Originally Posted by navin

"Naughty" is being really really naughty!

,,,,,,,,,, but also inclucde the vagrancies in the woofer/tweeters frequency response, baffle step (if there is a box involved), edge/difraction compensation, phase alignment, etc...

Obviously there are only so many problems a crossover cna practically correct for - after that the crossover starts to detract from the sound rather than make it better. Hence DIYers prefer to start with drivers that do not have too many issues to begin with. It makes crossover design simpler and also money can be spent on a few good components than many poor/average ones.

in short do not consider a speaker better or worse becuase of it's crossover's slope/topology but consider the entire design as a system and look at the entire design in it's totality.

i still stand by my observations because basically in a car there wouldnt be an enclosure to worry about for the higher frequency speakers so baffle stepping wouldnt be a problem since most of the time the drivers are also on a different plane and there are ways to compensate for edge diffraction ie to round off the edges of the baffle that the speaker is mounted on, as for phase alignment if the slope is constant it will matter not ie if one driver is 360 degrees out of phase because of a fourth order filter then all the drivers would be equally out of phase if they use a similar crossover hence effectively nothing is out of phase if you keep the slope constant but speaker positioning then becomes critical as does time delay

agreed that you need to start with drivers that have minimal issues in relation to each other but generally most people would consider the sleeper slope a more attractive proposition - thats why the makers of processors like the audison bit one and the JBL MS8 give you crossovers with adjustable slopes and topologies up to 48db/oct with a choice of either butterworth or linkwitz reilly configurations/topologies

but having said that the last statement you make is also the truth because its not only the slope of the crossover that matters when it comes to how good a speaker set would really be

generally i find car audio is a matter of compromises, and the idea is to use the compromises that you personally can live with and create the least disruption to good sound for your taste

[navin]

Originally Posted by naughty001

i still stand by my observations... i find car audio is a matter of compromises,

when i talk audio I include car audio as well has home audio with a little bit of pro audio thrown in.

yes the applications are all different but the basic principles are the same and like you said ALL audio is a matter of compromises. Car audio complicates things because the cabin is not nesscarily designed to be acoustically correct and the fact that making such a cabin (no road noise or traffic noise etc..) would be a great detrement to driving a car.

Originally Posted by navin

when i talk audio I include car audio as well has home audio with a little bit of pro audio thrown in.

yes the applications are all different but the basic principles are the same and like you said ALL audio is a matter of compromises. Car audio complicates things because the cabin is not nesscarily designed to be acoustically correct and the fact that making such a cabin (no road noise or traffic noise etc..) would be a great detrement to driving a car.

yes i agree that the basic underlying principles are generally the same

i have done a bit more reading as to the pro's and con's of steeper slopes and i have found the following further observations that we can throw into the mix

a steeper slope will increase group delay and as such constant slopes within a system become more important as long as you can compensate for speaker locations

but the more important point is that steeper slopes WILL REDUCE DISTORTION. This phenomenon occurs because you have harmful harmonics one octave away thus the steeper the slope the lower the harmonics at that point will sink toward the noise floor ie they become inaudible hence thus generally a steeper slope is more desirable

yet having said that we still need to take into account the group delay and phase characteristics that each slope linked to a particular topology will provide as well as what you actual crossover points are and where the actual intersection of crossover points occur for example a linkwitz reilly crossover would require the points to be set so that the intersection occurs at a -6db down point, and as Navin has meticulously pointed out above where the natural rolloff of each individual driver occurs. So we have to conclude that even though the steeper slope is more desirable it is not always possible which is why even crossover design or tuning (in the case of active crossovers) has to again be a balance of compromise and functionality

[navin]

Originally Posted by naughty001

So we have to conclude that even though the steeper slope is more desirable it is not always possible which is why even crossover design or tuning (in the case of active crossovers) has to again be a balance of compromise and functionality

I think we could repharse that to read as...

While a Steeper Slope is often desireable it may not nesscarily be the most "correct" slope to use for a particular set of drivers in a specific application.

[Mi10]

In a fully active setup, What is the role of the crossover in matching the sensitivity of different components?

To explain better, Say i have a pair of Mid bass whose sensitivity is at 88db and a pair of tweeters whose sensitivity is at 91db. Now if i am going active with this set of components, is it necessary is match the senstivity of these two ? (Either tone down the tweeter to 88 db or scale up the midbass to 91 db?)

Now how do i perform this?. Is it something that can be done via an electronic cross over ?

[Mi10]

Another Question:
I have crossover slopes from 6db to 24db i my HU. however I use a passive crossover. Which is say a second order filter (12db ?) so irrespective of what i keep from my HU my speakers can only take in what the crossover sends, am i right in my understanding Guru's ?

