Team-BHP > In-Car Entertainment


Reply
  Search this Thread
9,110 views
Old 17th December 2009, 11:44   #1
BHPian
 
pranava999's Avatar
 
Join Date: Aug 2006
Location: BANGALORE
Posts: 845
Thanked: 23 Times
How important are crossovers and how would one identify a good XO design/build?

Hi gurus, how important is the role of crossovers in the case of
  • HUs,
  • component speakers
  • and amplifiers
How would one identify good build and quality in terms of component speaker XOs? Is it possible?

From whatever little I have gathered from the forum and reading elsewhere I have arrived at a conclusion that crossovers, have a very important role to play in terms of the overall listening experience as well as SQ performance of speakers.

In theory, a 2way /3 way component set-up should enhance one's listening pleasure through better staging as well as clarity and warmth as a result of the crossovers splitting the audio signal frequencies and feeding them to their respective drivers (highs to tweeters, mids to mid bass)

Can a poor quality XO result in the highs and lows sounding distinct and separate and as result sounding less cohesive while listening to music?

Thanks a lot and sincerely appreciate it.

Last edited by pranava999 : 17th December 2009 at 11:52.
pranava999 is offline  
Old 17th December 2009, 12:22   #2
Team-BHP Support
 
navin's Avatar
 
Join Date: Mar 2004
Location: Mumbai
Posts: 25,156
Thanked: 9,238 Times

Crossovers are significnat just like every link in a chain.

For HU's and amplifiers where the signal level is low (of the order of 1V) making a decent crossover is a lot easier (using a good quality audio opamp like the 5532/OP275/etc..) that say for loudpsekaers where polyester/teflon capacitors or air core inductors can make things expensive.

Crossovers in loudspeakers do more than just sperate the audio spectrum into 2 or ore bands. THey compensate for spekaer anamolies such as frequency response aberations, they compensate for the box design (read as baffle step compensation, baffle edge difraction, etc..).
navin is offline  
Old 17th December 2009, 22:59   #3
BHPian
 
pranava999's Avatar
 
Join Date: Aug 2006
Location: BANGALORE
Posts: 845
Thanked: 23 Times

Thanks Navin. It definitely is as important as every other link in the chain.Good to know that they have a larger role when it comes to speakers. Now is there any way for laymen like the rest of us to visually identify the quality of XO construction, components or design and layout?
pranava999 is offline  
Old 17th December 2009, 23:26   #4
Senior - BHPian
 
nitrous's Avatar
 
Join Date: May 2004
Location: UAE/Lon/Madras
Posts: 6,965
Thanked: 322 Times

Quote:
Originally Posted by Navin
THey compensate for spekaer anamolies such as frequency response aberations, they compensate for the box design (read as baffle step compensation, baffle edge difraction, etc..).
Navin, I believe with an active head and tuning knowledge, we can eliminate the need for a passive crossover. Of course, not to forget this does bring into the scene the need for more amplifier channels.

Am I wrong?
nitrous is offline  
Old 17th December 2009, 23:56   #5
Senior - BHPian
 
greenhorn's Avatar
 
Join Date: Apr 2007
Location: KL-01
Posts: 7,737
Thanked: 4,371 Times

You'll still need a crossover, or at least an equivalent circuit to protect the tweeters. they're very sensitive, and are best wired with at least a cap in series to protect them from very LF content ( kinda like a fuse)
greenhorn is offline  
Old 18th December 2009, 00:13   #6
Senior - BHPian
 
nitrous's Avatar
 
Join Date: May 2004
Location: UAE/Lon/Madras
Posts: 6,965
Thanked: 322 Times

Quote:
Originally Posted by greenhorn View Post
You'll still need a crossover, or at least an equivalent circuit to protect the tweeters. they're very sensitive, and are best wired with at least a cap in series to protect them from very LF content ( kinda like a fuse)
Of course. We cant run 100 RMS to a tweeter and expect it to live.
What "equivalent circuit" do we use in place of the XO?
nitrous is offline  
Old 18th December 2009, 07:35   #7
Senior - BHPian
 
greenhorn's Avatar
 
Join Date: Apr 2007
Location: KL-01
Posts: 7,737
Thanked: 4,371 Times

Its not just the sheer power (well, that's there too. the tweeter in a 100W RMS speaker can handle only a fraction of that amount) . I dont think tweeters can handle large amounts of LF information ( of ex, a loud click or a thud from turning on/off the amp, or touching the inputs etc) .

