Hello diya community,
I need help with my first crossover.
I have questions about some correction elements and how I have to optimise them and what I have to pay attention to.
For example, I don't know exactly what to look for in the phase and whether I'm using reasonably suitable component values when setting the crossover frequencies.
The following is the crossover without any correction:
I have chosen a 2nd order Linkwitz-Riley circuit for the filters.
Circuit:
What would be the next step? Should I take care of the impedances?
If so, where is the best place to start? At the tweeter, midrange, or woofer?
Should I pay attention to the phase response all the time?
What is important to keep in mind about the phase curve?
Should the phase jumps of the drivers be close to each other, like with the woofer and the midrange? Should the phases run into each other?
How bad are peaks in the impedance?
When I put in the corrections where I think they are right, it looks like this::
However, I don't know if I have the right approach for the correctors and if I have the optimal component sizes.
I would be glad if someone could explain his procedure to me. My approach to the correction elements is more trial and error.
However, I would like to do it better.
Any help and suggestions are very welcome.
I need help with my first crossover.
I have questions about some correction elements and how I have to optimise them and what I have to pay attention to.
For example, I don't know exactly what to look for in the phase and whether I'm using reasonably suitable component values when setting the crossover frequencies.
The following is the crossover without any correction:
I have chosen a 2nd order Linkwitz-Riley circuit for the filters.
Circuit:
What would be the next step? Should I take care of the impedances?
If so, where is the best place to start? At the tweeter, midrange, or woofer?
Should I pay attention to the phase response all the time?
What is important to keep in mind about the phase curve?
Should the phase jumps of the drivers be close to each other, like with the woofer and the midrange? Should the phases run into each other?
How bad are peaks in the impedance?
When I put in the corrections where I think they are right, it looks like this::
However, I don't know if I have the right approach for the correctors and if I have the optimal component sizes.
I would be glad if someone could explain his procedure to me. My approach to the correction elements is more trial and error.
However, I would like to do it better.
Any help and suggestions are very welcome.
For now, just keep them nearby within the range of the overlap.
For now, don't worry about them.
Oh, sorry,
I forgot to mention that these are not real measurements but I used SPL Trace before SPL and Impendance, so they are the measurements from ScanSpeak data sheets.
I selected the following treiers; D2908/714000 Revelator, 12MU/8731T00 Illuminator and 22W/8851T00 Revelator.
I hope this is still suitable as an exercise before I take measurements myself.
I forgot to mention that these are not real measurements but I used SPL Trace before SPL and Impendance, so they are the measurements from ScanSpeak data sheets.
I selected the following treiers; D2908/714000 Revelator, 12MU/8731T00 Illuminator and 22W/8851T00 Revelator.
I hope this is still suitable as an exercise before I take measurements myself.
So where the phase shift comes they should be close to each other? or where they meet (circle)?
If i understood correctly you mean the circle at 0 degree.
Last edited:
Are the impedance peaks only important for a tube amp? So it doesn't matter if I have a 40 Ohm peak or more when I use a transistor amplifier?
I just have to make sure that I don't go below 4 ohms, otherwise my amp will burn out, no matter if it's a tube or a transistor.
Would it have a negative effect on the sound?
I changed the crossover frequence of the woofer and Mid
Do you mean the rotation (where it goes from bottom to top)?
No. You want phase to be near where the response is crossing.
Ok, I want the phase close to the crossover frequency.
Do you mean the point where the phase is 0 degrees and falls off at an angle (so the 0 degree point I made a circle around should be closer to the crossover frequency).
Or where the rotation is where it goes from bottom to top?
Sorry, I just don't want to misunderstand you.
It depends.
Impedance peaks of drivers can mess with the crossover. Therefore I think it's good you have raised woofer-midrange crossover point. That way the woofer impedance peak does not influence the crossover that much.
Any amp with low output impedance (usually transistor amps with negative feedback) is fine with impedance variations due to drivers or crossover as long as it does not get too low.
But of course you can still try to smoothen the impedance curve as an exercise - but before finalizing your project check availability and prices for a 37 mH/196 mOhm coil.
I happened to look up the price of some of the components just before your post. Well, i guess one would probably have to get closer to some values and see if you can achieve a similar result.
Where I'm not quite sure is the difference between a sable element (C-R) and a serial notch filter (R-L-C).
Is the sable element more for an impedance that rises and a serial notch filter for strong peaks?
And is it normal that the impedance rises so much at the crossover frequencies or is this a sign that the lowpass and highpass are too far apart. In my case, the peaks are 15 and 13 ohms. Or are these normal values?
Where I'm not quite sure is the difference between a sable element (C-R) and a serial notch filter (R-L-C).
Is the sable element more for an impedance that rises and a serial notch filter for strong peaks?
And is it normal that the impedance rises so much at the crossover frequencies or is this a sign that the lowpass and highpass are too far apart. In my case, the peaks are 15 and 13 ohms. Or are these normal values?
i don't understand "sable element". (ok just looked it up: you probably mean "zobel" 😊 - it's no animal but mr. zobel from bell labs in this case... )
caps, resistors and coils can be arranged in several ways and will act very differently. it's no simple matter!
yes they are. impedance nearly always rises at crossover points, unless you linearize it.
just look at some examples.
i suppose you speak german, so here is an example (scroll to the end) of a speaker with/without linearized crossover impedance peak:
https://www.donhighend.de/?page_id=8971