[navin]

Originally Posted by Mi10

In a fully active setup, What is the role of the crossover in matching the sensitivity of different components?

If you have an active crossover you will also have different amplifiers for the tweeter and woofer hence by turning down/up the gain of the amps you can adjust the relative level of the tweeter and woofer.

Originally Posted by Mi10

Another Question:
I have crossover slopes from 6db to 24db i my HU. however I use a passive crossover. Which is say a second order filter (12db ?) so irrespective of what i keep from my HU my speakers can only take in what the crossover sends, am i right in my understanding Guru's ?

The XO in your HU is an active XO. The object of this XO is to send your components ony freq above say 100Hz (or some other chosen subsonic frequency).

The frequencies above 100Hz then get processed into freq above 2kHz (or so) for the tweeter and between 100and 2 kHz for the woofer. Freq below 100 (just and example) are to be reproduced by a subwoofer.

[Mi10]

Originally Posted by navin

The XO in your HU is an active XO. The object of this XO is to send your components ony freq above say 100Hz (or some other chosen subsonic frequency).

The frequencies above 100Hz then get processed into freq above 2kHz (or so) for the tweeter and between 100and 2 kHz for the woofer. Freq below 100 (just and example) are to be reproduced by a subwoofer.

Thanks Navinji,
My Question is a bit more specific with respect to Slopes. Suppose i high pass from the HU at say 300Hz @ 24db/octave slope, This signal reaches the Crossover first before going to the Speakers. And suppose my passive crossover can only split frequencies at 12bd/octave slope will the end resultant signal be of a 12 db slope or 24 db slope or any other 'X' db slope ?

Sorry if i sound very naive (i am actually one )

[navin]

Originally Posted by Mi10

Thanks Navinji,
My Question is a bit more specific with respect to Slopes. Suppose i high pass from the HU at say 300Hz @ 24db/octave slope, This signal reaches the Crossover first before going to the Speakers. And suppose my passive crossover can only split frequencies at 12bd/octave slope will the end resultant signal be of a 12 db slope or 24 db slope or any other 'X' db slope ?

Sorry if i sound very naive (i am actually one )

If you cascade a 24db active XO with a 12db passive XO at exactly the same XO frequency you will get a 36db XO but there might be some strange phasing issues with such an arrangement.

[Mi10]

Suppose,
i buy a Tweeter from Brand 'X'
i buy a Mid-woofer from Brand 'Y'
i buy a Passive Crossover from Brand 'Z'

What are the criterias should i keep in mind before buying and assembling my 'custom Component' set.

>> Fs of the tweeter and Midbass
>> Crossover point the crossover (This should be less than or equal to my Fs of tweeter)
>> Power handling of the crossover.
>> Sensitivity of the Midbass and Tweeter

My Question is
Is the above setup possible?
If yes, are my considerations good enough / have i missed any other vital point
Can Bi-ampable crossover be used for such purposes

Thanks in advance

Originally Posted by Mi10

Suppose,
i buy a Tweeter from Brand 'X'
i buy a Mid-woofer from Brand 'Y'
i buy a Passive Crossover from Brand 'Z'

What are the criterias should i keep in mind before buying and assembling my 'custom Component' set.

>> Fs of the tweeter and Midbass
>> Crossover point the crossover (This should be less than or equal to my Fs of tweeter)
>> Power handling of the crossover.
>> Sensitivity of the Midbass and Tweeter

My Question is
Is the above setup possible?
If yes, are my considerations good enough / have i missed any other vital point
Can Bi-ampable crossover be used for such purposes

Thanks in advance

1) it is physically possible, but whether you should do it is very questionable. I wouldnt because of the different drivers and the specifications of them - when a manufacturer makes a component set he matches the characteristics of the drivers firstly and then designs a crossover that utilizes these characterisitics

2) crossover point is important but to be honest it should be at a minimum one octave more or less than the Fs of the drivers, never equal to the Fs because you get peaks and resonance that you cannot control at the Fs, so the further away from that your crossover point you go, the lower that these resonances will play ie they become inaudible compared to the frequencies that you hear louder

