few twisted bit of truth.

    Once again, the internal Crossovers in the AV Receiver, have no effect on the Passive Crossovers in the Speakers. Once the Amp crossovers are set, the Amp simply sends sound in the range of 80hz to 20khz to the front speakers. That Passive crossovers in the speakers simply react to that limited range of sound. There is NO interaction between the Amp and the Speaker crossovers. There are a whole lot of electronics in the Amp isolating those crossover settings from the actual speakers. So - No interaction.

    Notice in the Diagram I proved, the Subwoofer is on a completely separate line from the Main Speakers, and that is because they are truly isolated from each other. Two separate outputs drive them. The Sub Pre-Amp Out drives the Subwoofer, and speaker terminals drive the Front Speakers. Nothing connects the two, except the electronics deep inside the amp. Again, there is no way they can interact.

    You friend is free to believe anything he wants. He can believe the earth is flat. He can believe the Universe was literally created in 6 days. He can believe that Evolution doesn't happen. He can believe anything at all, but he has to realize that just because he believes it, doesn't make it true.

    Now, perhaps there is an aspect of what he is trying to say or in what you are trying to say that I don't understand. But I doubt it. I've got a Bachelor's degree in Electronics Engineering Technology from the Physics Department. What are your friends qualifications?

    Sorry.

    Steve/bluewizard
    Last edited: Apr 12, 2013
  12. BlueWizard

    BlueWizard Well-Known Member

    Joined:
    Jun 19, 2007
    Messages:
    14,930
    Trophy Points:
    166
    Ratings:
    +2,680
    Here is an incredibly crude Diagram of an Source, Amp, and Speakers -

    [Input]--=--[Pre-Amp]----[Active Crossover]----[Power Amp]---=---[Passive Crossover]----[Speakers]

    When you look back into the amp from the Passive Crossover, what do the Passive Crossovers see? They see the Power Amp, they have no awareness of the Active Crossovers. The Power Amp isolates the Passive Crossover from the Active Crossovers by acting as a Buffer between them. They are ISOLATE.

    There can be no adding or subtracting of the crossover slopes, because the speakers have not crossover in the 80hz range. To add or subtract, they both have to cover the same range, and they do not.

    Steve/bluewizard
  13. Olorin

    Olorin Member

    Joined:
    Sep 8, 2010
    Messages:
    131
    Trophy Points:
    16
    Ratings:
    +0
    Thanks very much for the explanation. I've sent you a PM. :)
  14. BlueWizard

    BlueWizard Well-Known Member

    Joined:
    Jun 19, 2007
    Messages:
    14,930
    Trophy Points:
    166
    Ratings:
    +2,680
    In the PM, is a link to a discussion in South Africa in which one member is arguing the "Cascading Crossover" point, the rest of the forum is taking my position against that member.

    The simplest thing is to use the crude text diagram I provided, (metaphorically) stand at the Passive Crossover and look back at the source of the signal. What do you see? You see the POWER AMPS not the Active Crossovers.

    At the output of the Amp, if we assume an 80hz crossover, out comes a standard signal in the range of 80hz to 20khz. Like any signal, this signal will roll off at 80hz and below.

    How can that possible interact with the passive crossover, when the passive crossovers are essentially out of the circuit in the 80hz and below range. In my Frequency Response Diagram, the lowest passive crossover is at 800hz and it is a Low Pass.

    The Filters simply can't interact because they simply do not cover the same range. There's no Passive Crossover anywhere near 80hz.

    So, for the filters to interact and produce any type of 'cascading' effect, both filters would have to be working on the same frequency range - they are not, and they do not.

    Steve/bluewizard
    Last edited: Apr 12, 2013
  15. Olorin

    Olorin Member

    Joined:
    Sep 8, 2010
    Messages:
    131
    Trophy Points:
    16
    Ratings:
    +0
    Thanks again Steve! Your explanations have been very detailed and informative and I appreciate the time and effort you put into each one. Again, thank you for your efforts in explaining this.
  16. Olorin

    Olorin Member

    Joined:
    Sep 8, 2010
    Messages:
    131
    Trophy Points:
    16
    Ratings:
    +0
    Steve, please confirm something for me. This is just an example I can play in my head.