equivalent circuit = a High pass filter, but with the lowest possible frequency the speaker can handle. You can set the actual cutoff frequency in your electronic crossover
greenhorn is offline  
Old 18th December 2009, 09:40   #8
Senior - BHPian
 
frankmehta's Avatar
 
Join Date: Mar 2006
Location: Mumbai
Posts: 3,999
Thanked: 3,575 Times

Quote:
Originally Posted by greenhorn View Post
You'll still need a crossover, or at least an equivalent circuit to protect the tweeters. they're very sensitive, and are best wired with at least a cap in series to protect them from very LF content ( kinda like a fuse)

WHEN YOU talk of an active crossover from the head unit, you automatically ARE putting a circuit in between to protect the tweeters. Best part about it is that you can even keep the level of the tweeter ZERO when starting up the head unit, so you are even more safe than an analog crossover in between.
frankmehta is offline  
Old 18th December 2009, 10:22   #9
Team-BHP Support
 
navin's Avatar
 
Join Date: Mar 2004
Location: Mumbai
Posts: 25,156
Thanked: 9,238 Times

Quote:
Originally Posted by nitrous View Post
Navin, I believe with an active head and tuning knowledge, we can eliminate the need for a passive crossover.
Quote:
Originally Posted by greenhorn View Post
You'll still need a crossover, or at least an equivalent circuit to protect the tweeters. ...protect them from very LF content ( kinda like a fuse)
I have run tweeters (SS 9900, SEAS Cresendo, etc..) active without a passive crossover but the tweeter amp (in this case a triode tube amp) was crossed over at about 2khz/12db. A good amp will not have a turn on thump, but yes a turn on thump will kill many a tweeter (I lost a pair of Morel MDT33 - capable of handling quite a bit of power - this way).

If the amp that is driving the tweeter is crossed over too low then some tweeter protection should be in place. tweeters have very little excursion (in the order of 0.2-0.5mm) even information in the lower midrange can drive them past this excusrion then the tweeter lead will disconnect from the voice coil and the tweeter will go silent without any warning (woofer and midrnges for example distort and give you some warning).
navin is offline  
Old 18th December 2009, 10:45   #10
naughty001
 
Posts: n/a

Quote:
Originally Posted by nitrous View Post
Of course. We cant run 100 RMS to a tweeter and expect it to live.
What "equivalent circuit" do we use in place of the XO?
i was using an amp capable of 200 watts RMS per channel on my tweeters in my previous component set and nothing at all untoward happened to them and they sounded fantastic. In my current setup im also accounting to use 200 watts for my current set of tweeters but those have an impedance of 8 ohm so the theoretical power they recieve would be closer to 150 watts RMS (note though when playing music the average amount of power recieved by a tweeter would be closer only to around 10 to 15 watts RMS no matter what the nominal rating of the amplifier is)

as pointed up above power is not the only factor

Quote:
If the amp that is driving the tweeter is crossed over too low then some tweeter protection should be in place. tweeters have very little excursion (in the order of 0.2-0.5mm) even information in the lower midrange can drive them past this excusrion then the tweeter lead will disconnect from the voice coil and the tweeter will go silent without any warning (woofer and midrnges for example distort and give you some warning).
agreed but there are some large format tweeters which can happily play a bit of upper midrange as well and which play ridiculously low as far as tweeters go, and im not referring to exotic solutions like horn loaded compression drivers some of which actually have a response down to around 500hz and which safely can be high passed at around 800hz or so, BUT there are some conventional dome designs, admittedly high end and expensive which are larger format that actually play lower than the average tweeter
 
Old 18th December 2009, 10:50   #11
Team-BHP Support
 
navin's Avatar
 
Join Date: Mar 2004
Location: Mumbai
Posts: 25,156
Thanked: 9,238 Times

Quote:
Originally Posted by naughty001 View Post
i was using an amp capable of 200 watts RMS per channel on my tweeters

agreed but there are some large format tweeters which can happily play a bit of upper midrange..BUT there are some conventional dome designs, admittedly high end and expensive which are larger format that actually play lower than the average tweeter
Sure naughty, but we were just talking generally. Today I see many tweeters with Fs of 500hz that can be safely XOed at 1k/24db (both the 9900 and Cresendo/Magnum are such designs but these are $200+ each!).
navin is offline  
Old 18th December 2009, 11:15   #12
Senior - BHPian
 
greenhorn's Avatar
 
Join Date: Apr 2007
Location: KL-01
Posts: 7,737
Thanked: 4,371 Times

I would never wire a tweeter straight to an amp.I'd better be safe than sorry.