3) yes you have missed something ..... in terms of vital points when a manufacturer builds a crossover for a specific set he takes into account the phase relationship as well as the sensitivities of the drivers, in terms of output he would match the sensitivities or even allow differing levels of output using L-pads and in terms of phase he would take into account slopes etc for example Dynaudio use a low pass of 6db/octave slope for the mid and 12db/oct high pass for the tweeter - now changing either of those will change the phase of the frequencies at the crossover point resulting in uncontrollable cancellations

an important hint to remember here is that phase relates to the arrival time of the waveform and not "polarity" - speaker polarity can affect phase but it is not phase and phase differences can occur in different degrees according to how high or low you are in a crest or trough of a wave that arrives to your ear

bi-ampable crossovers are IMHO better than than the veriety that use a single amp but only for matched sets that can match the drivers with the crossover. I would go active instead of using a different manufacturers passive crossover and introducing the absolute values that th epassive would create. Most active crossovers allow you to change the slope and crossover point but if you do go to a higher slope then you have to try and keep the phase relationship as constant as you can as for example in the dynaudio set if you do change the slope of the mid to 12db/oct then you need to increase the slope of the tweeter high pass to possibly 18db/oct so the relationship needs to be kept constant so that you affect the phase as little as possible otherwise the effects of altering the phase could work like a psuedo time alignment in relationship to arrival times of a particular frequency to your ears

moral of the story ...... dont mix and match it isnt a good idea unless you get very lucky with matching the characteristics of everything or are knowledgeable enough to take two drivers and match their characteristics and then build your own passive crossover - even going active requires a small measure of knowing exactly what you are doing because you could make changes that adversely affect the matching of the speakers to each other

this is why cars are difficult to get right compared to home speakers where most of the drivers are all in one box and crossed over how they should be

[Mi10]

Thanks Invisible. That made sense (after reading thrice)

Regarding going Active;

Suppose i have high end HU like 9887 which allows me to go full 3-way active.
Can i give the connection of fronts (RCA) directly to a 2 Ch Amp. From the 2 Ch Amp i ll be giving the connection to a pair of Mid basses. Now Here is how the tunnning looks;
in HU, I HPF the Fronts above 50 Hz. In the Amp i LPF the Signal at 500 Hz.
(Ideally its nothing but BP between 50Hz-500 Hz)

Now is where the twist lies,
I take the rear preout to give it to another 2 Ch Amp. From this 2 Ch Amp, i use a Passive X over to divide the Mid range and Tweeter. So the tunning here would be;
HU - I HPF at 500 Hz. The Amp is set to Full range. The Passive Xover takes care of spiltting the mid range and the tweeter

The Sub preout will be given to a mono and then to sub for playing frequencies less than 50 Hz (LPF)

Now my query is Is this connection possible. Will it work this way (assume that all equipments are of same brand)

Sorry for confusing you,

[nitrous]

Why would you High pass mids at 50Hz, Praveen?

Originally Posted by Mi10

Thanks Invisible. That made sense (after reading thrice)

Regarding going Active;

Suppose i have high end HU like 9887 which allows me to go full 3-way active.
Can i give the connection of fronts (RCA) directly to a 2 Ch Amp. From the 2 Ch Amp i ll be giving the connection to a pair of Mid basses. Now Here is how the tunnning looks;
in HU, I HPF the Fronts above 50 Hz. In the Amp i LPF the Signal at 500 Hz.
(Ideally its nothing but BP between 50Hz-500 Hz)

Now is where the twist lies,
I take the rear preout to give it to another 2 Ch Amp. From this 2 Ch Amp, i use a Passive X over to divide the Mid range and Tweeter. So the tunning here would be;
HU - I HPF at 500 Hz. The Amp is set to Full range. The Passive Xover takes care of spiltting the mid range and the tweeter

The Sub preout will be given to a mono and then to sub for playing frequencies less than 50 Hz (LPF)

Now my query is Is this connection possible. Will it work this way (assume that all equipments are of same brand)

Sorry for confusing you,

if you have a 9887 then no problems the sub will work as you say

(the rest here purely assumes that you indeed have a 9887 and not some other adjustable headunit)

then lets move on to the midbasses - now im assuming you adjust the switch on the chassis of the headunit to work in 2.2ch mode so it works as a three way setup - and is no longer front/rear/sub but is instead high/mid/sub. Now you can set to highpass at 50hz and also set to lowpass at 500hz in other words you can bandpass on the headunit and you can just run the amplifier fullrange. The crossover points are set by the headunit. the mid crossover will work for the midbass. and these will work from the RCA outputs for front

then from the RCA outputs for rear you normally set the tweeter and this normally works from 1000hz up when set in auto mode but if you set it to user mode it works the full frequency range and you can then high pass at 500hz. The thing is if you have a passive crossover that is the same brand as the midrange driver and the tweeter and it is not a three way passive because that would leave one set of outputs empty (then this could cause a condition in the passive whereby it sees that empty channel as a dead short so never ever leave a passive crossover channel empty because you will damage the crossover and eventually the speakers) - so if its a proper two way crossover and you link up a set of midranges and a set of tweeters into it then it will work fine

only thing then is that the time alignment will work as follows :