    If you take these speakers :

    EMP Tek E55Ti Tower Speakers

    It looks like the woofers are low-passed at 120 Hz. So in this case, with these speakers, if you used a 100 Hz-120 Hz crossover on the AVR it would interact and conflict with this speaker?

    Please correct me? I assumed speakers crossovers for the woofer would not be so low.
  17. PSM1

    PSM1 Well-Known Member

    Joined:
    Dec 20, 2008
    Messages:
    18,240
    Trophy Points:
    171
    Ratings:
    +2,906
    Since those speakers go down to 40Hz you would not have the amp crossover as high as 100-120Hz but probably in the 60 to 80Hz range hence there would be no conflict. Also what do you expect with this so called conflict. Even if they are close together all that it would mean is a slightly faster decay rate which is not going to cause a massive issue.
  18. Olorin

    Olorin Member

    Joined:
    Sep 8, 2010
    Messages:
    131
    Trophy Points:
    16
    Ratings:
    +0
    Well I guess what I'm saying is that in that example, if I chose a 120 Hz crossover the slopes should combine in some way, surely? So it would steepen quite significantly. I imagine that would have some effect on things, but I'm not the expert.
  19. BlueWizard

    BlueWizard Well-Known Member

    Joined:
    Jun 19, 2007
    Messages:
    14,930
    Trophy Points:
    166
    Ratings:
    +2,680
    No, the Wharfedale Diamond 9.1 3.5-way have a low-bass crossover of 150hz. But that has nothing to do with the Subwoofer or the AV amps internal crossovers. As I said, they do not effect the same range.

    Likely, though I'm guessing, the EMP is a 3-way with 3x 6.5" Low-Bass drivers, 2x 5" Mid-Bass Drivers, and a Tweeter.

    So best guess -

    3x 6.5" Low-Bass = 30hz to 120hz (3x meaning THREE drivers)

    2x 5" Mid-Bass = 120hz to 3000hz

    1x Tweeter = 3000hz to 20khz

    If you set the Sub crossover at 120hz, and set the system to SMALL, the Low-Bass drivers will do virtually nothing below 120hz. Any sound will be weak and fade quickly. But it has nothing to do with the Active Crossover in the Amp. Beyond sending a limited signal to the front speaker, there will be no interaction in the crossovers. At least no significant interaction.

    Now, in the Mid-Bass speakers, because they are fading out below 120hz, the fade out of the Amp below 120hz could add to the fade out of the Mid-Bass speakers below 120hz.

    However, as has been pointed out, you would never run this speaker with the amp set to 120hz or higher. It would be a waste of the front speaker. While there could technically be some interaction, it would only be because you made it willfully so. It would never happen in any realistic situation. If would be positively Dumb to run the speaker as described in this example.

    So, while you have create a situation in which the fading produced by the amp is added to the fading out produced by the speakers, it is a hopelessly unrealistic situation. Further, even if it is true, because you are not really in the working range of the Drivers, it would be debatable whether it would make a audible difference to the typical listener. One assumes this would be used with a Subwoofer, and the sound from the Subwoofer would overwhelm the fading sound of the Front Speakers below 120hz.

    This is a rare speaker, so to truly come across this possibility, and then to foolishly configure the system in the way described simply would not happen.

    Steve/bluewizard
    Last edited: Apr 13, 2013
  20. Julf

    Julf Member

    Joined:
    Apr 20, 2013
    Messages:
    1
    Trophy Points:
    1
    Ratings:
    +0
    The same discussion came up on Hydrogen Audio, and I made this scilab plot of how the effects of two low pass filters (one in the amp/receiver and one in the speaker) add up. I used the parameters suggested by the OP on HA, 80 Hz for the filter in the AVR and 500 Hz for the filter in the speaker (for the woofer).

    [​IMG]

Share This Page