Unless....
They are piezo. But piezo tweeters in hi fi applications are quite rare... (but seen them in a lot of cheap '5 way' 6x9's)

Last edited by greenhorn : 18th December 2009 at 11:17.
greenhorn is offline  
Old 18th December 2009, 15:47   #13
naughty001
 
Posts: n/a

Quote:
I would never wire a tweeter straight to an amp.I'd better be safe than sorry.
well you affect many other things when you do what you propose. For example if you use the amplifiers onboard crossover to highpass at 200hz (as a random figure) at a slope of 12db/octave, then you add another capacitor to further filter at 200hz the slope then becomes 18db/octave ie it will have the effect of cascading the slope as well as the crossover point

if the crossover point is at a different value ie at 500hz you will have a drop at 500hz to 200hz then a steeper drop at 200hz and hence you are stepping the crossover point which again is counterproductive to the idea of a linear frequency response - so the idea of duplicating the functions of a crossover by using a passive filter on top of an active one makes absolutely no sense at all. NOTE im not advocating that you stop this practice if you do engage in it because you need to work in a way that keeps you personally confident of how your system performs but im merely stating why i would not use this method...... ever

you have to consider that when you affect the slopes you are playing with the absolute phase of the speaker caused by the crossover. To be honest some of the best sounding systems ive ever heard have had the tweeter wired up directly to the amp and used only the protection afforded by the values that the active crossover used whether that crossover was the amps onboard or the headunits onboard or even a dedicated processor somewhere in the loop before the amps
 
Old 18th December 2009, 16:44   #14
Senior - BHPian
 
DerAlte's Avatar
 
Join Date: Mar 2007
Location: Bangalore
Posts: 8,069
Thanked: 2,919 Times

Quote:
Originally Posted by naughty001 View Post
... if you use the amplifiers onboard crossover to highpass at 200hz (as a random figure) at a slope of 12db/octave, then you add another capacitor to further filter at 200hz the slope then becomes 18db/octave ie it will have the effect of cascading the slope as well as the crossover point ...
Not quite! If you draw a response graph, it will be easier to understand. In the scenario you have highlighted, the max. of the 2 slopes - 12dB - will be the applicable slope. No way can 12dB and 6dB be added together - the maths doesn't work that way.

In your other scenario, i.e. one at 500Hz, it is quite likely that the second filter will not get signal at a significant level, in the range that was filtered out by the previous stage, for it do to anything!
DerAlte is offline  
Old 19th December 2009, 22:53   #15
naughty001
 
Posts: n/a

Quote:
Originally Posted by DerAlte View Post
Not quite! If you draw a response graph, it will be easier to understand. In the scenario you have highlighted, the max. of the 2 slopes - 12dB - will be the applicable slope. No way can 12dB and 6dB be added together - the maths doesn't work that way.

In your other scenario, i.e. one at 500Hz, it is quite likely that the second filter will not get signal at a significant level, in the range that was filtered out by the previous stage, for it do to anything!
no sir its nothing to do with maths, if you look at crossover design to go up higher in slope you either add one or two more components to the design.

it is common knowledge that cascading crossovers however you do it will affect the slope, however you combine them it will basically change the slope and the final slope that the speaker "see's" will be the combined total of all the crossovers you use

just to prove it to you i have asked the question in the "dumb question" section at DIYMA which is reserved for stuff that is either been discussed a lot or even a very basic question and you will see that there are not many responses because im sure most responses would have to do a search for a frequently discussed topic - nevertheless the one reply that is there agrees with me

cascading crossovers - DIY Mobile Audio

this is not something im guessing about, it is something ive been aware of ever since i researched crossovers many years ago and ive considered it a basic premise for a very long time

another thing is if you use non concurrent crossover points it will step the response ie the slope will be equivalent to the single crossover until it increases at the second point caused by the effect of increasing the slope of the combined crossovers - if the points are arranged in a manner where the combined crossovers are overlapping first you have a steeper slope which remains steep since the crossovers are still combining

how it really works is that if you are using an active crossover the signal is already affected by the time it reaches the amp and the amp will see the altered signal and amplify that, then you add the cap in series to the speaker cable but the signal doesnt know its already been altered, so it is re-altered and the speaker sees a combined altering equivalent to all of those changes - and if you know your crossover design ie how adding caps and coils will change the slope the end result here is the same as having it all in one unit

Last edited by naughty001 : 19th December 2009 at 22:59.
 
Reply

Most Viewed
Thread Tools Search this Thread
Search this Thread:

Advanced Search

Copyright ©2000 - 2024, Team-BHP.com
Proudly powered by E2E Networks