1) the sub T/A will affect the sub
2) the mid T/A will affect the midbass
3) the tweeter T/A will affect the midrange and tweeter equally since both those drivers are hokked up to the tweeter channel hence you need to have the midrange driver and tweeter as close as possible to each other and as close to almost no pathlength difference between each side ..... this means that if the right hand side mid is 1 metre away from you then the right hand side tweeter also needs to be 1 metre away from you and if the left hand side mid is 1.5 metres away from you then the left hand side tweeter also needs to be 1.5 metres away from you. Left and right can differ between the two channels but mid and tweeter on each side needs to be the same distance away as each other per side and if they are not then you cannot use the T/A functionality at all because you will mess up more than you will solve with T/A because you affect the phase and timing in unpredictable ways so you will not have a good soundstage nor will you get good tonality because of unpredictable cancellations

nothing that i have said above changes - all those conditions still apply ie in terms of brands you cannot have/rather should not have mix and match component speakers hooked up to a different brand crossover because even if the drivers overlap sufficently in terms of their parameters you could create complications by using a non-matching crossover where the slope or the crossover point or even the phase of the drivers is not suitable

it will work perfectly if you have a three way set and find a way to get everything to work active and leave out the passive crossover altogether. As an example i have a 9887 - and then i have a dynaudio system 242 set which is a 6,5" midbass (MW162) and a tweeter (MD102) and a passive crossover ..... but im in the process of getting rid of the passive crossover and getting a set of 3" midrange drivers (MW142) so that i can run the front stage three way as well. So i will be using the configuration as i have described above ie sub lowpassed at 50hz from the sub channel then the midbass bandpassed from 75hz to around 700hz then the tweeter channels will be hooked up to a 4 channel amplifier and i will use the active crossover in that amplifier to lowpass the mids at 4000hz and highpass the tweeters at 4000hz

in my case im lucky whereby the 4 channel amplifier has a 2ch/4ch switch so i can connect it in a way that i only need one set of rca's to be connected to the amplifier ie the remaining channels receive exact replicas of the signal that the headunit sends to the first two channels and the built in crossover will control the speakers - if you dont have that and need all 4 RCA channels to be connected - its simple, you can use a pair of RCA splitters to make your one pair of RCA's finish of as 4 channels

absolutely no passives anywhere in my signal path so it simplify's the whole situation. The T/A issue is sorted out because the tweeters are being mounted directly above the midrange drivers and both will be in the A-Pillars so its got a minimal pathlength difference ie it wont be more than around 5cm's of pathlength difference between those two drivers, so if i do use T/A the midrange and tweeter will work as if its just one big tweeter working from a very low frequency point (at least thats how it should seem on paper anyway)

my biggest problem will be crossover slopes but thats a subject for another discussion and i have that covered anyway because the 9887 allows for a choice of slopes between 6/12/18/24db/oct so i can adjust the slopes for the sub and the midbass according to the fixed slope of the crossover on the amps which will drive the midrange and tweeter

if you have a different headunit then the above details could change minimally in terms of settings and adjustments but the principles of the install will remain the same - as it stands i do really have the 9887 and that dynaudio set and im working on my (new) install as we speak so i basically have all those figures falling out of my ears and i hope i have not confused you too much ..... if there's anything you dont understand in what ive said above just shout and i will try and explain it better

[Mi10]

Originally Posted by nitrous

Why would you High pass mids at 50Hz, Praveen?

Nitrous, i am trying to BP between 50 Hz and 500 HZ

Originally Posted by naughty001

if you have a 9887 then no problems the sub will work as you say

(the rest here purely assumes that you indeed have a 9887 and not some other adjustable headunit)

then lets move on to the midbasses - now im assuming you adjust the switch on the chassis of the headunit to work in 2.2ch mode so it works as a three way setup - and is no longer front/rear/sub but is instead high/mid/sub. Now you can set to highpass at 50hz and also set to lowpass at 500hz in other words you can bandpass on the headunit and you can just run the amplifier fullrange. The crossover points are set by the headunit. the mid crossover will work for the midbass. and these will work from the RCA outputs for front

then from the RCA outputs for rear you normally set the tweeter and this normally works from 1000hz up when set in auto mode but if you set it to user mode it works the full frequency range and you can then high pass at 500hz. The thing is if you have a passive crossover that is the same brand as the midrange driver and the tweeter and it is not a three way passive because that would leave one set of outputs empty (then this could cause a condition in the passive whereby it sees that empty channel as a dead short so never ever leave a passive crossover channel empty because you will damage the crossover and eventually the speakers) - so if its a proper two way crossover and you link up a set of midranges and a set of tweeters into it then it will work fine

only thing then is that the time alignment will work as follows :

