Can I ask why you are curious?
Unless you have a very noisy active setup (poor quality DSP) - you can achieve excellent sonic results with either active or passive approaches.
If you are unhappy with how it sounds or wondering if it could be improved - then measuring is your best bet.
If you have just plugged in the order of crossover slopes and points (not even Z offset) then your acoustic frequency response could vary quite a bit. I have found issues in the 2Kh - 5Khz the most problematic when it comes to listening fatigue. Have you got the ability to measure frequency response?
So... if you want to improve what you have got - the first thing to do (active or passive crossover solution) is get a measurement microphone. If you still want passive - then you need to measure impedance as well (although I've had great success simply using manufacturer impedance measurements).
If you just want to copy what you have got, then a crossover simulation tool like VituixCAD will show you the electrical transfer function of your passive crossover (what your DSP shows) and you can design a passive crossover to mimic that. we largely ignore the transfer function in passive crossover design as it is the crossover combined with in-box on baffle driver responses (acoustic response) that matters.
I'd recommend you just import manufacturer curves and put a text book / on-line calculator crossover into VituixCAD to see what I mean.
Unless you have a very noisy active setup (poor quality DSP) - you can achieve excellent sonic results with either active or passive approaches.
If you are unhappy with how it sounds or wondering if it could be improved - then measuring is your best bet.
If you have just plugged in the order of crossover slopes and points (not even Z offset) then your acoustic frequency response could vary quite a bit. I have found issues in the 2Kh - 5Khz the most problematic when it comes to listening fatigue. Have you got the ability to measure frequency response?
So... if you want to improve what you have got - the first thing to do (active or passive crossover solution) is get a measurement microphone. If you still want passive - then you need to measure impedance as well (although I've had great success simply using manufacturer impedance measurements).
If you just want to copy what you have got, then a crossover simulation tool like VituixCAD will show you the electrical transfer function of your passive crossover (what your DSP shows) and you can design a passive crossover to mimic that. we largely ignore the transfer function in passive crossover design as it is the crossover combined with in-box on baffle driver responses (acoustic response) that matters.
I'd recommend you just import manufacturer curves and put a text book / on-line calculator crossover into VituixCAD to see what I mean.
Last edited:
Do I understand this correctly?: the passive crossover should sit behind the single amplifier, which then drives the whole speaker? So it should then be a loudspeaker with a classic passive crossover?
If so, I would first measure the frequency responses of each driver in the active setup.
Then I would disconnect all the active electronics and measure the impedance and frequency responses of each driver installed in the intended cabinet and import these measurements into a suitable simulation program (VituixCAD in my case) to design the passive crossover there. The measurements of the individual drivers of the active setup can then be displayed as an overlay.
One does not know what the DSP does apart from the separation between the drivers. Possibly there is still the use of a few PEQs. It becomes especially difficult if there are also delay time adjustments, because these cannot be reproduced passively.
Many greetings,
Azrael
edit:
Sorry, I didn't see that the DSP-settings were already postet.....
Translated with the help from www.DeepL.com/Translator (free version)
Last edited:
We could also simulate the in cabinet impedance curves using Jeff Bagby's Response Modeler if the OP would supply the driver names, the volume of each enclosure and the diameter and length of any ports. This assumes the driver sheets with impedance curves are available.
The OP has used textbook filters without any measurements, but likes the sound. And @AllenB has a method to duplicate the same effect using passive crossovers. It's certainly not how I would do it (no measurements), but que será, será.
I have no issue measuring, I have UMIK and DATS3 for a reason! I was just curious the best way to go about it. I have the speakers in a test cabinet right now and working on the cut list for the real one. I can take measurements when I get into the real cabinet.
The idea of it in cardboard is interesting however. I do have bunch laying around, but my design for the cabinets at this point is pretty good and I don't see changing much in terms of the baffle. This is going inside a larger cabinet so I would assume I would rather have a large baffle and try to compensate for it rather than have a cave behind the speakers creating some kind of echo.
The idea of it in cardboard is interesting however. I do have bunch laying around, but my design for the cabinets at this point is pretty good and I don't see changing much in terms of the baffle. This is going inside a larger cabinet so I would assume I would rather have a large baffle and try to compensate for it rather than have a cave behind the speakers creating some kind of echo.
Keep your microphone for designing a new crossover. Avoid using it for duplicating an existing crossover.
I suppose the confusion for me is your original post where you wanted to 'use the info from. These' being your active setup to inform a passive design.
I had assumed you wanted to implement the same transfer functions.
Since you can measure, pretend your active doesn't exist abd start from scratch. Use or measure and load those files into xsim or vituixcad and start modeling. I'm not aware of any modeling tool that given a transfer function, will generate a passive crossover for you
I had assumed you wanted to implement the same transfer functions.
Since you can measure, pretend your active doesn't exist abd start from scratch. Use or measure and load those files into xsim or vituixcad and start modeling. I'm not aware of any modeling tool that given a transfer function, will generate a passive crossover for you
I know how big a jump it can be before you start getting better results than you can by listening, and while there's nothing wrong with that it seemed off-topic.
However converting active to passive is a skill on its own. If electronic devices happen to fail then at the drop of a hat you could end up with a piece of junk.
However converting active to passive is a skill on its own. If electronic devices happen to fail then at the drop of a hat you could end up with a piece of junk.
Suppose I am using VituixCAD. So I load only the impedance curves of the involved, drivers and as overlay for the diagram "Filter" the filter curves from the active setup? Then I build the passive crossover so that in "Filter" the filter curves of the passive setup are congruent with that of the active setup?
If I think about it, this might even work, provided that all aspects of the active control of the drivers are completely included in the respective filter curves. For example, it could be that the volume ratios of the drivers to each other are controlled by the gain controls of the power amp channels involved. This would then not be reflected in the filter curve.
It also becomes difficult if delays have been used for runtime adjustment. But then you can't recreate this passively anyway, it's not a problem of the method we're talking about here.
Btw:
What is meant here by "response"?
Many greetings,
Azrael
Translated with the helpp from www.DeepL.com/Translator (free version)
Yes I think you have it right it's the filters that are being copied. So when I said response I meant level vs frequency measured at the driver terminals.
If you measure then the delays are included. If you don't measure then you can add delays in the simulator.
Delays can be done at speaker level.
One question i have, where in this approach is taken into account that the output of an amplifier(in case of active filtering) is different from that of a passive crossover?
Passive components introduce impedance varying with frequency, the amp output impedance is comparatively very low and independent of frequency.
The driver itself is not passive itself.
The driver itself is not passive itself.