1) the sub T/A will affect the sub
2) the mid T/A will affect the midbass
3) the tweeter T/A will affect the midrange and tweeter equally since both those drivers are hokked up to the tweeter channel hence you need to have the midrange driver and tweeter as close as possible to each other and as close to almost no pathlength difference between each side ..... this means that if the right hand side mid is 1 metre away from you then the right hand side tweeter also needs to be 1 metre away from you and if the left hand side mid is 1.5 metres away from you then the left hand side tweeter also needs to be 1.5 metres away from you. Left and right can differ between the two channels but mid and tweeter on each side needs to be the same distance away as each other per side and if they are not then you cannot use the T/A functionality at all because you will mess up more than you will solve with T/A because you affect the phase and timing in unpredictable ways so you will not have a good soundstage nor will you get good tonality because of unpredictable cancellations

nothing that i have said above changes - all those conditions still apply ie in terms of brands you cannot have/rather should not have mix and match component speakers hooked up to a different brand crossover because even if the drivers overlap sufficently in terms of their parameters you could create complications by using a non-matching crossover where the slope or the crossover point or even the phase of the drivers is not suitable

it will work perfectly if you have a three way set and find a way to get everything to work active and leave out the passive crossover altogether. As an example i have a 9887 - and then i have a dynaudio system 242 set which is a 6,5" midbass (MW162) and a tweeter (MD102) and a passive crossover ..... but im in the process of getting rid of the passive crossover and getting a set of 3" midrange drivers (MW142) so that i can run the front stage three way as well. So i will be using the configuration as i have described above ie sub lowpassed at 50hz from the sub channel then the midbass bandpassed from 75hz to around 700hz then the tweeter channels will be hooked up to a 4 channel amplifier and i will use the active crossover in that amplifier to lowpass the mids at 4000hz and highpass the tweeters at 4000hz

in my case im lucky whereby the 4 channel amplifier has a 2ch/4ch switch so i can connect it in a way that i only need one set of rca's to be connected to the amplifier ie the remaining channels receive exact replicas of the signal that the headunit sends to the first two channels and the built in crossover will control the speakers - if you dont have that and need all 4 RCA channels to be connected - its simple, you can use a pair of RCA splitters to make your one pair of RCA's finish of as 4 channels

absolutely no passives anywhere in my signal path so it simplify's the whole situation. The T/A issue is sorted out because the tweeters are being mounted directly above the midrange drivers and both will be in the A-Pillars so its got a minimal pathlength difference ie it wont be more than around 5cm's of pathlength difference between those two drivers, so if i do use T/A the midrange and tweeter will work as if its just one big tweeter working from a very low frequency point (at least thats how it should seem on paper anyway)

my biggest problem will be crossover slopes but thats a subject for another discussion and i have that covered anyway because the 9887 allows for a choice of slopes between 6/12/18/24db/oct so i can adjust the slopes for the sub and the midbass according to the fixed slope of the crossover on the amps which will drive the midrange and tweeter

if you have a different headunit then the above details could change minimally in terms of settings and adjustments but the principles of the install will remain the same - as it stands i do really have the 9887 and that dynaudio set and im working on my (new) install as we speak so i basically have all those figures falling out of my ears and i hope i have not confused you too much ..... if there's anything you dont understand in what ive said above just shout and i will try and explain it better

I am not sure if the Kenwood X693 supports 3 way active set up. In that case can a 3 way Active crossover help my cause ? (Something like Audio control X6S or IDX)

My first Goal is to eliminate the use of passive crossover.

I am kinda clear about the midbass stuff. I can directly give it from my amp with BP between 50 Hz - 500 Hz

My issue is with the tweeter. Most tweeters' Fs is above 1K. However my Amp is supporting only 50-500 HZ X over frequencies. So i would need a passive Tweeter crossover or atleast a Capacitor Circuit to protect my tweeters. Suppose i build a Capacitor circuit that can LPF from 1200 Hz, still i would be missing the Frequencies from 500 Hz - 1200 Hz. How do you expect me to solve this problem ?

The Following options come to my Mind

1. Get a Midrange that can take in from 500 Hz to 3000 Hz, however this would incur additional cost. also i would need another 2 Ch Amp which again is an additional cost.
2. Bandpass the Mid basses from 50 Hz - 2000 Hz (From HU) and then bring in the tweeter protection circuit so that tweets go LPF after 2Khz
3. Scrap all this, relax and get 3 Sony Xplod Coaxials